RE: An Ada IDE and discussions

RE: An Ada IDE and discussions

[Beard, Frank]

I was browsing around the Aonix Web site
(http://www.aonix.com/content/index.html)
and clicked on the Product button at the top of the screen.
I just happened to look at TeleUSE.  It is supposed to be a
high powered GUI/Development tool for Unix.  I noticed a new
product TeleUSE/Windows.  The product now claims you can do
development on either platform and run on the other. I quote

"You need the flexibility to build your applications from a
single source - and give them the look and feel of the platform
on which they are deployed, whether it be UNIX or Windows. "

It also says;

"The "Wintif" library - is capable of rendering GUI elements
that look and feel like Motif, Windows NT, or Windows 95."

and

"After creating an executable for Solaris, the developers
transfer all their TeleUSE files to TeleUSE/Win on Windows.
TeleUSE/Win, which includes precisely the same set of tools
that are provided by TeleUSE on Solaris, accepts all the
files. When the Application Builder is invoked, TeleUSE/Win
automatically invokes Visual C++ and links the application
with libraries that provide a native Windows look and feel." 

I know you can't really go by what a vendor says, but if it's
half as good as they claim, why don't they advertise it more,
or at least speak up on this list.

Has anyone used this product?
Does it run on Linux?

Frank

-----Original Message-----
From: nicolas [mailto:n.brunot@cadwin.com]
Sent: Wednesday, July 18, 2001 4:14 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: An Ada IDE and discussions


"Stefan Skoglund" <stetson@ebox.tninet.se> a écrit dans le message news:
3B54BF44.646290DC@ebox.tninet.se...
> Using the windows look&feel in the GIMP wouldn't work !!

Is there a technical reason for that ?

We have a proprietary GUI originally developped for DOS
It was then ported to Unix
And finally to Windows when it came to evidence that in our market every DOS
user and 99% of unix users switched to Windows.

Originally it was a Motif look GUI.
A few years ago we changed the look to a Windows look GUI

There was some strong debate about that, but a few weeks after the change,
nobody (us internally as well as customers) would even think in coming back
to Motif

I have a RH 7.0 where a lot of things have a Windows look.
When you use swing, you have a choice, but you have a Windows look

By the way, are you sure it is very useful to re-invent swing again ?



_______________________________________________
comp.lang.ada mailing list
comp.lang.ada@ada.eu.org
http://ada.eu.org/mailman/listinfo/comp.lang.ada

Re: RE: An Ada IDE and discussions

[Ted Dennison]

In article <mailman.995648003.8404.comp.lang.ada@ada.eu.org>, Beard, Frank
says...
>I was browsing around the Aonix Web site
>(http://www.aonix.com/content/index.html)
>and clicked on the Product button at the top of the screen.
>I just happened to look at TeleUSE.  It is supposed to be a
>high powered GUI/Development tool for Unix.  I noticed a new
>product TeleUSE/Windows.  The product now claims you can do

Its not all that new. They were touting it around the time my six-year-old was
born.

>Has anyone used this product?
I've used it on Unix, but I never tried out a "portable" GUI with it.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html
          home email - mailto:dennison@telepath.com

Re: An Ada IDE and discussions

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 873 bytes --]

We never tried it, since we have our own product evolving for our needs.

I think that to be successfull, a Ada GUI product should

1/
be a standard product packaged with main Ada compilers
swing for Java is a good example

2/
be absolutely compatible and familiar to developpers with widely used
products for popular languages.
Especially for Windows and Linux developpers.

We have Java applications, for which the portability is even better than Ada
applications.
We don't even have to recompile for different platforms.

"Beard, Frank" <beardf@spawar.navy.mil> a �crit dans le message news:
mailman.995648003.8404.comp.lang.ada@ada.eu.org...
I know you can't really go by what a vendor says, but if it's
half as good as they claim, why don't they advertise it more,
or at least speak up on this list.

Has anyone used this product?
Does it run on Linux?

Frank

Java portability (was: An Ada IDE and discussions)

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 761 bytes --]


"nicolas" <n.brunot@cadwin.com> a �crit dans le message news: %cR67.1385$mz6.4505748@nnrp3.proxad.net...
> We have Java applications, for which the portability is even better than Ada
> applications.
> We don't even have to recompile for different platforms.
>
This has nothing to do with Java, and all to do with the JVM. Any language compiled to the JVM (including and not limited to Ada)
will do the same.

Remember that Java is the less portable of all programming languages, since it runs on only one machine. It is so unportable that to
run it, you must emulate the only machine on which it runs !

--
---------------------------------------------------------
           J-P. Rosen (rosen@adalog.fr)
Visit Adalog's web site at http://www.adalog.fr

Re: Java portability (was: An Ada IDE and discussions)

[Gerhard Häring]

On Mon, 23 Jul 2001 10:53:36 +0200, Jean-Pierre Rosen <rosen@adalog.fr> wrote:
>This has nothing to do with Java, and all to do with the JVM. Any language
>compiled to the JVM (including and not limited to Ada) will do the same.
>
>Remember that Java is the less portable of all programming languages, since it
>runs on only one machine. It is so unportable that to run it, you must emulate
>the only machine on which it runs !

Native compilers for Java do exist as well.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://highqualdev.com              public key at homepage
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y: x+y, [chr(ord(x)^42) for x in list('zS^BED\nX_FOY\x0b')])

Re: Java portability (was: An Ada IDE and discussions)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1445 bytes --]

What I meant is that we write code, we have a working application, and we
can provide it as it is for different platforms.
If the JVM is the only thing that must be ported, at least it's done one for
all for a given platform.

Ok, you can find different versions of the JVM, compatibility between
Microsoft and Sun can be a problem.
But globally , the concept is excellent, in practice, it works not so bad,
and nobody can deny it is a success.

After all, one could say you have to port the Ada compiler for each platform
you want to use.

"Jean-Pierre Rosen" <rosen@adalog.fr> a �crit dans le message news:
9jgopl$5v6$1@s1.read.news.oleane.net...
>
> "nicolas" <n.brunot@cadwin.com> a �crit dans le message news:
%cR67.1385$mz6.4505748@nnrp3.proxad.net...
> > We have Java applications, for which the portability is even better than
Ada
> > applications.
> > We don't even have to recompile for different platforms.
> >
> This has nothing to do with Java, and all to do with the JVM. Any language
compiled to the JVM (including and not limited to Ada)
> will do the same.
>
> Remember that Java is the less portable of all programming languages,
since it runs on only one machine. It is so unportable that to
> run it, you must emulate the only machine on which it runs !
>
> --
> ---------------------------------------------------------
>            J-P. Rosen (rosen@adalog.fr)
> Visit Adalog's web site at http://www.adalog.fr
>
>

Re: Java portability (was: An Ada IDE and discussions)

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 912 bytes --]


"nicolas" <n.brunot@cadwin.com> a �crit dans le message news: lpS67.1427$OA6.4515044@nnrp3.proxad.net...
> What I meant is that we write code, we have a working application, and we
> can provide it as it is for different platforms.
> If the JVM is the only thing that must be ported, at least it's done one for
> all for a given platform.
>
> Ok, you can find different versions of the JVM, compatibility between
> Microsoft and Sun can be a problem.
> But globally , the concept is excellent, in practice, it works not so bad,
> and nobody can deny it is a success.
>
True, but what I meant is that it is a feature of the JVM, not of the Java language.
If you compile Ada for the JVM (Ada-Magic, JGNAT), you'll get exactly the same binary compatibility.

--
---------------------------------------------------------
           J-P. Rosen (rosen@adalog.fr)
Visit Adalog's web site at http://www.adalog.fr

Re: Java portability (was: An Ada IDE and discussions)

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 919 bytes --]


"Gerhard H�ring" <gerhard.nospam@bigfoot.de> a �crit dans le message news: kvqgj9.e9r.ln@gargamel.hqd-internal...
> On Mon, 23 Jul 2001 10:53:36 +0200, Jean-Pierre Rosen <rosen@adalog.fr> wrote:
> >This has nothing to do with Java, and all to do with the JVM. Any language
> >compiled to the JVM (including and not limited to Ada) will do the same.
> >
> >Remember that Java is the less portable of all programming languages, since it
> >runs on only one machine. It is so unportable that to run it, you must emulate
> >the only machine on which it runs !
>
> Native compilers for Java do exist as well.
>
And is the result really portable? Even for numerics on a non-IEEE machine ? Even for multi-threaded programs (that depend on the
scheduling algorithm) ?

--
---------------------------------------------------------
           J-P. Rosen (rosen@adalog.fr)
Visit Adalog's web site at http://www.adalog.fr

Re: Java portability (was: An Ada IDE and discussions)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 477 bytes --]

"Jean-Pierre Rosen" <rosen@adalog.fr> a �crit dans le message news:
9jh2cs$aon$1@s1.read.news.oleane.net...
> True, but what I meant is that it is a feature of the JVM, not of the Java
language.
> If you compile Ada for the JVM (Ada-Magic, JGNAT), you'll get exactly the
same binary compatibility.

Does it work for GUI components ?
If you develop an application with GtkAda for example, and use JGNAT,
do you get something like what you would get with Swing and Java ?

Re: Java portability (was: An Ada IDE and discussions)

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1068 bytes --]


"nicolas" <n.brunot@cadwin.com> a �crit dans le message news: 2sU67.1485$di7.4670499@nnrp3.proxad.net...
> "Jean-Pierre Rosen" <rosen@adalog.fr> a �crit dans le message news:
> 9jh2cs$aon$1@s1.read.news.oleane.net...
> > True, but what I meant is that it is a feature of the JVM, not of the Java
> language.
> > If you compile Ada for the JVM (Ada-Magic, JGNAT), you'll get exactly the
> same binary compatibility.
>
> Does it work for GUI components ?
> If you develop an application with GtkAda for example, and use JGNAT,
> do you get something like what you would get with Swing and Java ?
>
I don't think there is support for Gtk in Java either...
If you develop in Ada for the JVM, of course, you will use AWT/Swing/Whatever-is-the-newest-toolkit-today.

(and before you ask the question: of course, all Ada compilers on the JVM are 100% interfaced with existing classes, and can use all
Java classes).

--
---------------------------------------------------------
           J-P. Rosen (rosen@adalog.fr)
Visit Adalog's web site at http://www.adalog.fr

Re: Java portability (was: An Ada IDE and discussions)

[Marc A. Criley]

Jean-Pierre Rosen wrote:
> 
> I don't think there is support for Gtk in Java either...
> If you develop in Ada for the JVM, of course, you will use
> AWT/Swing/Whatever-is-the-newest-toolkit-today.
> 
> (and before you ask the question: of course, all Ada compilers on the JVM
> are 100% interfaced with existing classes, and can use all Java classes).

And I will vouch for this.  I've been working with JGNAT quite a bit
recently, using the Java class libraries, specifically Swing for the
GUI.  While I've certainly not used every single available class, every
one that I have used has interoperated just fine.

Marc A. Criley

Re: An Ada IDE and discussions

[Warren W. Gay VE3WWG]

"Beard, Frank" wrote:
> I was browsing around the Aonix Web site
> (http://www.aonix.com/content/index.html)
> and clicked on the Product button at the top of the screen.
> I just happened to look at TeleUSE.  It is supposed to be a
> high powered GUI/Development tool for Unix.  I noticed a new
> product TeleUSE/Windows.  The product now claims you can do
> development on either platform and run on the other. I quote
...snip...
> I know you can't really go by what a vendor says, but if it's
> half as good as they claim, why don't they advertise it more,
> or at least speak up on this list.
> 
> Has anyone used this product?
> Does it run on Linux?
> 
> Frank

Linux? Dunno..

We looked at it briefly. We were looking for a rapid X-Window
development package, and for this, it didn't seem to make the
grade for us. I am convinced that if you become well-versed in
their product, including their "D" language, it might work well.

However, if you have the situation where your developers are
not using the product daily/weekly, and/or they're not quick
on the technical uptake, you won't find it's use "rapid". It
does purport to support the "whole development life cycle", 
which may justify the "extra effort". For us, quick and _easy_
development was more important, since for maintenance, we rarely
make big changes to the screens after the fact. What I found with
my initial trials was that I could have coded the darn thing by
hand faster (at least for smaller screens).

What I also found was that when I got it to generate code, I often
ended up choosing names that conflicted with "something else",
and so it would seldom work for me the first time (there would
be conflicts in their "D" language with either other objects or
keywords). This would be OK, but there was never any errors or
warnings -- just the compiled code would not work.
After much investigation, you could eventually make 
it work -- but this defeated the "quick development time" that 
we were looking for.

We ended up going for X-Designer, which was much easier to
learn (you might be able to use it without documentation). It
did not require proprietary widget libraries either. THe downsides
are that it has fewer widgets to choose from (but you can add to
the selection), and it does not try to address the full "development
life cycle". 

X-Designer is available for Linux, and I think it
can support Ada if you use some downloadable perl script to do
some script-magic.

FWIW, Warren.

-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Java portability (was: An Ada IDE and discussions)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 694 bytes --]

We never tried Ada for the JVM
Is there a standard specification when you want to access Java libraries, or
is it specific for each vendor (Ada-Magic, JGNAT, others ?)
If we write Ada code for the JVM, will it be the same for all compiler
vendors which support it ?

"Marc A. Criley" <mcqada@earthlink.net> a �crit dans le message news:
3B5C4A92.647FC2EC@earthlink.net...
> Jean-Pierre Rosen wrote:
>
> And I will vouch for this.  I've been working with JGNAT quite a bit
> recently, using the Java class libraries, specifically Swing for the
> GUI.  While I've certainly not used every single available class, every
> one that I have used has interoperated just fine.
>
> Marc A. Criley

Re: Java portability (was: An Ada IDE and discussions)

[Marc A. Criley]

nicolas wrote:
> 
> We never tried Ada for the JVM
> Is there a standard specification when you want to access Java libraries, or
> is it specific for each vendor (Ada-Magic, JGNAT, others ?)
> If we write Ada code for the JVM, will it be the same for all compiler
> vendors which support it ?

Since there's no officially recognized Java standard yet, there's no Ada
standard specification for interfacing to it (as there is for C,
Fortran, and COBOL, and just recently for C++), so vendors are free to
forge their own specification.

I'm nearly certain that each vendor has their own means of interfacing
to the Java libraries.  I've not looked at the specifics of how AdaMagic
does it, but I know that JGNAT added Java-specific pragmas to support
interfacing.  In addition, the documentation describing how Java
Interfaces are implemented (as discriminants) suggests some "magic"
going on inside the compiler.

It's not the _Ada_ code for a JVM that will differ between vendors, it's
the interfacing to the class libraries.  I would expect vanilla Ada code
to be quite portable, but the interfaces to Java classes will resemble
one another, and differ in the details.

Marc A. Criley

Re: Java portability (was: An Ada IDE and discussions)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 924 bytes --]

"Marc A. Criley" <mcqada@earthlink.net> a �crit dans le message news:
3B5D5B79.F2DC527E@earthlink.net...

> I'm nearly certain that each vendor has their own means of interfacing
> to the Java libraries.  I've not looked at the specifics of how AdaMagic
> does it, but I know that JGNAT added Java-specific pragmas to support
> interfacing.

Unfortunetaly that's exactly what should be avoided.
We chose Ada more then 12 years ago for safety and portability concerns, and
check our code against different compilers.
Being tied to one compiler vendor is exactly what you are likely to dislike
when you choose Ada.

Ada world is small enough to make you feel extremely uncomfortable if you
depends on one single company.
It is everything but nice to depend on Microsoft, it's not a very good
solution to find yourself in the same case,
but with a company which could disappear or drop Ada language within a few
years.

Re: Java portability (was: An Ada IDE and discussions)

[Florian Weimer]

"Jean-Pierre Rosen" <rosen@adalog.fr> writes:

> And is the result really portable? Even for numerics on a non-IEEE
> machine ?

Yes, I think so. Some people claim that the JVM support for floating
point operations was designed so that it runs fast and correctly only
on Sun's hardware. ;-)

Re: Java portability (was: An Ada IDE and discussions)

[Marc A. Criley]

nicolas wrote:
> 
> Ada world is small enough to make you feel extremely uncomfortable if you
> depends on one single company.
> It is everything but nice to depend on Microsoft, it's not a very good
> solution to find yourself in the same case,
> but with a company which could disappear or drop Ada language within a few
> years.

Though I would doubt that _Ada_ Core Technologies would drop Ada any
time in the near future :-)

Marc

Re: Java portability (was: An Ada IDE and discussions)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 602 bytes --]

"Marc A. Criley" <mcqada@earthlink.net> a �crit dans le message news:
3B5DCE74.C12AA2D8@earthlink.net...
> nicolas wrote:
>
> Though I would doubt that _Ada_ Core Technologies would drop Ada any
> time in the near future :-)

I doubt it too ...
Unfortunately, at least in Windows world, this is the only one company for
which you can be quite certain about that.
Ada developments are ususally designed to last more than 'near future'
Still remains the problem that when there is only one company left, you'd
better be satisfied with its products,
and remember that if they close, 1 - 1 = 0 ...

Re: Java portability (was: An Ada IDE and discussions)

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 896 bytes --]


"Florian Weimer" <fw@deneb.enyo.de> a �crit dans le message news: 877kwycfw7.fsf@deneb.enyo.de...
> "Jean-Pierre Rosen" <rosen@adalog.fr> writes:
>
> > And is the result really portable? Even for numerics on a non-IEEE
> > machine ?
>
> Yes, I think so. Some people claim that the JVM support for floating
> point operations was designed so that it runs fast and correctly only
> on Sun's hardware. ;-)
What makes you think so ? The trouble is that Java *requires* IEEE arithmetic. On non IEEE hardware you have a choice of a) emulate
IEEE in software (correct and slow) or b) ignore the requirement (incorrect and fast). On a non-IEEE number-crunching machine (like
SGI), my bet is that the solution b) will be taken, hence my concern...

--
---------------------------------------------------------
           J-P. Rosen (rosen@adalog.fr)
Visit Adalog's web site at http://www.adalog.fr

Re: Java portability (was: An Ada IDE and discussions)

[David C. Hoos, Sr.]

----- Original Message -----
From: "Jean-Pierre Rosen" <rosen@adalog.fr>
Newsgroups: comp.lang.ada
To: <comp.lang.ada@ada.eu.org>
Sent: July 25, 2001 3:40 AM
Subject: Re: Java portability (was: An Ada IDE and discussions)


>
> "Florian Weimer" <fw@deneb.enyo.de> a �crit dans le message news:
877kwycfw7.fsf@deneb.enyo.de...
> > "Jean-Pierre Rosen" <rosen@adalog.fr> writes:
> >
> > > And is the result really portable? Even for numerics on a non-IEEE
> > > machine ?
> >
> > Yes, I think so. Some people claim that the JVM support for floating
> > point operations was designed so that it runs fast and correctly only
> > on Sun's hardware. ;-)
> What makes you think so ? The trouble is that Java *requires* IEEE
arithmetic. On non IEEE hardware you have a choice of a) emulate
> IEEE in software (correct and slow) or b) ignore the requirement (incorrect
and fast). On a non-IEEE number-crunching machine (like
> SGI), my bet is that the solution b) will be taken, hence my concern...
Which SGI machine doesn't do IEEE arithmetic? all the SGI machines
I ever used did.

To be sure, the concern, in general, is a valid one.

>
> --
> ---------------------------------------------------------
>            J-P. Rosen (rosen@adalog.fr)
> Visit Adalog's web site at http://www.adalog.fr
>
>
> _______________________________________________
> comp.lang.ada mailing list
> comp.lang.ada@ada.eu.org
> http://ada.eu.org/mailman/listinfo/comp.lang.ada
>

Re: Java portability (was: An Ada IDE and discussions)

[Florian Weimer]

"Jean-Pierre Rosen" <rosen@adalog.fr> writes:

>> Yes, I think so. Some people claim that the JVM support for floating
>> point operations was designed so that it runs fast and correctly only
>> on Sun's hardware. ;-)

> What makes you think so ? The trouble is that Java *requires* IEEE
> arithmetic.

A particular incarnation of it.  For example, you can't use x86 80 bit
extended precision floating point operations with the JVM.  OTOH,
there is some double-extended-exponent type, and I would be suprised
if no SPARC processor implemented it. ;-)

Re: Java portability (was: An Ada IDE and discussions)

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1176 bytes --]


"Florian Weimer" <fw@deneb.enyo.de> a �crit dans le message news: 874rs0pwbk.fsf@deneb.enyo.de...
> "Jean-Pierre Rosen" <rosen@adalog.fr> writes:
>
> >> Yes, I think so. Some people claim that the JVM support for floating
> >> point operations was designed so that it runs fast and correctly only
> >> on Sun's hardware. ;-)
>
> > What makes you think so ? The trouble is that Java *requires* IEEE
> > arithmetic.
>
> A particular incarnation of it.  For example, you can't use x86 80 bit
> extended precision floating point operations with the JVM.  OTOH,
> there is some double-extended-exponent type, and I would be suprised
> if no SPARC processor implemented it. ;-)

There is much more in IEEE arithmetic than the format of numbers. You have to support NaNs, infinities, and a lot of other stuff.
And of course, when I'm talking about adhering to a standard, I meant 100%. Being 95% conformant is sooooo much easier! Ask Ada
compiler vendors. (and the defunct Artek compiler, that never made it from 95% to 100%)

--
---------------------------------------------------------
           J-P. Rosen (rosen@adalog.fr)
Visit Adalog's web site at http://www.adalog.fr

Re: Java portability

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Marc A. Criley" <mcqada@earthlink.net> a ?crit dans le message news:
: 3B5D5B79.F2DC527E@earthlink.net...
: 
:> I'm nearly certain that each vendor has their own means of interfacing
:> to the Java libraries.  I've not looked at the specifics of how AdaMagic
:> does it, but I know that JGNAT added Java-specific pragmas to support
:> interfacing.
: 
: Unfortunetaly that's exactly what should be avoided.

Well, what can you do? Does your company do some lobbying
to get this standardized?
Besides, methinks we all should be using DWIM :-)

Re: Java portability

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Marc A. Criley" <mcqada@earthlink.net> a ?crit dans le message news:
: 3B5DCE74.C12AA2D8@earthlink.net...
: and remember that if they close, 1 - 1 = 0 ...

GNAT is Free Software...

Re: Java portability

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 598 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9jp5dc$e2b$1@a1-hrz.uni-duisburg.de...
> Well, what can you do? Does your company do some lobbying
> to get this standardized?
> Besides, methinks we all should be using DWIM :-)
>
What else can we do ? the answer is easy.
Compare available options and choose the best one
If there is a good way to access Java through Ada, which doesn't make you
dependant of just one vendor, why not ?
If it's easier to write in Java directly, why spending time in useless
lobbying and complicated or not portable interfaces ?

Re: Java portability

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 403 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9jp5eo$e2b$2@a1-hrz.uni-duisburg.de...
> GNAT is Free Software...
So what ?
If maintaining your own compiler, operating system or whatever else is a
risk high enough,
no company will go for it, it would be much more productive to change the
tool.
That's why it's so important to have standards and different vendors.

Re: Java portability

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
: news: 9jp5dc$e2b$1@a1-hrz.uni-duisburg.de...

: If there is a good way to access Java through Ada, which doesn't make you
: dependant of just one vendor, why not ?
: If it's easier to write in Java directly, why spending time in useless
: lobbying and complicated or not portable interfaces ?

May I suggest that you try to answer these questions?

Georg

Re: Java portability

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
: news: 9jp5eo$e2b$2@a1-hrz.uni-duisburg.de...
:> GNAT is Free Software...
: So what ?
: If maintaining your own compiler, operating system or whatever else is a
: risk high enough,
: no company will go for it, it would be much more productive to change the
: tool.
: That's why it's so important to have standards and different vendors.

Nicolas, there are many ifs and there is, if I may say so,
bold company prophecy of vague kind in what you write.

Re: Java portability

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 301 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9jrcmm$mc0$1@a1-hrz.uni-duisburg.de...
> May I suggest that you try to answer these questions?
What makes you think it's not already done ? Given current standards and
tools, there is no real question to answer ...

Re: Java portability

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1269 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9jrdl3$mh2$1@a1-hrz.uni-duisburg.de...
> Nicolas, there are many ifs and there is, if I may say so,
> bold company prophecy of vague kind in what you write.

I usually hear 2 kind of reaction about Ada from 99.9% of people involved in
software development.
1/ Ada ? what is it ? How, that's a language. Never heard about it ...
2/ Ada ? I heard about it a few years ago, this language looked really great
! Does it still exist ?

Current use of Ada language in software develoment is everything but 'ifs'
I don't know if the way Ada is promoted is a good one. But I know the
current situation.
What percentage of Free Software is develop in Ada ?

Very few people use Ada. They are usually convincing one another among
already convinced people.
The rest of the world is usually considered as stupid enough not to use Ada
I don't think it's a very efficient promotion.

I imagine a car vendor telling you :
Here is your car !
Of course there are no seats ... stupid you are ... There are plenty of seat
vendors. Find and chose it yourself.
Lights ? Are you sure you really need to use it at night ... ? That's not
safe ... Anyway I guess some people must provide lights somewhere ...

Re: Java portability

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1088 bytes --]

"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> a �crit dans le
message news: Yjoj5DGkwoqg@eisner.encompasserve.org...
> There was no international standard for Java at the time the Ada95
> standard was developed.  In fact, the only guidance on what constitutes
> "Java" still comes from just one vendor (Sun), doesn't it ?

Yes, but notice that "standard" can be a matter of fact and wide acceptance
which doesn't necessarily require a RM, at least in the beginning.

> Typically those who adopt Ada are interested in "more correct" and
> "easier to maintain" rather than "easier to write".

we adopted Ada before 1990. "correct" is not so easy to define.
actually if there is no standard of fact, and too few users
it can be "easier to maintain" as well as "easier to write" to write
directly in Java,
than to use an interface which could disappear or let you in a nightmare if
each provider chose different options.

We have a strong standard and a RM for Ada95, and you sometime still have to
care if you want your Ada sources to be the same for Gnat, Objectada and
Apex

Re: Java portability

[Larry Kilgallen]

In article <9jrcmm$mc0$1@a1-hrz.uni-duisburg.de>, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:
> nicolas <n.brunot@cadwin.com> wrote:
> : "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
> : news: 9jp5dc$e2b$1@a1-hrz.uni-duisburg.de...
> 
> : If there is a good way to access Java through Ada, which doesn't make you
> : dependant of just one vendor, why not ?

There was no international standard for Java at the time the Ada95
standard was developed.  In fact, the only guidance on what constitutes
"Java" still comes from just one vendor (Sun), doesn't it ?

Of course, that is not the only consideration that went into Ada95,
since there is no binding to Pascal, and Pascal has had an international
standard for many years.

> : If it's easier to write in Java directly, why spending time in useless
> : lobbying and complicated or not portable interfaces ?

Typically those who adopt Ada are interested in "more correct" and
"easier to maintain" rather than "easier to write".

> May I suggest that you try to answer these questions?

I was not the one who posted them, but since your message was posted
rather than sent in email, I presume the invitation to answer was
open to anyone.

Re: Java portability

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
: news: 9jrcmm$mc0$1@a1-hrz.uni-duisburg.de...
:> May I suggest that you try to answer these questions?
: What makes you think it's not already done ? Given current standards and
: tools, there is no real question to answer ...

Then I must have misunderstood, because I had thaught you asked...

Re: Java portability

[Marin David Condic]

A very nice analogy in some ways. Maybe more accurate to characterize it
along the lines of "luxury" items in a car since Ada provides all the
essentials that any programming language needs to provide. (No missing seats
or headlights.) What's missing are things like Air Conditioning, CD Player,
Tinted Windows, Power Door Locks, etc. The buyer takes a look at the car and
says "Nice car with lots of room and a powerful engine, but I really need an
air conditioner and tinted glass since I live where it is hot." The response
is "Well you can always take it to an aftermarket shop and have that stuff
installed for you at additional expense and hastle and, no, it won't look
like it was an exact match for the car, but it will get the job done..." The
buyer sees the Lexus dealer down the street who has the full package
available, nicely integrated to fit the design of the car, factory warranty,
sitting on the lot - maybe at a somewhat higher sticker price, but its all
right there and ready to go - just write a check. Which way is he likely to
go?

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"nicolas" <n.brunot@cadwin.com> wrote in message
news:%hb87.917$%w2.3730577@nnrp3.proxad.net...
>
> I imagine a car vendor telling you :
> Here is your car !
> Of course there are no seats ... stupid you are ... There are plenty of
seat
> vendors. Find and chose it yourself.
> Lights ? Are you sure you really need to use it at night ... ? That's not
> safe ... Anyway I guess some people must provide lights somewhere ...
>

Re: Java portability

[Warren W. Gay VE3WWG]

Marin David Condic wrote:
> A very nice analogy in some ways. Maybe more accurate to characterize it
> along the lines of "luxury" items in a car since Ada provides all the
> essentials that any programming language needs to provide. (No missing seats
> or headlights.) What's missing are things like Air Conditioning, CD Player,
> Tinted Windows, Power Door Locks, etc. The buyer takes a look at the car and

I'd like to offer a slightly different "analogy" :

_flamesuit on_

The shopper looks at the Ada car, and notices that it has seatbelts and
of course 4 doors on it. The programmer says, but I don't want to be 
restrained from accessing the outdoors -- I go through drive-throughs
a lot..

The shopper than goes down the street and settles on the C++ car
which does not have the restrictive seat belts, and of course,
_NO_ doors to get in the way ;-)  The Java car has seatbelts, but
no doors...

_flamesuit off_

-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Java portability

[Marin David Condic]

Oh, I think most of us would accept that Ada has superior safety features
and that this is important. I think Stanley Allen drew the analogy to a
Volvo in an article on AdaPower. He also observed that Volvo had to add some
"Sex Appeal" features because people weren't buying it in droves.

I'm not saying that lack of safety is a sex appeal feature. I'll bet that
Volvos may have gotten sexier but that underneath it all there is still a
lot of safety engineering.

I think I'd suggest that the fundamentals of Ada are very superior to the
fundamentals of most other general purpose languages. I'd even suggest that
it had a lot of features that other languages don't have. (Tasking &
generics come to mind - analogous to a Borg Warner T-56 six speed racing
transmission and an Edelbrock Nitrous Oxide System) Those are all wonderful
language attributes, but somewhere along the line the buyer seems to want
plush velour seats, a hood scoop, wheel flares and a spoiler and doesn't
want them to look like someone installed them with a pop-rivet gun and
sprayed Rustoleum on it. The parallel with Ada would be all of the
dingleberries hanging around other languages in the way of development
tools. They've got them and we don't - at least not in the spiffy package
that the buyer has come to expect.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Warren W. Gay VE3WWG" <ve3wwg@home.com> wrote in message
news:3B619A6D.5DD6E782@home.com...
> I'd like to offer a slightly different "analogy" :
>
> _flamesuit on_
>
> The shopper looks at the Ada car, and notices that it has seatbelts and
> of course 4 doors on it. The programmer says, but I don't want to be
> restrained from accessing the outdoors -- I go through drive-throughs
> a lot..
>
> The shopper than goes down the street and settles on the C++ car
> which does not have the restrictive seat belts, and of course,
> _NO_ doors to get in the way ;-)  The Java car has seatbelts, but
> no doors...
>
> _flamesuit off_

Re: Java portability

[Stefan Skoglund]

Marin David Condic wrote:
> along the lines of "luxury" items in a car since Ada provides all the
> essentials that any programming language needs to provide. (No missing seats

When Ada is the Rolls-royce company of old.
You could buy your rolls ready-made from the factory
or with no carriage/seats but technically complete
and then commission your local horse-wagoon manufacturer.

It exist a number of rolls here in Sweden with Stockholm-made
carriage.

Re: Java portability

[Stefan Skoglund]

Marin David Condic wrote:
> I'm not saying that lack of safety is a sex appeal feature. I'll bet that
> Volvos may have gotten sexier but that underneath it all there is still a
> lot of safety engineering.
Oh, the current V70 is sexy.
As a minimum it is pretty fast (the fastest one is nasty)
Very good balance. Somewhat sensitive brakes (they cuts real fast so
a sensitive feet is recommended)

Very nice power steering.
And everything with front-wheel steering.

But the latest word is that the S/V 60 is better.

Re: Straight Jackets

[Warren W. Gay VE3WWG]

Marin David Condic wrote:
> Oh, I think most of us would accept that Ada has superior safety features
> and that this is important. I think Stanley Allen drew the analogy to a
> Volvo in an article on AdaPower. He also observed that Volvo had to add some
> "Sex Appeal" features because people weren't buying it in droves.
> 
> I'm not saying that lack of safety is a sex appeal feature. I'll bet that
> Volvos may have gotten sexier but that underneath it all there is still a
> lot of safety engineering.

I think the real reason has to do with "design effort". If you are accustomed
to C/C++/Java/Assembly language, where you can express "what you like" (or
"almost" in some cases), then you will find Ada/Ada95 a restrictive
"straight jacket". To love Ada, you need to learn how to go about
things in a different (safer) way, and this requires _effort_, and only
then starts to feel less like a "straight jacket".

For an example, the C++ programmer can easily share file descriptors 
(or FILE pointers) in and out of a thread (task), or between threads,
if he thinks this is safe to do so. In Ada, the File_Type cannot be 
copied (like any limited type), and so you can only pass the
File_Type to the rendevous; but this use cannot be extended past the 
rendevous in the task (with good reason).

This makes perfect safety sense but programmers new to this straight
jacket, often refuse to like it.  Other examples involving access
type scope etc. can also be made -- in each case, the programmer 
new to Ada feels that this is cramping his style. So rather than
put the effort to go beyond learning the language, and learning to
apply it, they fall back to their old evil ways and tools.

Soooo,

The real difficulty, IMHO, is getting programmers to accept that these
programming "straight jackets" are a "good thing" (TM). ;-)

Warren.

> I think I'd suggest that the fundamentals of Ada are very superior to the
> fundamentals of most other general purpose languages. I'd even suggest that
> it had a lot of features that other languages don't have. (Tasking &
> generics come to mind - analogous to a Borg Warner T-56 six speed racing
> transmission and an Edelbrock Nitrous Oxide System) Those are all wonderful
> language attributes, but somewhere along the line the buyer seems to want
> plush velour seats, a hood scoop, wheel flares and a spoiler and doesn't
> want them to look like someone installed them with a pop-rivet gun and
> sprayed Rustoleum on it. The parallel with Ada would be all of the
> dingleberries hanging around other languages in the way of development
> tools. They've got them and we don't - at least not in the spiffy package
> that the buyer has come to expect.
> 
> MDC
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/
> 
> "Warren W. Gay VE3WWG" <ve3wwg@home.com> wrote in message
> news:3B619A6D.5DD6E782@home.com...
> > I'd like to offer a slightly different "analogy" :
> >
> > _flamesuit on_
> >
> > The shopper looks at the Ada car, and notices that it has seatbelts and
> > of course 4 doors on it. The programmer says, but I don't want to be
> > restrained from accessing the outdoors -- I go through drive-throughs
> > a lot..
> >
> > The shopper than goes down the street and settles on the C++ car
> > which does not have the restrictive seat belts, and of course,
> > _NO_ doors to get in the way ;-)  The Java car has seatbelts, but
> > no doors...
> >
> > _flamesuit off_

-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Java portability

[Marin David Condic]

Well, for that matter, when you go buy a Freightliner Tractor, you can buy
it without an engine & get one of your own choosing put in there. Or you
could buy an F-15 Eagle and choose to put a GE engine (The Bad Guys) or a
Pratt & Whitney engine (The Good Guys) in it - but they never came "factory
equipped" with one or the other as the default. This is the point where the
analogy breaks down ;-)

The point being that in the days when language vendors are supplying big
class libraries, GUI builders, debuggers, IDEs that do more than edit files,
etc. it is going to be hard for Ada to compete unless the same sort of
things are available. Saying "Well you can download X from
http://whozits.com/ and Y from http://whatchamacallit.org/ and Z from
http://somebodyorother.net/ and go compile it yourself and maybe with a few
thousand lines of glue software and a few hundred hours of tinkering you can
kind of have an 'integrated' environment - and oh, by the way, when you get
done with all that code, please be good enough to put it under the GPL and
send it to us so we can package it with our stuff and go sell it to someone
else..." - its not quite the same thing to the buyer as "Here's your 'B-'
compiler with everything you need or want to build apps on platform
Whatever. Where's my check for $29.95?" For a niche language with a bunch of
enthusiastic hobbyists, hackers & students, maybe the former is "Good
Enough" - but then that's where Ada is going to stay.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Stefan Skoglund" <stetson@ebox.tninet.se> wrote in message
news:3B61C4AF.5B5F4694@ebox.tninet.se...
> Marin David Condic wrote:
> > along the lines of "luxury" items in a car since Ada provides all the
> > essentials that any programming language needs to provide. (No missing
seats
>
> When Ada is the Rolls-royce company of old.
> You could buy your rolls ready-made from the factory
> or with no carriage/seats but technically complete
> and then commission your local horse-wagoon manufacturer.
>
> It exist a number of rolls here in Sweden with Stockholm-made
> carriage.

Re: Java portability (was: An Ada IDE and discussions)

[Lao Xiao Hai]

nicolas wrote:

> "Marc A. Criley" <mcqada@earthlink.net> a �crit dans le message news:
> 3B5DCE74.C12AA2D8@earthlink.net...
> > nicolas wrote:
> >
> > Though I would doubt that _Ada_ Core Technologies would drop Ada any
> > time in the near future :-)
>
> I doubt it too ...
> Unfortunately, at least in Windows world, this is the only one company for
> which you can be quite certain about that.

I am currently advisor on a project that will be presented at this year's
SigAda which has been written in GtkAda.   I originally suggested the
developers
consider CLAW, but they decided they wanted platform independence.  That
is, they want their application to run on Wintel, Linux, or any variation
of Unix or anything else they could name.

The result of their effort is quite good.  Not only have they achieved the
desired
level of portability, one of them told me last week they would not have been
able to accomplish the same quality had they limited themselves to Java.  As
it
stands now, their source code compiles, unchanged, on an NT machine or a
Linux machine and runs exactly the same on both.   This has impressed one of
our colleagues who originally intended to use JGNAT to the point that he has
decided to explore the potential of GtkAda instead.

I would be nice if there were a version of GtkAda for Windows CE or the
Palm OS, but that is probably a long way off since we don't have Ada compilers

for those operating systems either.

Richard Riehle
AdaWorks Software Engineering
richard@adaworks.com

Re: Java portability (was: An Ada IDE and discussions)

[Stefan Skoglund]

Lao Xiao Hai wrote:
> stands now, their source code compiles, unchanged, on an NT machine or a
> Linux machine and runs exactly the same on both.   This has impressed one of

They should try to get a not-new SGI with
a good 2D accelerated graphics adapter.

Gtk+ works very well on IRIX so should GtkAda too.

Re: Java portability

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1129 bytes --]

"Warren W. Gay VE3WWG" <ve3wwg@home.com> a �crit dans le message news:
3B619A6D.5DD6E782@home.com...
> I'd like to offer a slightly different "analogy" :
>
> _flamesuit on_
>
> The shopper looks at the Ada car, and notices that it has seatbelts and
> of course 4 doors on it. The programmer says, but I don't want to be
> restrained from accessing the outdoors -- I go through drive-throughs
> a lot..
>
> The shopper than goes down the street and settles on the C++ car
> which does not have the restrictive seat belts, and of course,
> _NO_ doors to get in the way ;-)  The Java car has seatbelts, but
> no doors...
>
> _flamesuit off_
>

I wouldn't see good and easy to use debuggers, standard libraries, GUI, IDE
coming with the compiler, etc  ...
as tools going against safety concerns ....
Good luck if you want to convince Java or C++ programmers ... It could be a
lot of fun, but I'm not sure they will even listen to you.
The point is that, whatever good is the language, I dont see how Ada could
justify not to provide elementary stuff expected by almost every programmer.
There is no safety concern about that.

Re: Java portability (was: An Ada IDE and discussions)

[Dave Adlam]

I seem to remember a post in the past (about the time JGNAT came out) from
Robert Dewar that suggested that the JGNAT Java Pragmas followed a draft AI.

This is a vague recollection so please let me apologise in advance if I got
that wrong (especially to Robert Dewar).

If you go to the jgnat_ug HTML file in the docs directory of your jGNAT
installation and follow the following series of links

What This Guide Contains ->Java-Specific Pragmas -> Handling Circularities
using With-Type Clauses -> A Word of Caution

you will find a statement on this issue.

Marc A. Criley wrote in message <3B5D5B79.F2DC527E@earthlink.net>...
>nicolas wrote:
>>
>> We never tried Ada for the JVM
>> Is there a standard specification when you want to access Java libraries,
or
>> is it specific for each vendor (Ada-Magic, JGNAT, others ?)
>> If we write Ada code for the JVM, will it be the same for all compiler
>> vendors which support it ?
>
>Since there's no officially recognized Java standard yet, there's no Ada
>standard specification for interfacing to it (as there is for C,
>Fortran, and COBOL, and just recently for C++), so vendors are free to
>forge their own specification.
>
>I'm nearly certain that each vendor has their own means of interfacing
>to the Java libraries.  I've not looked at the specifics of how AdaMagic
>does it, but I know that JGNAT added Java-specific pragmas to support
>interfacing.  In addition, the documentation describing how Java
>Interfaces are implemented (as discriminants) suggests some "magic"
>going on inside the compiler.
>
>It's not the _Ada_ code for a JVM that will differ between vendors, it's
>the interfacing to the class libraries.  I would expect vanilla Ada code
>to be quite portable, but the interfaces to Java classes will resemble
>one another, and differ in the details.
>
>Marc A. Criley

Re: Proving Correctness (was Java Portability)

[Warren W. Gay VE3WWG]

nicolas wrote:
> "Warren W. Gay VE3WWG" <ve3wwg@home.com> a �crit dans le message news:
> 3B619A6D.5DD6E782@home.com...
> > I'd like to offer a slightly different "analogy" :
> >
> > _flamesuit on_
> >
> > The shopper looks at the Ada car, and notices that it has seatbelts and
> > of course 4 doors on it. The programmer says, but I don't want to be
> > restrained from accessing the outdoors -- I go through drive-throughs
> > a lot..
> >
> > The shopper than goes down the street and settles on the C++ car
> > which does not have the restrictive seat belts, and of course,
> > _NO_ doors to get in the way ;-)  The Java car has seatbelts, but
> > no doors...
> >
> > _flamesuit off_
> 
> I wouldn't see good and easy to use debuggers, standard libraries, GUI, IDE
> coming with the compiler, etc  ...
> as tools going against safety concerns ....

It's difficult for me to understand what the precise jist of your message
is here. One of the things you have meantioned here, suggest that you're
implying that YFL (your favourite language) provides a debugger, and so
that qualifies it to be just as safe as Ada. Am I reading too much into this?

This will be the only issue that I'll respond to here:

First of all, there are debuggers for Ada. The real issue however is more
along these lines, that an example might best highlight WRT safety :

Let's say your country has contracted two individuals to write a text based
chess program, that will compete in the global Chess Olympics (you are one
of the ones chosen and free to use YFL). The committee
is not sure what the implementation should be, but they want the 
final implementation to be rock solid, and
to not only obey all the rules of play under all circumstances, 
but it must not core dump.

Further, it must carry out an algorithm that 
has been designed by your country's sharpest chess master.

So you write a text based chess game in YFL (your favourite
language, but not Ada), according to the committees requirements.
You write the code, and eventually go through several iterations 
of testing in and out of YFD (your favourite debugger).

Now, another programmer uses Ada to accomplish the same task. He may well 
use a debugger, but he might not need it. This largely depends upon the 
skill of the programmer, of course, in any language. (This is just more
likely in Ada, but we can ignore this if it bothers you).

Now, both implementations are considered "complete" by their programmers,
and the sponsoring committee wants to have assurances from both, about
their correctness and readiness to carry out the chess match. The committee
must now choose which implementation to use. If the chosen implementation
should core dump, break a rule or fail performance wise, the country will
lose the match, with "egg on face." The chess master behind the software
chess algorithms will also lose face. Consequently, the committee is 
eager to choose the best implementation (it is "mission critical").

The non-Ada person has to submit that his program is correct on the basis
of testing. But the sponsoring committee asks "but how can you be sure
that you've done enough testing?" You might answer "there are too many
possibilities to check.. I cannot check all possibilities, but my testing
was _extensive_."

You're basically expecting to have the committee accept your implementation
on the basis that you tested "extensively" (the qualifications of the two
programmers where judged equal at the start).

The Ada person has the same challenge. He answers the sponsoring committee
that "I have tested _extensively_ and found no faults". But he can add
that "I have also had the source code audited by Ada professionals. This
has been done over and above what the 'computer' has checked at compile
time, during the project's development <insert Ada strengths about
strong type checking, module [package] isolation etc. etc.>."

The bottom line here is that Ada provides much greater assurances about
software correctness, than any other language that I am aware of. It is
also quite easy to read, making it easy to audit. Auditing C/C++ is a
nightmare, and easily error prone for humans. Java is an improvement, 
but still falls short.

In short, given that both implementations tested OK in front of the
committee, the committee chose to go with the Ada implementation. It's
not hard to see why.

The time is now past when people are concerned about efficiency. If it
is not already here, it will be here in a few more years as CPUs jump
another order of magnitude or so, forward in speed. 

As software becomes bloated by another order of magnitude (which ususally
follows CPU speed), then we'll be even more concerned about software 
reliability than ever before. We have built "towers of babel" on C/C++
frameworks. This is costly in security terms, and for reliability. If
we keep building "higher towers", we'll need more reliable footings
on which to build.

People abandoned assembly language for operating systems for the same
reasons. The time is coming when people will abondon C for operating
systems for "safer language".

Whether for applications or operating systems, I believe the time is
coming soon when people will be forced to look at Ada as an option
for building that next "big system".  People are already scratching
their heads in this area. At some point, the light bulb will start
going on and a few more people will discover Ada.  It is a language
that has already been designed to solve the problems that others are
looking for solutions in. It has the advantage that it has a lot
of experience, which is something that a designer of a new language
would be loath to throw away.

Some already have figured out, that Ada is a "good thing". More
will follow.

> Good luck if you want to convince Java or C++ programmers ... It could be a
> lot of fun, but I'm not sure they will even listen to you.

I was poking fun. I have to use C/C++ in my day job.

-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3190 bytes --]

"Warren W. Gay VE3WWG" <ve3wwg@home.com> a �crit dans le message news:
3B6636BA.96FD8348@home.com...
> > I wouldn't see good and easy to use debuggers, standard libraries, GUI,
IDE
> > coming with the compiler, etc  ...
> > as tools going against safety concerns ....
>
> It's difficult for me to understand what the precise jist of your message
> is here. One of the things you have meantioned here, suggest that you're
> implying that YFL (your favourite language) provides a debugger, and so
> that qualifies it to be just as safe as Ada. Am I reading too much into
this?
>

Not at all, just saying that those tools don't go against safety, and so
that safety cannot be an excuse for their absence.

First note that we've been writing 99% of our code in Ada since around 1987,
so we are not especially new to Ada, nor denying its qualities.
I would also say that it is not a very good thing to have a "favourite
language", because a good programmer should choose a language as a tool for
a specific purpose, not for his personal feelings. It is nothing more than a
tool.
A very big problem of Ada, is that a lot of Ada users are so fan of it, that
they deny obvious problems which prevent its wider use, and don't get fixed.

The problem is more about IDE, GUI and libraries, I know there are excellent
debuggers for Ada
we use or have been using Alsys, Objectada, Apex, Verdix etc ...
we tried GVD, which seems very promising, but still wondering how to display
the value of a string variable (not always easy in -O0, almost impossible
in -O1 or -O2 ...)
This is an excellent example of amazing capabilities, and incredible
elementary problems as well.

To get back to the the original point, it was there is no integrated and
frendly complete development kit for Ada, comparable to what any debutant
find for popular languages.
And unfortunately, a common Ada fan reaction to this is to give examples to
show that Ada is safer than other languages.

1/ This is something every Ada user, and a lot of other people even if they
are not using Ada, aknowledge,
2/ It has absolutely nothing to do with the question.

The question could be
- Ada has a lot of excellent qualities, but it is everything but popular.
The lack of integrated environment and standard libraries coming with the
compiler has certainly something to do with it.

The answer was
- Ada is safer ...

Yes, I know it is safer, everybody knows, that's not the question, and if
this is the only answer which can be given to someone thinking  in using
Ada, I think he'll wait for a real answer to his question.

Even among Ada users I saw more than once this kind of situation :

- Hey, there is a problem
- No, there is no problem, you must be wrong
- No there is a problem, here it is, how can I solve it ?
- you must be doing it wrong ... (No answer to the problem ...)
- Here is what, according to me, proves there is problem, but may be I
missed something ...
- Are you sure you really need to do that ? (change subject ...)
- Yes I do, I think there is a problem and nobody gives a solution ...
- Ada is much safer  (So what ? we all know that ... nothing more about the
problem ... end of the discussion ...)

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

I understood what you were getting at: Having or not having tools doesn't
have anything to do with safety issues, but it does have to do with helping
to make the language more popular.

There is a certain amount of advocacy that goes on here - I am as guilty as
the next guy. :-) I agree that it doesn't really help anything to avoid
discussing the issues that may be keeping Ada from gaining acceptance.
Changing the subject and concentrating entirely on where Ada is superior
tends to not get the other issues addressed. Having a nice integrated
toolset including class libraries, GUI builder, debugger, IDE, etc. are all
things that would help make Ada more popular. Discussion about what those
tools should look like, how they should get made, how they should get
distributed, etc., are all going to contribute something positive.

It also helps to take the advocacy from C.L.A. and apply it elsewhere so we
aren't just preaching to the choir.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"nicolas" <n.brunot@cadwin.com> wrote in message
news:TLt97.9335$Xa.790630@nnrp4.proxad.net...
>
> Not at all, just saying that those tools don't go against safety, and so
> that safety cannot be an excuse for their absence.
>
> First note that we've been writing 99% of our code in Ada since around
1987,
> so we are not especially new to Ada, nor denying its qualities.
> I would also say that it is not a very good thing to have a "favourite
> language", because a good programmer should choose a language as a tool
for
> a specific purpose, not for his personal feelings. It is nothing more than
a
> tool.
> A very big problem of Ada, is that a lot of Ada users are so fan of it,
that
> they deny obvious problems which prevent its wider use, and don't get
fixed.
>
> The problem is more about IDE, GUI and libraries, I know there are
excellent
> debuggers for Ada
> we use or have been using Alsys, Objectada, Apex, Verdix etc ...
> we tried GVD, which seems very promising, but still wondering how to
display
> the value of a string variable (not always easy in -O0, almost impossible
> in -O1 or -O2 ...)
> This is an excellent example of amazing capabilities, and incredible
> elementary problems as well.
>
> To get back to the the original point, it was there is no integrated and
> frendly complete development kit for Ada, comparable to what any debutant
> find for popular languages.
> And unfortunately, a common Ada fan reaction to this is to give examples
to
> show that Ada is safer than other languages.
>
> 1/ This is something every Ada user, and a lot of other people even if
they
> are not using Ada, aknowledge,
> 2/ It has absolutely nothing to do with the question.
>
> The question could be
> - Ada has a lot of excellent qualities, but it is everything but popular.
> The lack of integrated environment and standard libraries coming with the
> compiler has certainly something to do with it.
>
> The answer was
> - Ada is safer ...
>
> Yes, I know it is safer, everybody knows, that's not the question, and if
> this is the only answer which can be given to someone thinking  in using
> Ada, I think he'll wait for a real answer to his question.
>
> Even among Ada users I saw more than once this kind of situation :
>
> - Hey, there is a problem
> - No, there is no problem, you must be wrong
> - No there is a problem, here it is, how can I solve it ?
> - you must be doing it wrong ... (No answer to the problem ...)
> - Here is what, according to me, proves there is problem, but may be I
> missed something ...
> - Are you sure you really need to do that ? (change subject ...)
> - Yes I do, I think there is a problem and nobody gives a solution ...
> - Ada is much safer  (So what ? we all know that ... nothing more about
the
> problem ... end of the discussion ...)
>
>
>
>

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2227 bytes --]

"Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit dans
le message news: 9k6aug$mtq$1@nh.pace.co.uk...
> I understood what you were getting at: Having or not having tools doesn't
> have anything to do with safety issues, but it does have to do with
helping
> to make the language more popular.
>

And not only more popular ... It would be also extremely profitable for
current Ada users.
How many Ada programmers are looking around for easy interface with whatever
you can easily find in popular languages IDEs
sometimes you find several of them, good ones, bad ones (which one to choose
?)
sometimes it simply not exists, good luck for endless and tedious pragma
interface (most common situation when you buy a commercial C library)

This is a complete loss of time and a boring job, going against any basic
reason why Ada users made the Ada choice.
Where are the library components, reuse etc ... ?

Some Ada fans seems to deny that, without having written a single big enough
Windows application, meeting interface requirements of a normal PC user.
They should really have a look of what is available elsewhere
They should really try to start from scratch a simple graphical Windows
application
really with nothing more that what comes with their compiler, not using
other tools they spent monthes to find or write themselves.
I bet any Java debutant can provide a working application before they have
made the choice of their GUI ...

We don't have this kind of problem because we have had our own tools for a
lot of years, but I fully understand reluctance of people starting today
with Ada, and realizing what they miss compare to popular languages.
Most applications cannot afford such a loss of time, even if it improves
safety. Most softwares are not embedded critical ones.

The first thing to do, is starting to aknowledge that people using popular
languages are not completely stupid, that their concerns are valid ones, and
that there is absolutely no valid reason to miss basic standard tools.
After all they release a lot of excellent products, and if there is a lot of
junk software, may be it's also because they release hundreds of products
while Ada programmers release just one.

Re: Java portability

[John Doe]

> The shopper looks at the Ada car, and notices that it has seatbelts and
> of course 4 doors on it. The programmer says, but I don't want to be
> restrained from accessing the outdoors -- I go through drive-throughs
> a lot..

I don't really need a hood either.  Slows me down getting to the engine.
It's my job, not Detroit's, to keep birds and children from falling into
it.  Huh?  Thieves might steal the engine?  Of course not!  I'll install
a security system called lint and I'll never forget to turn it on.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau

Re: Java portability

[Robert Dewar]

> I'm nearly certain that each vendor has their own means of
> interfacing to the Java libraries.  I've not looked at the specifics 
> of how AdaMagic does it, but I know that JGNAT added Java-specific 
> pragmas to support interfacing.

Well it is amazing to see how often the blind lead the blind on CLA
:-)

In fact there are *no* "Java-specific" pragmas in GNAT, and the entire
discussion of these imaginary beasts is a tribute to people's
willingness to believe anything they read :-)

Note that anyone can look this up in the GNAT RM, just search for the
word Java, you will not find it.

Re: Java portability

[Warren W. Gay VE3WWG]

John Doe wrote:
> 
> > The shopper looks at the Ada car, and notices that it has seatbelts and
> > of course 4 doors on it. The programmer says, but I don't want to be
> > restrained from accessing the outdoors -- I go through drive-throughs
> > a lot..
> 
> I don't really need a hood either.  Slows me down getting to the engine.
> It's my job, not Detroit's, to keep birds and children from falling into
> it.  Huh?  Thieves might steal the engine?  Of course not!  I'll install
> a security system called lint and I'll never forget to turn it on.
> 
> --
> Wes Groleau
> http://freepages.rootsweb.com/~wgroleau

I used lint in the old days for C. Now all the good warnings and checks
are already part of GCC these days, and many newer compilers. Of course,
this is not universal (HP comes to mind).

Note however, lint has no advantage over a regular Ada compile. 
It's not even in the "running".

-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Java portability

[Larry Kilgallen]

In article <5ee5b646.0108011843.12cc0e9a@posting.google.com>, dewar@gnat.com (Robert Dewar) writes:
>> I'm nearly certain that each vendor has their own means of
>> interfacing to the Java libraries.  I've not looked at the specifics 
>> of how AdaMagic does it, but I know that JGNAT added Java-specific 
>> pragmas to support interfacing.
> 
> Well it is amazing to see how often the blind lead the blind on CLA
> :-)

It has been proven effective for drawing others out of their shell :-)

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Warren W. Gay VE3WWG" <ve3wwg@home.com> a ?crit dans le message news:
: 3B6636BA.96FD8348@home.com...
: 
: The problem is more about IDE, GUI and libraries, I know there are excellent
: debuggers for Ada [...]
: This is an excellent example of amazing capabilities, and incredible
: elementary problems as well.
: 
: To get back to the the original point, it was there is no integrated and
: frendly complete development kit for Ada, comparable to what any debutant
: find for popular languages.
: And unfortunately, a common Ada fan reaction to this is to give examples to
[safety]

You write about debutants, could you provide some reasoning why for
these people there should be a more full grown environment than AdaGIDE?
Is the Aonix product so far off for this group of people???


You could of course say, pressing F9 to get a runnable executable
is so much easier that getting to know Emacs & Co conventions (or whatever)
but that looks like a shortsighted self deception to me, at best.
Can you provide any convincing empirical data to show that a
Click-1-Button solution is better _for debutants_ than a
Get-to-Know-Your-Tools craftspeople approach? WRT Short term,
long term, programmer productivity, general knowledge of what
compilation tools do, ease of adaption to different
tools on different platforms... (Yes I know the lament of
debutants about the additional percentage of time devoted
to studying compiler/IDE docs, but I also know debutants and programmers
who actually _enjoyed_ learning about their toolset, because
of the interesting concepts and "visible" machinery. (And I'm
not debating the merits of powerful environments.)

To give an example, creating a project in your favourite Windows
IDE is possibly "just a matter of dragging this here and that
there and then chosing that from this drop down list and clicking
bang" (depends on you IDE, this was Visual Age for C++:-) But
this does in fact need learnuing as with every IDE (which have
extensive documentation, so can't be that simple really), and
likely entails non-portable build scripts (because the Click-1
programmer wouldn't know how to translate to a portable one, as there
is no button for this). (side note: I can't help thinking this
"Gimme an _easy_ IDE" thing has similar annoing side effects as
being proud on having autoconfed otherwise portable software.
Or as thinking "I know how to use my Word processor to its best
because I can see the printouts."  Ever compiled a book from
more than one authors, all using the "same" Word processor? ;-)

: - Ada has a lot of excellent qualities, but it is everything but popular.
: The lack of integrated environment 
[claimed by the author, but see above]
: and standard libraries coming with the
: compiler has certainly something to do with it.

What exactly? Certainly C++ became popular before there was
today's STL. Then, certainly many Ada compilers come with libraries,
since the RM has so much to say about them, e.g. for string
processing. So I guess you mean container libraries or similar?
Standard Database Interfaces? Numerical Algorithms? Graphics
libraries? Encryption libraries?

I'm sure you know both the old and the new Booch components
and see that the requested libraries have been there for a long
time. Only they are decoupled from each vendors compiler and
are free for download instead. Compare them to their equivalents
(more or less) in Java 2 (not 1!).
Would you agree that Ada will become more popular if every ADA IDE
comes with a sticker
		"Prebuilt
	Booch Components Inside"? 
Could be a point...



--  Georg
---
Microsoft Windows--a fresh perspective on information hiding

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a ?crit dans
: le message news: 9k6aug$mtq$1@nh.pace.co.uk...

: Where are the library components, reuse etc ... ?

You're not seriously asking this?

: Some Ada fans seems to deny that, without having written a single big enough
: Windows application, meeting interface requirements of a normal PC user.


: They should really try to start from scratch a simple graphical Windows
: application
: really with nothing more that what comes with their compiler, not using
: other tools they spent monthes to find or write themselves.

'Monthes' is by far an order of magnitude too large
for the time it took me.
As always, if you start at http://www.adapower.com...


: I bet any Java debutant can provide a working application before they have
: made the choice of their GUI ...

Choosing a GUI is a time consuming procedure if you choose not
by taste, fashion, prestige, or hearsay, but by comparison,
reading docs, feature lists, about extensibility, integration
with 3rd party tools, etc.
So that only shows that IDE can be complex, despite marketing
departement efforts to make it look easy, geared towards the
momentarily lazy, wishfully thinking programmer..


(side note: given many approaches to GUIs, it is
a question whether the choice of "standard" Windows
applications (a very time dependent term in the Windows world :-)
is appropriate for each and every job. See non-standard, but 
heavily used mail clients for an example.
(And the ever changing mouse interface to system administration
tasks drives Administrators crazy. I've been told you can get
a nice text driven programmable interface form M$ for some $$$$. :-)


: The first thing to do, is starting to aknowledge that people using popular
: languages are not completely stupid, that their concerns are valid ones,

Yes, valid, but not technical ones for the most part, if you
accept that multi language programs are a reality.  As you
implied, it took someone time to find libraries, you find it
inconvenient to not be given beforehand what you might be after,
and more.  Indeed if you can chose your tool collection
from various providers, there might be some work ahead, like
configuring a more recent GTK+ in a non-standard directory,
if necessary, say. Tedious.  An alternative is an all-in-one
solution from one provider, most likely M$. Good for quick hacks,
and one time products, but, to stress the term "standard" in M$
universe, for investment in long term projects?

All these are convenience concerns, aren't they.

: that there is absolutely no valid reason to miss basic standard tools.

Could you define this term, please?

WRT to GUI, there is good and valid reason on the ground
of user interface research (which is an activity not constrained
to the current Windows GUI or to computer science),
and it is done, see touch screens in
shops or train stations. They may or may not be windows like.

If you restrict Ada programs to "standard
Windows applications" and base your observations cencerning
Ada programming environments on this,
this isn't really a broad base for you arguments?

-- Georg

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 4642 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9kb3ub$hdo$1@a1-hrz.uni-duisburg.de...
> You write about debutants, could you provide some reasoning why for
> these people there should be a more full grown environment than AdaGIDE?

I don't write only about debutants. Exactly the reasoning apply to
experimented programmers.
What I meant is that, in those languages, even debutants can find very
easily simple tools,
that experimented programmers are (or at least should be) very happy to
find.

> Is the Aonix product so far off for this group of people???

Aonix is smart enough to provide something looking like Visual C++, Apex is
not bad too.
But when people get stuck with one compiler because of the absence of
standard, while they hear about Ada concerns for reuse, portability etc ...
I understand they can be somewhat surprised ...

> You could of course say, pressing F9 to get a runnable executable
> is so much easier that getting to know Emacs & Co conventions (or
whatever)
> but that looks like a shortsighted self deception to me

Of course I could say !!  and by the way, most popular languages programmers
say it ...
They are not so stupid, because it is really much easier ...
However experimented I am, I have a very strange way of thinking :
When something is quite complicated and require experimented people,
if there is a way to safely simplify it so that a debutant can do it,
the experimented guy is likely to go much faster and safer using the simple
solution.
If he is not, give the job to the debutant ...

> (Yes I know the lament of
> debutants about the additional percentage of time devoted
> to studying compiler/IDE docs, but I also know debutants and programmers
> who actually _enjoyed_ learning about their toolset, because
> of the interesting concepts and "visible" machinery. (And I'm
> not debating the merits of powerful environments.)

Remember what Marin said about softwares
You should put the CD, see a standard install procedure pop up, and find
menus "file", "edit", "tools" in their expected palce, with expected
functions and behavior.
When standard and expected things are found, only after that, check the docs
to go further,
If not just throw the software away. There is no excuse.
He is right, and almost every software users do the same.
May be the problem of Ada language is that such an evidence is not an
evidence for too many Ada users.

> : - Ada has a lot of excellent qualities, but it is everything but
popular.
> : The lack of integrated environment
> [claimed by the author, but see above]

The fact that I claim it has absolutely no importance, we already use Ada,
and I am just one person.
The important point is that it is claimed by 99% of people who could think
in using Ada
If your product doesn't satisfy them, change the product or they'll forget
about your poduct, and they will be very wise to do so.
They know what they want, if you do not provide it, and if others provide
it, no hesitation ...
Every industry has understood such elementary evidence, software industry
started to understand it, those who don't understand will disappear ...

> What exactly? Certainly C++ became popular before there was
> today's STL. Then, certainly many Ada compilers come with libraries,
> since the RM has so much to say about them, e.g. for string
> processing. So I guess you mean container libraries or similar?
> Standard Database Interfaces? Numerical Algorithms? Graphics
> libraries? Encryption libraries?

Just have a look at a Java compiler. It is self-explanatory ...

>Only they are decoupled from each vendors compiler and
> are free for download instead.

May be the heart of the problem :-)

>Compare them to their equivalents
> (more or less) in Java 2 (not 1!).

Did you see how fast Java figured out the problem of providing libraries  ?
Once again compare a Java compiler and any Ada compiler.
while we hear on c.l.a. endless discussions about syntax, and that 18 year
after Ada83 and 6 years after Ada95 ?
Don't you think there could be some lesson to learn ?

> Would you agree that Ada will become more popular if every ADA IDE
> comes with a sticker
> "Prebuilt
> Booch Components Inside"?
> Could be a point...

Could be ?
Let's be serious
Everybody knows what popular languages compilers provide
Everybody knows what Ada compilers provide
Everybody can compare how widely they are used.

I cannot understand why so much time is lost in useless discussion while the
gap is growing.

Almost any software company knows that the marketing and the packaging of
the product is at least as important as its qualities.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3157 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9kb7s9$hr3$1@a1-hrz.uni-duisburg.de...
> : Where are the library components, reuse etc ... ?
> You're not seriously asking this?

Of course I am, there are not coming with the compiler ...
What means the I in IDE ???

> 'Monthes' is by far an order of magnitude too large
> for the time it took me.
> As always, if you start at http://www.adapower.com...
> Choosing a GUI is a time consuming procedure if you choose not
> by taste, fashion, prestige, or hearsay, but by comparison,
> reading docs, feature lists, about extensibility, integration
> with 3rd party tools, etc.

'minutes' is by far an order of magnitude too large for a lot of people ...
Oh, I imagine a software company, after long and smart studies, deciding to
release their softwares exclusively for OS/2 ...
(If OS/2 doesn't suit you, put whatever you want except Windows Mac, and
Linux ...)
Seriously, they should be a common standard GUI with every compiler, so that
people who don't want to look elsewhere have something easy and ready.
I cannot imagine how this could prevent you to choose carefully another GUI
if you feel like it ...

> (side note: given many approaches to GUIs, it is
> a question whether the choice of "standard" Windows
> applications (a very time dependent term in the Windows world :-)

A lot of things are like "standard" Windows on my linux machine ...
I'm happy with that.
I also like to see brakes at the same place on most cars and motorbikes I
ride, without spending hours reading the doc ...

> Yes, valid, but not technical ones for the most part, if you
> accept that multi language programs are a reality.

I had never thought in even asking the question ...
I sometime hear about the existence of strange exotic languages which are
not Ada ...

 >An alternative is an all-in-one
> solution from one provider, most likely M$. Good for quick hacks,
> and one time products, but, to stress the term "standard" in M$
> universe, for investment in long term projects?

I would to see an equivalent of Java Swing for Ada, nothing more.

> : that there is absolutely no valid reason to miss basic standard tools.
>
> Could you define this term, please?

Once more, check a Java compiler ...
People who do not use Ada won't spend hours explaining what everybody expect

> If you restrict Ada programs to "standard
> Windows applications" and base your observations cencerning
> Ada programming environments on this,
> this isn't really a broad base for you arguments?

My arguments rely on the fact that Ada programs should at least aknowledge
that Windows applications cannot be ignored.
You cannot call that a restriction, if you agree that Ada programs should
not be restricted to one platform application

Today being multi-platform means being fully compliant and comparable to
other products with :
first Windows
then Mac and Linux
after that whatever you want, the more you have, the best it is.

Astonishingly, in Ada world, Windows seems not to be at the beginning of the
list
(because of anti-MS quasi religious concerns ?)
May be that explains everything ....

Re: Java portability

[Marc A. Criley]

Robert Dewar wrote:
> 
> > I'm nearly certain that each vendor has their own means of
> > interfacing to the Java libraries.  I've not looked at the specifics
> > of how AdaMagic does it, but I know that JGNAT added Java-specific
> > pragmas to support interfacing.
> 
> Well it is amazing to see how often the blind lead the blind on CLA
> :-)
> 
> In fact there are *no* "Java-specific" pragmas in GNAT, and the entire
> discussion of these imaginary beasts is a tribute to people's
> willingness to believe anything they read :-)

I do not recall stating there were "Java-specific" pragmas in GNAT, I
believe the text reads "JGNAT added Java-specific pragmas to support
interfacing."

To wit, from the JGNAT 1.1p User Guide:
  pragma Java_Constructor (function-name);

Please accurately read my postings to avoid attributing statements to me
that I did not make.
> 
> Note that anyone can look this up in the GNAT RM, just search for the
> word Java, you will not find it.

But you will find it throughout the JGNAT User Guide.

Marc A. Criley
Senior Staff Engineer
Quadrus Corporation
www.quadruscorp.com

Re: Proving Correctness (was Java Portability)

[Jacob Sparre Andersen]

Nicolas:

> Everybody knows what popular languages compilers provide

I am not sure I do, but I will try to make a list and guess
which things you consider important. I work on Digital Unix
and Linux systems.

 * Ada     - the standard libraries and some vendor
dependent stuff (GNAT.*)

 * Basic          - don't know
 * C              - the standard libraries
 * C++            - the standard libraries
 * COBOL          - don't know
 * Fortran        - the standard libraries
 * Java           - don't know
 * Borland Pascal - integrated GUI builder and DB builder,
no _standard_ libraries

I addition to this you can install GUI builders and various
graphical and text based development environments and
debuggers as well as additional libraries for practically
any purpose you can think of.

Borland Pascal (aka Delphi) seems to be the most popular
programming language in Denmark (before C) even though it
until recently only was available for a single
processor-operating system combination, so I assume that
most programmers consider an integrated GUI builder and/or
database builder the most important factor, when choosing a
programming language. Apparently you don't even have to live
up to the standard your language is named after.

So the answer seems to be:

  "Make an _integrated_ package consisting of:

    * An Ada compiler.
    * A GUI builder (where you can go _freely_ back and
forth between graphical and source view).
    * A database builder.
    * Whatever libraries you can think of."

<slightly off-topic rant>
My short experience with Object Ada gave me the impression
that Aonix for some reason had decided to use a rather
primitive parser (even the number of spaces appeared to be
important in some places) in stead of using the one from the
compiler.
</slightly off-topic rant>

Jacob
-- 
Det virker p� en eller anden m�de absurd at det anses for
god tone at sige
at "vi har sundheds/skole/social..-systemet for at give
erhverslivet gode
konkurrencevilk�r", mens det anses for samfundsomstyrtende
at foresl� at
"erhverslivet hovedopgave er at financiere
velf�rdssystemer".
-- Sune Rastad Bahn

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Just for the record, this is not me. The cut you did with my name in it must
have come from an earlier message.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:9kb7s9$hr3$1@a1-hrz.uni-duisburg.de...
> nicolas <n.brunot@cadwin.com> wrote:
> : "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a ?crit
dans
> : le message news: 9k6aug$mtq$1@nh.pace.co.uk...
>
> : Where are the library components, reuse etc ... ?
>
> You're not seriously asking this?
>
> : Some Ada fans seems to deny that, without having written a single big
enough
> : Windows application, meeting interface requirements of a normal PC user.
>
>
> : They should really try to start from scratch a simple graphical Windows
> : application
> : really with nothing more that what comes with their compiler, not using
> : other tools they spent monthes to find or write themselves.
>
> 'Monthes' is by far an order of magnitude too large
> for the time it took me.
> As always, if you start at http://www.adapower.com...
>
>
> : I bet any Java debutant can provide a working application before they
have
> : made the choice of their GUI ...
>
> Choosing a GUI is a time consuming procedure if you choose not
> by taste, fashion, prestige, or hearsay, but by comparison,
> reading docs, feature lists, about extensibility, integration
> with 3rd party tools, etc.
> So that only shows that IDE can be complex, despite marketing
> departement efforts to make it look easy, geared towards the
> momentarily lazy, wishfully thinking programmer..
>
>
> (side note: given many approaches to GUIs, it is
> a question whether the choice of "standard" Windows
> applications (a very time dependent term in the Windows world :-)
> is appropriate for each and every job. See non-standard, but
> heavily used mail clients for an example.
> (And the ever changing mouse interface to system administration
> tasks drives Administrators crazy. I've been told you can get
> a nice text driven programmable interface form M$ for some $$$$. :-)
>
>
> : The first thing to do, is starting to aknowledge that people using
popular
> : languages are not completely stupid, that their concerns are valid ones,
>
> Yes, valid, but not technical ones for the most part, if you
> accept that multi language programs are a reality.  As you
> implied, it took someone time to find libraries, you find it
> inconvenient to not be given beforehand what you might be after,
> and more.  Indeed if you can chose your tool collection
> from various providers, there might be some work ahead, like
> configuring a more recent GTK+ in a non-standard directory,
> if necessary, say. Tedious.  An alternative is an all-in-one
> solution from one provider, most likely M$. Good for quick hacks,
> and one time products, but, to stress the term "standard" in M$
> universe, for investment in long term projects?
>
> All these are convenience concerns, aren't they.
>
> : that there is absolutely no valid reason to miss basic standard tools.
>
> Could you define this term, please?
>
> WRT to GUI, there is good and valid reason on the ground
> of user interface research (which is an activity not constrained
> to the current Windows GUI or to computer science),
> and it is done, see touch screens in
> shops or train stations. They may or may not be windows like.
>
> If you restrict Ada programs to "standard
> Windows applications" and base your observations cencerning
> Ada programming environments on this,
> this isn't really a broad base for you arguments?
>
> -- Georg

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2055 bytes --]

"Jacob Sparre Andersen" <sparre@nbi.dk> a �crit dans le message news:
3B695521.54BBF4D3@nbi.dk...
> Nicolas:
>
> > Everybody knows what popular languages compilers provide
>
> I am not sure I do, but I will try to make a list and guess
> which things you consider important. I work on Digital Unix
> and Linux systems.

Well I certainly exagerated.
What I meant actually is that Windows and Mac are likely to be the only
platforms where one can really figure out, how high are the requirements of
interactivity and ease of use for software users.

You don't really have a high level user interface for embedded systems.
HP Sun Irix and others unixes are extremely rare for workstations
Linux is still not really ready for a PC you can buy in a supermarket.
For example I think that RH 7.1 is a regression regarding RH 7.0
(XConfigurator or Linuxconf are likely to ruin your system without any
warning)
Talking about XConfigurator, any method to configure your screen (colors,
resolution etc ...)  under Linux is totally unacceptable for a Mac or
Windows user.

In fact what you need is not so complicated to define.
Choose some small application where graphical user interface is important.
Try to develop it a C/C++  version
Try to develop a Java version
Try to develop an Ada version

Anything you have to look for in Ada, and you don't have to look for in Java
or C is definitely something which must be corrected.

>  * Ada     - the standard libraries and some vendor
> dependent stuff (GNAT.*)
>
>  * Basic          - don't know
>  * C              - the standard libraries
>  * C++            - the standard libraries
>  * COBOL          - don't know
>  * Fortran        - the standard libraries
>  * Java           - don't know
>  * Borland Pascal - integrated GUI builder and DB builder,
> no _standard_ libraries

Let 's be honest, real competition is with C, C++ and Java
Despite its qualities, Ada has a lot of problem to be accepted in front of
those languages
Things are likely to be far worse in front of C# if it does what it says.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 676 bytes --]

Sorry if this was a mistake. I just remembered the message and thought it
was from you.
I cannot find it now, I must have read it about 2 or 3 weeks ago.
Didn't you (or somebody else on c.l.a) made a post about that ?
It said that most software users put the CD in.
Expect an autostart, a standard installation procedure, and standard menu
with expected functions.
And don't look any further if this is not the case.

"Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit dans
le message news: 9kbkll$m73$1@nh.pace.co.uk...
> Just for the record, this is not me. The cut you did with my name in it
must
> have come from an earlier message.
>
> MDC
> --

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1368 bytes --]

No, that *was* me. I was referring to the post as displayed after my sig. I
believe I was improperly quoted by Georg Bauhaus - the actual text, I
believe was yours.

I wasn't necessarily disagreeing - just that I didn't want to have remarks
attributed to me that were someone else's. One of those easily committed
errors in newsgroup postings :-)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"nicolas" <n.brunot@cadwin.com> wrote in message
news:8uda7.3152$6n3.4285485@nnrp6.proxad.net...
> Sorry if this was a mistake. I just remembered the message and thought it
> was from you.
> I cannot find it now, I must have read it about 2 or 3 weeks ago.
> Didn't you (or somebody else on c.l.a) made a post about that ?
> It said that most software users put the CD in.
> Expect an autostart, a standard installation procedure, and standard menu
> with expected functions.
> And don't look any further if this is not the case.
>
> "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit dans
> le message news: 9kbkll$m73$1@nh.pace.co.uk...
> > Just for the record, this is not me. The cut you did with my name in it
> must
> > have come from an earlier message.
> >
> > MDC
> > --
>
>
>

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <marin.condic.auntie.spam@pacemicro.com> wrote:
: Just for the record, this is not me. The cut you did with my name in it must
: have come from an earlier message.

That was a delete one line off in an ansi Windows telnet terminal,
on which nicolas had quoted you and which I've missed ;-) I'm Sorry.

-- Georg

Re: Proving Correctness (was Java Portability)

[Wes Groleau]

> That was a delete one line off in an ansi Windows telnet terminal,
> on which nicolas had quoted you and which I've missed ;-) I'm Sorry.

Oh, a one-off error.  Must be a C programmer.  :-) :-) :-)

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
: news: 9kb3ub$hdo$1@a1-hrz.uni-duisburg.de...
:> You write about debutants, could you provide some reasoning why for
:> these people there should be a more full grown environment than AdaGIDE?
: 
: I don't write only about debutants. Exactly the reasoning apply to
: experimented programmers.
: What I meant is that, in those languages, even debutants can find very
: easily simple tools,
: that experimented programmers are (or at least should be) very happy to
: find.
: 
:> Is the Aonix product so far off for this group of people???
: 
: Aonix is smart enough to provide something looking like Visual C++, Apex is
: not bad too.

So we have settled this?
(Like: every young man in the 90ies has had to have a Zippo lighter.
every debutant programmer has to have a Visual C++ like IDE. :-)

: Remember what Marin said about softwares
: You should put the CD, see a standard install procedure pop up, and find
: menus "file", "edit", "tools" in their expected palce, with expected
: functions and behavior.

Yes, I've read his posting. True, if you are or have to be impatient
because of deadlines say, you want a DWIM IDE with much reusable
software and a slot in
it that reads: "Put your needed libraries in here, I'll care
for the rest". Or better, they are already there.
This is important for people who cannot afford reading
documentation accompanying software components, and install after.
They won't buy otherwise.  So the Microsoft delusion
technique has caught on. 

I agree that to convince some people it must look like you are
doing profitable work right after the install wizard has 
finished it's work. But I doubt this will lead to effective
use of an IDE. Wich is not relevant for some marketing, true.

I think we could arrive at a much better explaination looking
at how mainstream programming and programming as a social
process relate. Counting PC installations, I think it is safe
to say that Ada is not popular, as are quite some other nice
languages.
<wildly fantasizing, or modelling you can call that>
Programmers or managers
in charge of deciding which tools to use can be viewed as
members of a pack. Leaders of each pack communicate via
known respectable channels (private or public) to agree
on the best solution wihch will help getting the job done. 
Very simple known and proven social effect:
The solution which the most people say is good, is good.
The leaders decide.
(For one of many references, see aany group study from
the mid of the last century. See also the book by T.C. Schelling,
wich has been mentioned a few times here. ) </>

CONSEQUENCE:
============
see below

: When standard and expected things are found, only after that, check the docs
: to go further,

Yes, the Microsoft delusion technique is successful, and
one might try to do the same to make Ada more successful,
if necessary. See below.
But to me, the evidence is still not in the 99% range.
Could you provide some evidence?

: If not just throw the software away. There is no excuse.

I've seen some unsuccessful undertakings using this approach.
So by current methodological standards, this is not a law.
Still, see below.

: He is right, and almost every software users do the same.

Almost every... May I again ask for evidence?

: May be the problem of Ada language is that such an evidence is not an
: evidence for too many Ada users.

Does "evidence" mean the same in French as it does in German, namely
"self-evidence"? If so,  might I ask again for evidence?

: The important point is that it is claimed by 99% of people who could think
: in using Ada

I must say, this number strikes me again and again.
You must have some data that I don't have. :-)
Remember, not every programmer in the world, and in business,
was brought up on Borland IDEs (was there a standard Pascal
library ? :-)

: If your product doesn't satisfy them, change the product or they'll forget
: about your poduct, and they will be very wise to do so.

Well, certainly Rational Rose os not a click and go product,
very unsatisfying if you are used to "install, be productive
from the start". Still some do not throw it away. I think
you might have to be a bit more specific about "products".

: They know what they want, if you do not provide it, and if others provide
: it, no hesitation ...

Yes, but in some cases they come back, repentant.
Any parents reading this? :-)

: Every [!] industry has understood such elementary evidence [!],
software industry
: started to understand it[!], those who don't understand will disappear[!]...

Have you considered working as a stockbroker?
You could advise potential monopolists' shareholders :-)
(Sorry.)


: Just have a look at a Java compiler. It is self-explanatory ...

That's what I'm doing a few days every week, and from
both Emacs and an alledgedly _very_ popular Java IDE by IBM which
does not fit your criteria for the entries in the menues,
for example. Plus the class browser, source views, etc which
remind me of Smalltalk, not Windows.

It prooves a few things:
1 The libraries are not really for a ready to go approach,
  even excluding the sort
  of libraries I have named, which are not considered "standard"
  (like linear algebra packages etc). Example: for java.sql.*
  to work, remotely, you need 3rd party drivers, outside the IDE.
  Much the same with GNADE, from what I've read.
  After all, the Son docs invite you to either use or extend the
  library classes. Or to implemement the interfaces...
  So it is disguise to some extent, but apparently
  effective as far as appreciation in the windows world
  is concerned.
2 The new interfaces in Java 2 seem to have been influenced
  by the GoF patterns. Implmentations also available from adapower,
  due to wonderful work by Matthew Heany and friends, if I may
  mention this.
3 Are there graph classes and related algorithms somwhere
  in the java.* hierarchy? Have I missed them? Aren't they
  pretty standard? Present in the Booch components.
  
So it boils down to the question: where is the student's
(or private use) price CD that integrates them for the
windows user using a 1-click installation wizard?
Where is Visual Ada?

If Nicolas is right, there is a chance to make money.
Just package the available software into an IDE, including
available libraries.  Need not be a powerful IDE since the CD
is used for convincing people who are used to Windows GUI
standards IDEs.

(Besides, where do so many people get their nice full blown
MS Visual Xyz licenses from? Are they all students?)

: Once again compare a Java compiler and any Ada compiler.
: while we hear on c.l.a. endless discussions about syntax, and that 18 year
: after Ada83 and 6 years after Ada95 ?
: Don't you think there could be some lesson to learn ?

Look at the Pizza, Aspect J, ..., and GJ discussions. 
 
: Everybody knows what popular languages compilers provide
: Everybody knows what Ada compilers provide
: Everybody can compare how widely they are used.

I seriously doubt that any of these 3 sentences is based on facts,
and is well defined. 

Let me boldly try to translate:
"I, like some of my collegues, have grown accustomed to some
 popular language programming environment. I like what is there
 after I do the usual install. This is more compicated and tedious
 in Ada environments. There seem to be more installations of that
 pop lang environment on PCs and Macs."

" So the obvious conclusion is: This correlation is a cause-effect
  relation!"

I beg to differ. See below.
 
: I cannot understand why so much time is lost in useless discussion while the
: gap is growing.

What gap? Starting when? Which significance?
This sounds like priest's words wanting to frighten someone.
Misleading rhetoric. One-factor explainations.
(VB 2001/ VB 2001) > (Ada 2001/Ada 2000) implies dangerous gap?
Fact, facts, please. They could be really helpful.

--  Georg Bauhaus
---
GTK bound--how to defeat GNOME

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1560 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9kchn1$lng$1@a1-hrz.uni-duisburg.de...
> : He is right, and almost every software users do the same.
>
> Almost every... May I again ask for evidence?

Have you ever met people considering that computers are just tools ?
I guess the misunderstanding comes from the fact that
- Most programmers like computers, this is a hobby, they have a lot of fun
playing with them
- the rest of the world use a computer or a software like a TV set, a coffee
machine or whatever you want.
the important point is that it must work more or less, be easy to use,
nobody cares what's inside ...
Nobody cares if it comes from the evil Microsoft or the good whatever you
want ...
Nobody wants to spend 5 hours reading a doc before a basic use of the
product

You'd never accept to have to read the doc of a new car or motorbike before
you can find where are the brakes ...
Would you ?

The day programmers will understand how the rest of the world thinks, this
will be a giant step forward
Now, about all what we were talking about, I guess softwares would much
better meet users requirements if programmers had not so much fun playing
with computers.

Popular languages are used by a lot of people, more and more programming
just for work, and having fun with everything but computers outside the
office.
So they are evolving to be useful tools, because those guys don't accept
weird things you can only appreciate if you have fun with computers.
If Ada doesn't go that way, we have to worry.

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
: news: 9kchn1$lng$1@a1-hrz.uni-duisburg.de...

:> Almost every... May I again ask for evidence?

None has arrived, afaics.
 
: Have you ever met people considering that computers are just tools ?

Yes.

: the important point is that it must work more or less, be easy to use,
: nobody cares what's inside ...

you claim.

: Nobody cares if it comes from the evil Microsoft or the good whatever you
: want ...

evidence? Besides, I'm not saying, M$ is evil, whatever the truth
may be.

: Nobody wants to spend 5 hours reading a doc before a basic use of the
: product

you claim.  Book publishers might not agree. I could, as far as
Word processor users are concerned, but
hey, we talked about programmers considering Ada and chosing
an IDE and libraries, not coffe computer users you talk about now.

: You'd never accept to have to read the doc of a new car or motorbike before
: you can find where are the brakes ...
: Would you ?

_Before_ I by a car (oh, hopefully that will happen once more
some day :-) I try to find out what it is like. I think brakes
are at a much too basic level for an appropriate analogy.

: The day programmers will understand how the rest of the world thinks, this
: will be a giant step forward

I think the M$ marketing department has understood,
but the brave of course make every pedagogical effort
at saving the general public from believing that,
computing = clicking Windows. Life is not just business.
Please consider what is happening in schools: Word,
*not* text processing, is considered a basic skill to be taught
to children. Why? Because they think they are fulfilling
user (= labour market) requirements. Perceived user
requirements, listening to locally elected expert teachers.
That's shortsighted!


: Now, about all what we were talking about, I guess softwares would much
: better meet users requirements if programmers had not so much fun playing
: with computers.

Could you elaborate? Fun: I know of at least a few prominent
computer scientist who wrote a note _in favour of reintroducing
or keeping_ fun in programming (Knuth, Perlis, ...). There must
be some real world background to that...

Meeting your flavour of Windows user requirements:
And should we surrender to what you naively call "user
requirements"?  It is naive because user requirements, as you
have defined them in your recent postings, are not
necessarily free inventions of users but are of course largely
influenced by the pressure of large groups, peer groups,
the crowd, fashion, and other things I've mentioned during this
discussion. How many times have _you_ heard the sentence,
"I know you are right, but give me what everybody else has."?

Now back to Ada: From what we know about mass movements
in the small or in the large, we can conclude that they may
have desasterous effects. So it is a question of social
responsibility to _not_ surrender to your "user requirements"
if one can figure a way to avoid this. For example if in my software
it is essential to avoid one-off errors, then a language which
offers built-in support for catching them is a big win.
That comes first in times. If it helps to build a CD with
prebuild libraries and a basic IDE made from available
sources to foster the use of Ada by way of exploiting
the Visual Xyz delusion, then there is a chance to not only
make money, but also to help the world suffering less from
the consequences of overflow in programs written in a 
language where this may be considered endemic.

: Popular languages are used by a lot of people, more and more programming
: just for work, and having fun with everything but computers outside the
: office.
: So they are evolving to be useful tools, because those guys don't accept
: weird things you can only appreciate if you have fun with computers.

Hm, are you saying that only progrmmers having fun with their
computers after work accept to expose themselves to weird languages
like Ada? 

: If Ada doesn't go that way, we have to worry.
 

Well, I don't. Do as you like. Maybe you can start a
profitable project of building a very convincing debutant's
Ada CD? :-)

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

There is an argument that goes something like this: It doesn't particularly
matter *why* people are choosing development environments that have all
sorts of integrated tools and it doesn't particularly matter if this is a
*good* thing or not. The fact remains that the market is out there buying
things like MSVC++, etc. If Ada wants to gain popularity in those markets,
it has to come to the table with at least what customers are currently
buying. (id est, some kind of IDE that is integrated with source level
debugging, a GUI builder, class libraries, etc.)

Now that may not be the smart decision for buyers to make and maybe one can
suggest to the buyer that it isn't smart and offer some alternative answer.
But remember that you do need an alternative answer - I doubt it will fly to
say "Go assemble the pieces you want from the Internet and cobble it
together for yourself." (At least not in any sort of mass manner. It hasn't
so far, has it?)

So, IMHO, Ada needs to be more "Customer Centered" and offer the users &
potential users the kinds of things they are currently buying. There is a
huge cost to an organization to switch from technology they are currently
using to something new, so they need to see that they are getting some sort
of productivity advantage in doing so. They want to see that they have all
the tools they get with their current technology & then some. Lots of
developers are being driven by time-to-market and minimal cost solutions
because of highly competitive situations. Expecting a large investment in
researching the best possible tools, assembling the pieces from a variety of
places, gluing them together in some manner and training people to use the
cobbled-together pieces is not very attractive when a competitor can come
along and say "See? I've got it all in one box and there's a good collection
of documents that work your people through how to use it and you're up and
running right away."

Or we can say "What you want doesn't make sense and when you smarten up,
you'll choose Ada" - but that hasn't worked well so far in terms of
capturing a large market.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:9kchn1$lng$1@a1-hrz.uni-duisburg.de...
>
> Yes, I've read his posting. True, if you are or have to be impatient
> because of deadlines say, you want a DWIM IDE with much reusable
> software and a slot in
> it that reads: "Put your needed libraries in here, I'll care
> for the rest". Or better, they are already there.
> This is important for people who cannot afford reading
> documentation accompanying software components, and install after.
> They won't buy otherwise.  So the Microsoft delusion
> technique has caught on.
>
> I agree that to convince some people it must look like you are
> doing profitable work right after the install wizard has
> finished it's work. But I doubt this will lead to effective
> use of an IDE. Wich is not relevant for some marketing, true.
>

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

I like the way you think. :-)

There was a time when I was interested in being a "geek" and I got a great
deal of satisfaction out of understanding all the intracacies of an OS and
how to plug software into it and so on. That time is now past. I want to
turn it on and get a job done. The less I need to know about the appliance,
the more time I have free to dedicate to more profitable endeavors. Maybe
I'm atypical of the population of programmers out there, but I think that it
isn't hard to imagine most people appreciating a software product that just
plugs in and starts to work. (Do most people *really* want to spend time
fooling with registry entries, paths, file locations, compile/install
scripts, etc? Or would they rather just plug the disk in, say "Install" and
get started playing with the product? It leaves more time for sitting on the
beach, sucking down a Pina Colada and getting a suntan. :-)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"nicolas" <n.brunot@cadwin.com> wrote in message
news:R1ta7.4077$ke.4769807@nnrp6.proxad.net...
>
> Have you ever met people considering that computers are just tools ?
> I guess the misunderstanding comes from the fact that
> - Most programmers like computers, this is a hobby, they have a lot of fun
> playing with them
> - the rest of the world use a computer or a software like a TV set, a
coffee
> machine or whatever you want.
> the important point is that it must work more or less, be easy to use,
> nobody cares what's inside ...
> Nobody cares if it comes from the evil Microsoft or the good whatever you
> want ...
> Nobody wants to spend 5 hours reading a doc before a basic use of the
> product
>

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1413 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9ke8b4$qiv$1@a1-hrz.uni-duisburg.de...

> _Before_ I by a car (oh, hopefully that will happen once more
> some day :-) I try to find out what it is like. I think brakes
> are at a much too basic level for an appropriate analogy.

This is a quite appropriate analogy according to me.
Millions of people use cars and motorbikes, ask them
You can rent one, a lot of people do that
This is much cheaper, and you can change each week-end
(If you don't spend the entire week-end reading the doc ...)

> :> Almost every... May I again ask for evidence?
> None has arrived, afaics.
> : Have you ever met people considering that computers are just tools ?
> Yes.
> : the important point is that it must work more or less, be easy to use,
> : nobody cares what's inside ...
> you claim.
> : Nobody cares if it comes from the evil Microsoft or the good whatever
you
> : want ...
> evidence? Besides, I'm not saying, M$ is evil, whatever the truth
> may be.

Well I think we are getting a little too childish.
You don't see any evidence of what I claim,
I don't see any evidence of what you claim
and I guess that sometime common sense should be enough.
But never mind, that's not very important,
the day software world is convinced to use Ada with current mentality,
I'll be very happy to aknowledge you were right and I was wrong ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 763 bytes --]

"Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit dans
le message news: 9ke9pa$lks$1@nh.pace.co.uk...
> So, IMHO, Ada needs to be more "Customer Centered" and offer the users &
> potential users the kinds of things they are currently buying.

I agree. No product, how excellent it is, can afford to forget this
elementary rule.
And Ada how excellent it is, is everything but "Customer Centered"

> Or we can say "What you want doesn't make sense and when you smarten up,
> you'll choose Ada" - but that hasn't worked well so far in terms of
> capturing a large market.

Right.
I saw Ada being promoted that way for about 12 years
Everybody can see the result today.
And I had don't see very well why it would work better in the future.

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <marin.condic.auntie.spam@pacemicro.com> wrote:
 
:  (Do most people *really* want to spend time
: fooling with registry entries, paths, file locations, compile/install
: scripts, etc? Or would they rather just plug the disk in, say "Install" and
: get started playing with the product? It leaves more time for sitting on the
: beach, sucking down a Pina Colada and getting a suntan. :-)

I guess most people will agree that most people are or become
sane enough to avoid system configuration as much as possible.
(Having come back home from Bretagne, and missing the delicious
combination of sun, wind, and the see, and more, I do certainly
agree. :-) 

Only, this cannot be used to claim that there are no libraries for
Ada, or there is no equivalent to SUN class libraries, as Nicolas
has done (as far as I have understood his earlier postings.)
It's just not as convenient to use them as the use of a select set
of libraries coming with some Visual compilers.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1359 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9keduf$qvc$1@a1-hrz.uni-duisburg.de...
> Marin David Condic <marin.condic.auntie.spam@pacemicro.com> wrote:
> Only, this cannot be used to claim that there are no libraries for
> Ada,
I don't remember having said that ...
I heard about some Ada libraries
I sometime check Adapower :-)

>or there is no equivalent to SUN class libraries, as Nicolas
> has done (as far as I have understood his earlier postings.)

I don't know yet a good enough XML library (with DTD and so on)
we tried some which all had far too many limitations,
and I know at least one project where it was decided to develop a home-made
limited one for specific use.

I'm not sure you can easily find exhaustive equivalent of :

java.util.zip
java.util.jar
java.net
java.rmi
java.sql
javax.sound
javax.transaction
javax.swing
org.omg.corba

... etc ...

without spending quite a lot of time searching or writing yourself too many
things,
while it just took me a few seconds to know their existence and availability
with the HTML doc of a java compiler
By the way, the automatic generation of docs in Java is really great ...

> It's just not as convenient to use them as the use of a select set
> of libraries coming with some Visual compilers.
>
not as convenient ... , that's the least you can say !!!

Re: Proving Correctness (was Java Portability)

[Larry Kilgallen]

In article <9kea9a$lsc$1@nh.pace.co.uk>, "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> writes:
> I like the way you think. :-)
> 
> There was a time when I was interested in being a "geek" and I got a great
> deal of satisfaction out of understanding all the intracacies of an OS and
> how to plug software into it and so on. That time is now past. I want to
> turn it on and get a job done. The less I need to know about the appliance,
> the more time I have free to dedicate to more profitable endeavors. Maybe
> I'm atypical of the population of programmers out there, but I think that it
> isn't hard to imagine most people appreciating a software product that just
> plugs in and starts to work. (Do most people *really* want to spend time
> fooling with registry entries, paths, file locations, compile/install
> scripts, etc? Or would they rather just plug the disk in, say "Install" and
> get started playing with the product?

I call those people "Macintosh users" :-)

Last night we went to a Kinko's that told me on the phone they had
scanners on Macintosh, and they didn't.  I stormed out.  My wife
said that if push came to shove she would be willing to do it on
a PC (which that Kinko's did offer).

We went to another Kinko's that _did_ have scanners on Macintosh
(2, in fact) and had a rather easy time (aside from waiting for
the scanner to scan) pulling something into Photoshop and putting
it on diskette in JPEG for a web site my wife is setting up.  This
is without having ever used Photoshop or a G4 Macintosh (the Flower
key is displaced a bit from what I use).

This experience is _considerably_ better than I have had with
copying machines.

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Well, I think that this is possible on just about any platform - it just
means that the software vendor needs to do the work to make it as seamless
as is possible. I've had apps install nicely on WinNT and pop up and run
just fine and present me with an interface I could recognize & work with
right from the start. I've also had apps that did the same on VMS with a
command line interface. Its just a matter of taking the time to make it work
properly and use whatever interface is customary on the platform of
interest.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message
news:pLczjM8J5xm3@eisner.encompasserve.org...
>
> I call those people "Macintosh users" :-)
>

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <marin.condic.auntie.spam@pacemicro.com> wrote:

: So, IMHO, Ada needs to be more "Customer Centered" and offer the users &
: potential users the kinds of things they are currently buying. 

Or, only as an alternative, something completely different.
That variant is quite a market success in times.
E.g., I wasn't tooooo impressed when I tried ISE Eiffel (only the
text editing facilities), but it does really have more power than vi
and adamode is heading towards it.  They now have completed
Visual Studio integration, but I doubt they have given up all
the nifty browsing features, which afaik couldn't be bought for
C++, say. Let's see how Eiffel# will raise the number
of Eiffel installations.


As to programmer capabilities, aka human resources(?), and given
the number of circulating Visual C++ copies, I realize there is
a market for Visual Ada, but <Conjecture>there is also quite a
number of programmers who have learned to use a "Language/Emacs"
combination, or better, or worse</>, but different enough to
make the effort to learn to use Visual Ada less cost effective
than assembling the components in the manner they are somewhat
used to.

:  I've got it all in one box and there's a good collection
: of documents that work your people through how to use it and you're up and
: running right away."

Does this product exist? :-) We've not been _that_ successful with
the alledgedly multipotent Java libraries, in that a few tedious
work arounds were necessary.

: Or we can say "What you want doesn't make sense and when you smarten up,
: you'll choose Ada" - but that hasn't worked well so far in terms of
: capturing a large market.

Agreed, yes, the masses are best captured by positive mass psychological
means, which sometimes involves being silent about bitter truths. :-) :-)

 But have you considered what mass market Ada programs will look
like? :-) :-) :-)

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
: "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
:> Only, this cannot be used to claim that there are no libraries for
:> Ada,
: I don't remember having said that ...

Euh, yes , sorry, only I've had difficulties to see that your many
'Nobody does...'
'Everybody knows...'
'99% says...'
"Where are the library components, reuse etc ... ?
which I started to interpret as rhetorically chosen were meant
literally, in part. Uhm,... O.K. not every comes with your compiler,
when they exist.

: I don't know yet a good enough XML library (with DTD and so on)

Is this one insufficient?:
"It includes support for parsing XML files, including DTDs,
as well as a full support for SAX, and an almost complete support
for the core part of the DOM"  http://libre.act-europe.fr/xmlada/


: I'm not sure you can easily find exhaustive equivalent of :

Neither am I.  For some of them, I didn't even have to look,
because of language features. ;) One of them does not apply in
Ada context.  Most of them probably don't come with compilers,
yes.  CORBA has a directory on adapower. An unzip library can
be found at http://www.mysunrise.ch/users/gdm/unzipada.htm.
The GNADE project appears to make some progress.  In some
spots, this list isn't exhaustive, but mere existence of a Java
package doesn't solve every problem, sad as it is.  I am not
saying this to demean these useful libraries.  E.g., not every
popular image/sound format is handled by java.awt.* or javax.
Same library tediousness as with Ada. But it sure makes one feel
comfortable to have HashMaps, Lists, and Sets.

(I think I#ll compare java.sql.*/Tomcat to GNADE/AWS and
see if they match :-)

Re: Proving Correctness (was Java Portability)

[Warren W. Gay VE3WWG]

Marin David Condic wrote:
> I like the way you think. :-)
> 
> There was a time when I was interested in being a "geek" and I got a great
> deal of satisfaction out of understanding all the intracacies of an OS and
> how to plug software into it and so on. That time is now past. I want to
> turn it on and get a job done. The less I need to know about the appliance,
> the more time I have free to dedicate to more profitable endeavors. Maybe
> I'm atypical of the population of programmers out there, but I think that it
> isn't hard to imagine most people appreciating a software product that just
> plugs in and starts to work. 

Here here! I hear ya.

> (Do most people *really* want to spend time
> fooling with registry entries, paths, file locations, compile/install
> scripts, etc? Or would they rather just plug the disk in, say "Install" and
> get started playing with the product? It leaves more time for sitting on the
> beach, sucking down a Pina Colada and getting a suntan. :-)

When you're young you have enthusiasm and the energy to learn all of those
"intracacies". As you get "beyond young" you loose enthusiasm for the detail,
and lack the energy to care.. yes, you begin to focus on getting the job
done. ;-)

I like Ada for many reasons, but one of them is to avoid having to deal with
the tireless details! Especially for tasking (threaded) code.

> MDC
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/

-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Proving Correctness (was Java Portability)

[AG]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 848 bytes --]


"nicolas" <n.brunot@cadwin.com> wrote in message
news:bmya7.18382$Iu6.4283510@nnrp1.proxad.net...
> "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit dans
> le message news: 9ke9pa$lks$1@nh.pace.co.uk...
> > So, IMHO, Ada needs to be more "Customer Centered" and offer the users &
> > potential users the kinds of things they are currently buying.

Well, if you only offer what they are currently buying you aren't going to
pull them
away from what they are currently buying:) I thought the whole point was to
offer
something better.

>
> I agree. No product, how excellent it is, can afford to forget this
> elementary rule

If that was true we would still be using stone hammers to hunt wooly
mammothes
(or whatever the plural is for mammoth). After all - that was exactly what
the
customers were buying at the time.

Re: Proving Correctness (was Java Portability)

[rob]

In article <9k6aug$mtq$1@nh.pace.co.uk>, "Marin says...
>
 
>I agree that it doesn't really help anything to avoid
>discussing the issues that may be keeping Ada from gaining acceptance.
 
Ada will become popular when Borland makes a Turbo Ada :)

May be ACT should talk to Borland about this. GNAT will be the engine, and
Borland will write the IDE and the rest of the env. to package it. I think
a Turbo Ada would be a killer app! (make sure it has a button to click on
to generate byte code from Ada source also).

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

rob@nospam <rob_member@newsguy.com> writes:

> In article <9k6aug$mtq$1@nh.pace.co.uk>, "Marin says...
> >
>  
> >I agree that it doesn't really help anything to avoid
> >discussing the issues that may be keeping Ada from gaining acceptance.
>  
> Ada will become popular when Borland makes a Turbo Ada :)

Amazing, I have thought the same thing for long time... since 1985-86 when I
have used for the first time Turbo Pascal (what version 3.x if my memory is
right). But now we have GNAT an affordable Ada compiler and very stable... so
I'm not sure I still share your idea today... Note that for the IDE Aonix have
done a nice job. Of course if Delphi was based on Ada... But that's another
story :)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1394 bytes --]

"AG" <ang@xtra.co.nz> a �crit dans le message news:
LH_a7.739$fg7.176384@news.xtra.co.nz...
>
> "nicolas" <n.brunot@cadwin.com> wrote in message
> news:bmya7.18382$Iu6.4283510@nnrp1.proxad.net...
> > "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit
dans
> > le message news: 9ke9pa$lks$1@nh.pace.co.uk...
> > > So, IMHO, Ada needs to be more "Customer Centered" and offer the users
&
> > > potential users the kinds of things they are currently buying.
>
> Well, if you only offer what they are currently buying you aren't going to
> pull them
> away from what they are currently buying:) I thought the whole point was
to
> offer
> something better.

And better usually means at least as good, which doesn't prevent to you from
offering something better.
We agree ...
I wouldn't see any reason to use Ada if it was only as good as other
languages.

> > I agree. No product, how excellent it is, can afford to forget this
> > elementary rule
>
> If that was true we would still be using stone hammers to hunt wooly
> mammothes
> (or whatever the plural is for mammoth). After all - that was exactly what
> the
> customers were buying at the time.
>

May be the danger of mammoth hunting ... or their extinction ...
Great, we just have to wait for the extinction of other languages to see Ada
qualities widely recognized.
I'm afraid I'll be retired at that time ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3603 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9kelv1$riq$1@a1-hrz.uni-duisburg.de...
> nicolas <n.brunot@cadwin.com> wrote:
> : "Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a ?crit dans le message
> :> Only, this cannot be used to claim that there are no libraries for
> :> Ada,
> : I don't remember having said that ...
>
> Euh, yes , sorry, only I've had difficulties to see that your many
> 'Nobody does...'
> 'Everybody knows...'
> '99% says...'
> "Where are the library components, reuse etc ... ?
> which I started to interpret as rhetorically chosen were meant
> literally, in part. Uhm,... O.K. not every comes with your compiler,
> when they exist.

I guess we have a very different view of what surrounds us.
By the way 99% is in my opinion very optimistic ...
I doubt 1% of programmers in the world have heard about Ada ...
I'm not sure you very often provide software to South America, Japan, China,
Korea etc ... to have such an opinion.

>
> : I don't know yet a good enough XML library (with DTD and so on)
>
> Is this one insufficient?:

No, and this is so obvious that it's very bad news somebody can think it is
sufficient.
Don't get me wrong, it's great to have this library for students or Ada
fans.
But it is definitely not ok for a professional  use.
You may say that we just have to be Gnat supported customers
You just forget that not every Ada project, even professional ones, use
Gnat.

> "It includes support for parsing XML files, including DTDs,
> as well as a full support for SAX, and an almost complete support
> for the core part of the DOM"  http://libre.act-europe.fr/xmlada/

Sax relies on Gnat specific packages.
I even remember a post a few monthes ago saying that GNAT.Spitbol use Gnat
specific pragmas

The library was developped and tested with Gnat 3.14a
tested on
GNU linux x86
Sparc Solaris

Which percentage of Ada users are Gnat supported users ?

I would expect an Ada library to be tested at least against

Gnat public version
Aonix Objectada
Rational Apex

and obviously on

Windows ...
 (You may have heard about it, a little operating system that a very few
number of stupid people are still using ...)

It's incredible not to understand how such elementary things are obvious ...
And don't tell me that may be this works, we just have to try or may be to
change 1 or 2 little things ...
I won't do it
If this is not easy, I don't have to spend my time on this
If this is easy, a sensible Software manager expects that the library
provider does the test

> : I'm not sure you can easily find exhaustive equivalent of :
>
> Neither am I.  For some of them, I didn't even have to look,
> because of language features. ;) One of them does not apply in
> Ada context.  Most of them probably don't come with compilers,
> yes.  CORBA has a directory on adapower. An unzip library can
> be found at http://www.mysunrise.ch/users/gdm/unzipada.htm.
> The GNADE project appears to make some progress.  In some
> spots, this list isn't exhaustive, but mere existence of a Java
> package doesn't solve every problem, sad as it is.  I am not
> saying this to demean these useful libraries.  E.g., not every
> popular image/sound format is handled by java.awt.* or javax.
> Same library tediousness as with Ada. But it sure makes one feel
> comfortable to have HashMaps, Lists, and Sets.
>
> (I think I#ll compare java.sql.*/Tomcat to GNADE/AWS and
> see if they match :-)

I don't think you understood what is the problem.
Don't spend your time trying to convince me, you have the whole rest of the
world to convince ...

Re: Proving Correctness (was Java Portability)

[Mike]

In article <dVsb7.5035$2i4.5346431@nnrp6.proxad.net>, "nicolas" says...
>
 
>I would expect an Ada library to be tested at least against
>
>Gnat public version
>Aonix Objectada
>Rational Apex
>
>and obviously on
>
>Windows ...
 
The Ada standard library development model is not up to the modern times.

Look at how Sun manages Java. They have reqular releases of an updated 
Java (mostly in the standard libraries). This is how Ada should work. Ada
standard libraries should be updated with more packages all the time. 

Without a more powerfull and rich Ada standard library, Ada will not be used
as much as it could. If the XML library was part of the Ada standard, then
all compilers will have be shipped with it and support it. (Java JDK 1.4
for example, added more Java XML standard packages).

The problem with Ada is there there is no one big company or an
organization behind it. lets see:

Java/IBM/Borland --> Java
MS --> C++, VB, C#
Borland --> Delphi, Kylix
??? --> Ada

Defence used to be the big one "taking care" of Ada, and it dropped it. So, now
we need someone else. 

Ada standard libraries must be enhanced and extended for Ada to grow. 

just my 2 cents.....

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1239 bytes --]

"Mike@xx" <Mike_member@newsguy.com> a �crit dans le message news:
9klokd0nif@drn.newsguy.com...
> The Ada standard library development model is not up to the modern times.
>
> Look at how Sun manages Java. They have reqular releases of an updated
> Java (mostly in the standard libraries). This is how Ada should work. Ada
> standard libraries should be updated with more packages all the time.
>
Thanks a lot ... I'd like to hear that kind of things more often
I don't understand how Ada could stay up-to-date with one revision of
standard libraries in 18 years ...
And who knows when will be the next, if there is one ...
No need to change language rules and syntax, but at least, as far as
libraries are concerned, you are absolutely right.
Ada users should be able to count on a complete set of libraries, absolutely
compatible with main available compilers and main platforms, gathered on one
official download site.
If everybody releases his own version of the same library for his favourite
compiler, on his favourite platform, we'll get nothing really useful, and
moreover, we'll prove to others languages users, that Ada users give lessons
to everybody about software development, and are unable to apply them in
practice.

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> > : I don't know yet a good enough XML library (with DTD and so on)
> >
> > Is this one insufficient?:
> 
> No, and this is so obvious that it's very bad news somebody can think it is
> sufficient.
> Don't get me wrong, it's great to have this library for students or Ada
> fans.
> But it is definitely not ok for a professional  use.

Why ?

> You may say that we just have to be Gnat supported customers

No. Just grab the library and changes it to be whatever-compiler
compatible. You have the source, use them...

> You just forget that not every Ada project, even professional ones, use
> Gnat.

Right. So ask you vendor to adapt an Open Source component or sell an
equivalent component.

I don't see a problem with Aonix, Rational taking the XML library, adapt it to
their compiler and release the result as an Open Source add-on for their
compiler.

> > "It includes support for parsing XML files, including DTDs,
> > as well as a full support for SAX, and an almost complete support
> > for the core part of the DOM"  http://libre.act-europe.fr/xmlada/
> 
> Sax relies on Gnat specific packages.

So why ? (see above). XML has been done by an Ada fan working for ACT. That's
not a big surprise. I don't understand why you are so upset ?

And BTW, it is no fair to be upset by something that has been done. Why are
you not upset by the fact that these components have not been ported by you
Ada vendor for their compilers ?

> I even remember a post a few monthes ago saying that GNAT.Spitbol use Gnat
> specific pragmas
> 
> The library was developped and tested with Gnat 3.14a
> tested on
> GNU linux x86
> Sparc Solaris
> 
> Which percentage of Ada users are Gnat supported users ?
> 
> I would expect an Ada library to be tested at least against
> 
> Gnat public version
> Aonix Objectada
> Rational Apex
> 

Possible, but not something that should be done by ACT for obvious reason :)

Let me add that we have tried in AWS to be compiler independant (see
AWS.OS_Lib) but since 2 or 3 releases this is not true anymore. I don't have
the time to try on all available compilers, the GNAT library is so nice that I
just can't bypass it (GNAT.Regexp for example). So yes now AWS is mostly
working with GNAT... But I'd love to see others porting some stuff to be able
to avoid this dependences :)

> and obviously on
> 
> Windows ...
>  (You may have heard about it, a little operating system that a very few
> number of stupid people are still using ...)
> 
> It's incredible not to understand how such elementary things are obvious ...
> And don't tell me that may be this works, we just have to try or may be to
> change 1 or 2 little things ...
> I won't do it
> If this is not easy, I don't have to spend my time on this

Others have spent THEIR time on XML, GtkAda :), mostly because they are Ada
fans !

Sorry but I'm very upset by your mail. If nobody does the job the job wont
gets done that it !

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> "Mike@xx" <Mike_member@newsguy.com> a �crit dans le message news:
> 9klokd0nif@drn.newsguy.com...
> > The Ada standard library development model is not up to the modern times.
> >
> > Look at how Sun manages Java. They have reqular releases of an updated
> > Java (mostly in the standard libraries). This is how Ada should work. Ada
> > standard libraries should be updated with more packages all the time.

I've never seen libraries updating itself right :) ? So who should do 
the job ? This is the real question.

> >
> Thanks a lot ... I'd like to hear that kind of things more often
> I don't understand how Ada could stay up-to-date with one revision of
> standard libraries in 18 years ...

This is too much fun guys !!!

As you said SUN add Java libraries. But ACT add GNAT libraries too right ?

The big difference is that "Java == SUN" but Ada is a standard and of course
ACT is not driving the changes, nor Aonix, nor Rational...

I agree that Ada is changing slowly compared to Java, but it stays under
control and well designed. I don't think you can play on both sides.

Now, you should submit library additions to the ARG to have a chance to see
them added in next revision.

Maybe all Ada vendors should put some money on some sort of organization to
have the Ada libraries updated using standard Ada without vendor-specific
additions. This would benefit the whole Ada community. Does such an
organization exist ?

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 5627 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
ulmkxuy80.fsf@wanadoo.fr...
> Sorry but I'm very upset by your mail. If nobody does the job the job wont
> gets done that it !
I'll start to reply your last remark.
There is absolutely no attack about this XML library. It's great to have it
for students and Gnat users.

But we were talking about standard and portable libraries designed to be
independant of compilers.
This is what Ada is designed for. This is about what Ada users are giving
lessons to everybody.
George gave me this XML library as an example to show that this type of
library is available.

This is not the case.
We don't use Gnat because Gnat doesn't satisfy our requirements. That's our
problem.
That doesn't mean Gnat is a bad product, no product is perfect, and no
product can satisfy everybody.
Gnat, especially public version, is great for students, for anybody who
wants to try Ada, and certainly for professionals who made Gnat choice.
A lot of people are using others compilers than Gnat, a lot of people work
with Windows.
You definitely cannot take this XML library example to show that standard
portable libraries are available for Ada users.

> No. Just grab the library and changes it to be whatever-compiler
> compatible. You have the source, use them...

No, I shouldn't have to do that,
Anyway, Gnat.spitbol is not available for other compilers, and I don't have
to spend my time retrieving it and trying to adapt it to my compiler.
It's ok for Ada fans ready to spend more time on their favourite language
than the time they would spend on a language they don't like.
This is not ok for people considering languages as tools, who choose the
most convenient one for their needs.
It's a nonsense to say to Ada programmers
"I made my own specific library, adapt it yourself if you don't want to use
my compiler"
This is exactly what people who choose Ada don't want to hear, and they are
right about that.

> Right. So ask you vendor to adapt an Open Source component or sell an
> equivalent component.

All the thread is about the fact that this is not an option and goes against
Ada promotion
Ada is designed to write code as much as possible independant from platforms
and compilers.
An XML library has nothing specific to a platform or a compiler.
You can develop one with code totally independant of the compiler, and there
is no reason to do otherwise.

You may ask why we didn't develop that ourseleves :  it is for very simple
reasons.
- We are not paid for that
- Java does it much better.
- When we are not working, we are more interested in others things in life
than playing with computers

> I don't see a problem with Aonix, Rational taking the XML library, adapt
it to
> their compiler and release the result as an Open Source add-on for their
> compiler.

It's clear to me that a standard library shouldn't have to be adapted to
each compiler.
We have more than 1 million lines of Ada code and are very careful to have
them compiler independant.
That's the least you can expect from an Ada library.
Ada language is designed to be platform and compiler independant.
If Ada users starts to release such basic libraries specific to one compiler
or one platform, the only thing we do is to prove to others languages users
that we are unable to apply what we are trying to promote.

> > Sax relies on Gnat specific packages.
> So why ? (see above). XML has been done by an Ada fan working for ACT.
That's
> not a big surprise. I don't understand why you are so upset ?

I am not upset, Georges told me I should be satisfied with this XML library.
I am not satisfied because we don't use Gnat for our software releases, and
because libraries tied to one compiler without any valid reason, go against
Ada promotion.
The only consequence is that I don't use it.
No big deal ...

> And BTW, it is no fair to be upset by something that has been done. Why
are
> you not upset by the fact that these components have not been ported by
you
> Ada vendor for their compilers ?

Once again I'm not upset about anything
I strongly believe that the lack of standard available library you can use
'as it is' without any modification, whatever compiler you use, goes against
Ada rationale ...

> Possible, but not something that should be done by ACT for obvious reason
:)

I don't care who should do it .... I'm a user and like any other user I
choose the best tool available for my needs.
If ACT does everything to force Ada world to be dependant of Gnat compiler,
it is ACT's decision.
But I think it could kill Ada, and therefore ACT ...

> Let me add that we have tried in AWS to be compiler independant (see
> AWS.OS_Lib) but since 2 or 3 releases this is not true anymore. I don't
have
> the time to try on all available compilers, the GNAT library is so nice
that I
> just can't bypass it (GNAT.Regexp for example). So yes now AWS is mostly
> working with GNAT... But I'd love to see others porting some stuff to be
able
> to avoid this dependences :)

I think it's a dead-end if everybody does his own job specific for his own
tools, and wait for others to adapt it for others tools.
This shows all the limitations of the process.
We are giving to others languages users, all what they need, to be convinced
that Ada users are themselves unable to stick to what they promote.
If you can't bypass GNAT.Regexp, obviously GNAT.Regexp should be in the set
of standard libraries, it shouldn't be named GNAT, and should be available
as it is for any compiler.
Once again Ada users give strong lessons to the rest of the world, and do
exactly the opposite in practice ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1028 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
uhevluxrf.fsf@wanadoo.fr...
> This is too much fun guys !!!

At least if it makes you laugh ... not everything is bad ... :-)

> I agree that Ada is changing slowly compared to Java, but it stays under
> control and well designed. I don't think you can play on both sides.

DOS too stays under control, and by the way doesn't change any more ...
Couldn't change any slower :-)
Do you want Ada to end like that ?

> Maybe all Ada vendors should put some money on some sort of organization
to
> have the Ada libraries updated using standard Ada without vendor-specific
> additions. This would benefit the whole Ada community. Does such an
> organization exist ?

May be they should ?
Guess what : Yes I think so ...
That's what we were talking about in case you didn't notice ...
I'm not an Ada vendor, ACT is, Aonix is, Rational is.
Don't count on us to do compilers vendors job ...
When you buy a TV set, the vendor doesn't ask you to build the factory ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 757 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
uhevluxrf.fsf@wanadoo.fr...
> This is too much fun guys !!!
> As you said SUN add Java libraries. But ACT add GNAT libraries too right ?

Hum ... are you comparing SUN and ACT, is it what you have fun about ?

> The big difference is that "Java == SUN" but Ada is a standard and of
course
> ACT is not driving the changes, nor Aonix, nor Rational...

The real big difference is that the young Java is widely used and evolving
with user's needs, and the old Ada is not especially widely used, and
doesn't evolve a lot ...

> Now, you should submit library additions to the ARG to have a chance to
see
> them added in next revision.

That's not my job ... That's compiler vendors job ...

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> That's not my job ... That's compiler vendors job ...

I agree. And ACT is doing a quite good job with the GNAT.xx hierarchy. So
let's say that we have ACT/Ada, Aonix/Ada as we have SUN/Java,
Microsoft/Java... That's my point.

The big difference is that SUN is quite bigger than ACT, Aonix, RAtional :)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> > Maybe all Ada vendors should put some money on some sort of organization
> to
> > have the Ada libraries updated using standard Ada without vendor-specific
> > additions. This would benefit the whole Ada community. Does such an
> > organization exist ?
> 
> May be they should ?
> Guess what : Yes I think so ...
> That's what we were talking about in case you didn't notice ...

I did notice in fact, hence my constructive proposal ;)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Well, this brings up a problem that is really a two-part contraption. Part
one is that there are certain "standard libraries" that are part of the Ada
Reference Manual standard. The other is that there are/can-be libraries out
there that address things like GUI interfaces, databases, etc. The second
part is difficult to get merged with the first part because it probably
doesn't belong in a language standard. (Although I think it would be
possible to extend the standard with certain mathematical libraries..... But
that's another problem)

Ada unfortuantely doesn't have a "standard" interface to GUI's databases,
OS's, etc. I agree that it would be useful to have some agreed-upon
collection of package specs that basically say "If you provide an interface
to BlahBlahBlah, it ought to conform to this..." At one time there was a
group trying to form some sort of standard library of this sort, but it
didn't get very far. I suspect that it would happen better if there was some
vendor supported interface that got increasingly formalized & enhanced such
that all vendors eventually adopted it. If, for example, Gnat came delivered
with a collection of this sort of thing and it wasn't Gnat specific
(adoptable by other vendors) it might catch on.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"nicolas" <n.brunot@cadwin.com> wrote in message
news:gkvb7.6275$R8.5682297@nnrp6.proxad.net...
> Thanks a lot ... I'd like to hear that kind of things more often
> I don't understand how Ada could stay up-to-date with one revision of
> standard libraries in 18 years ...
> And who knows when will be the next, if there is one ...
> No need to change language rules and syntax, but at least, as far as
> libraries are concerned, you are absolutely right.
> Ada users should be able to count on a complete set of libraries,
absolutely
> compatible with main available compilers and main platforms, gathered on
one
> official download site.
> If everybody releases his own version of the same library for his
favourite
> compiler, on his favourite platform, we'll get nothing really useful, and
> moreover, we'll prove to others languages users, that Ada users give
lessons
> to everybody about software development, and are unable to apply them in
> practice.
>
>
>

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> We are giving to others languages users, all what they need, to be convinced
> that Ada users are themselves unable to stick to what they promote.
> If you can't bypass GNAT.Regexp, obviously GNAT.Regexp should be in the set
> of standard libraries, it shouldn't be named GNAT, and should be available
> as it is for any compiler.

I agree. But at least ACT has done a job that nobody else have done before!

Maybe a regular expression library should be added into the standard Ada
library (I'm all for), but this is another story. I'm sure we think that the
library revision process is too slow, but face the truth, Ada market is small
and I doubt that the ARG have as much money than SUN :)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

There was an attempt to have a working group via SIGAda do this, but it
rather fell apart. Maybe a better way of doing it would be to glom onto some
existing collection of stuff and start enhancing/modifying it & making it
available to all vendors as easily licensable Ada source. (The license is
important - many people would not want to build apps with it if it came
under GPL for fear of infecting their own code.)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Pascal Obry" <p.obry@wanadoo.fr> wrote in message
news:uhevluxrf.fsf@wanadoo.fr...
>
> Maybe all Ada vendors should put some money on some sort of organization
to
> have the Ada libraries updated using standard Ada without vendor-specific
> additions. This would benefit the whole Ada community. Does such an
> organization exist ?
>

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2497 bytes --]

I think I would agree that the goal is not simply to *duplicate* what
someone else is doing. I have repeatedly emphasized in this forum that I
thought this would be a "Me Too!!!" response that gets us nowhere. However,
this is different from saying that the customer is buying a particular
*kind* of product and Ada should meet that need.

If Ada has available an answer that says "Yeah, here's your basic compiler
and you can go searching the net for thousands of disparate tools, bindings,
etc." and The Other Guy is saying "Yeah, here's your basic compiler bundled
with a whole integrated solution" and everyone is buying The Other Guy,
maybe Ada should strive to meet that same class of buyer where he is rather
than expect him to come around to liking downloading and integrating for
himself?

That said, I'd advice (again) that Ada needs to not only meet that
expectation, but *exceed* it as well. Look at the top 3 or 4 selling
development kits for a given segment and say "Yeah, I can give you all that
*and*then*some!*" Ada has a fair amount of "*and*then*some!*" built in - its
more reliable and easier to use. If it has the "all that" part and possibly
adds some additional tools beyond what The Other Guy has, then you've really
got something there.

How to get there is the *real* problem! :-)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"AG" <ang@xtra.co.nz> wrote in message
news:LH_a7.739$fg7.176384@news.xtra.co.nz...
>
> "nicolas" <n.brunot@cadwin.com> wrote in message
> news:bmya7.18382$Iu6.4283510@nnrp1.proxad.net...
> > "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com> a �crit
dans
> > le message news: 9ke9pa$lks$1@nh.pace.co.uk...
> > > So, IMHO, Ada needs to be more "Customer Centered" and offer the users
&
> > > potential users the kinds of things they are currently buying.
>
> Well, if you only offer what they are currently buying you aren't going to
> pull them
> away from what they are currently buying:) I thought the whole point was
to
> offer
> something better.
>
> >
> > I agree. No product, how excellent it is, can afford to forget this
> > elementary rule
>
> If that was true we would still be using stone hammers to hunt wooly
> mammothes
> (or whatever the plural is for mammoth). After all - that was exactly what
> the
> customers were buying at the time.
>
>
>
>

Re: Proving Correctness (was Java Portability)

[Darren New]

Pascal Obry wrote:
> And BTW, it is no fair to be upset by something that has been done. Why are
> you not upset by the fact that these components have not been ported by you
> Ada vendor for their compilers ?

Since the bulk of XML is string parsing and data representations, I'm
kind of surprised there's any portability problem at all. Isn't Ada
portable enough to manage parsing a string into a tree without having to
be ported to each compiler?

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST). Cryptokeys on demand. dnew@san.rr.com
"You probably noticed you can't breath underwater. Hence the tank."
   -- PADI instruction manual, page 107.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1329 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
u4rrlur8g.fsf@wanadoo.fr...
> I agree. But at least ACT has done a job that nobody else have done
before!

Nobody else have done it, and that's the first biggest problem.
The second biggest problem is that GNAT.xx is truly great for people who
decided to be tied to Gnat compiler, and almost useless to others, unless
they are ready to loose their time to adapt it, and follow its evolution,
each one finally having its own slightly different version of the same
things, each one loosing its time making again the same job in parallel.
I don't think being tied to one compiler is a sensible decision for a
company choosing Ada.
This goes against main reasons of Ada choice, and there is no justification
for such a decision.

I don't see why this standardisation job couldn't be made with GNU.
It wouln't be very difficult to have a standard Ada library tree, with first
absolute rule that those libraries must run with main Ada compilers on main
platforms, Windows in first place.
It's up to Ada vendors to settle that, not to Ada fans to run each one in
their own direction, telling :
"I've done a great job, specific to my compiler/platform, why don't you
adapt it to your compiler/platform ?"

No need to wait for Gnat integration in GCC 3.xx for that.

Re: Proving Correctness (was Java Portability)

[Stephen Leake]

"nicolas" <n.brunot@cadwin.com> writes:

> "Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
> ulmkxuy80.fsf@wanadoo.fr...
> > Sorry but I'm very upset by your mail. If nobody does the job the job wont
> > gets done that it !
> I'll start to reply your last remark.
> There is absolutely no attack about this XML library. It's great to have it
> for students and Gnat users.
> 
> But we were talking about standard and portable libraries designed to be
> independant of compilers.
> This is what Ada is designed for. This is about what Ada users are giving
> lessons to everybody.
> George gave me this XML library as an example to show that this type of
> library is available.
> 
> This is not the case.

ok; you have one example of one library that is not compiler-portable.

> <snip> 
> > Right. So ask you vendor to adapt an Open Source component or sell an
> > equivalent component.
> 
> All the thread is about the fact that this is not an option and goes against
> Ada promotion
> Ada is designed to write code as much as possible independant from platforms
> and compilers.
> An XML library has nothing specific to a platform or a compiler.
> You can develop one with code totally independant of the compiler, and there
> is no reason to do otherwise.
> 
> You may ask why we didn't develop that ourseleves :  it is for very simple
> reasons.
> - We are not paid for that
> - Java does it much better.

How many Java libraries are portable across _different_ Java
implemntations, from _different_ vendors? I don't follow closely, but
I seem to remember that even Sun's libraries are not portable across
different Sun implementations of Java. So what are we comparing to here?

> It's clear to me that a standard library shouldn't have to be adapted to
> each compiler.

That is a desirable goal. We are discussing how to get there. Clearly,
the first version of a library might be compiler dependent. If the
people developing the library want compiler independence, they have to
put in some effort.

I don't see how Ada is any different from C++ in this regard. Can you
use MS Foundation classes in the Borland C++ environment, with
complete portability? I don't think so.

> We have more than 1 million lines of Ada code and are very careful
> to have them compiler independant. That's the least you can expect
> from an Ada library. 

Well, maybe its the least _you_ expect :). I have different
expectations. 

> I strongly believe that the lack of standard available library you can use
> 'as it is' without any modification, whatever compiler you use, goes against
> Ada rationale ...

I agree. But that doesn't mean it's a point against Ada, in favor of
Java or C++.

> I don't care who should do it .... I'm a user and like any other user I
> choose the best tool available for my needs.
> If ACT does everything to force Ada world to be dependant of Gnat compiler,
> it is ACT's decision.

Shades of Microsoft, don't you think? Works for them :).

> But I think it could kill Ada, and therefore ACT ...

Obviously, they disagree. It's their jobs on the line, I expect they
know better.

> > Let me add that we have tried in AWS to be compiler independant
> > (see AWS.OS_Lib) but since 2 or 3 releases this is not true
> > anymore. I don't > have the time to try on all available
> > compilers, the GNAT library is so nice > that I just can't bypass
> > it (GNAT.Regexp for example). So yes now AWS is mostly working
> > with GNAT... But I'd love to see others porting some stuff to be >
> > able to avoid this dependences :)
> 
> I think it's a dead-end if everybody does his own job specific for his own
> tools, and wait for others to adapt it for others tools.
> This shows all the limitations of the process.

Yes. So what is a better process? 

> We are giving to others languages users, all what they need, to be
> convinced that Ada users are themselves unable to stick to what they
> promote. If you can't bypass GNAT.Regexp, obviously GNAT.Regexp
> should be in the set of standard libraries, it shouldn't be named
> GNAT, and should be available as it is for any compiler. Once again
> Ada users give strong lessons to the rest of the world, and do
> exactly the opposite in practice ...

Ok. How do we make that happen? ACT isn't going to do it without some
monetary return; that's what businesses do. 

-- 
-- Stephe

Re: Proving Correctness (was Java Portability)

[Stephen Leake]

"nicolas" <n.brunot@cadwin.com> writes:

> "Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
> u4rrlur8g.fsf@wanadoo.fr...
> > I agree. But at least ACT has done a job that nobody else have done
> before!
> 
> <snip>
> 
> I don't see why this standardisation job couldn't be made with GNU.

All of the code in the GNU project is contributed either by
volunteers, or by companies being paid to produce GNU code (ACT,
RedHat). Volunteers do what they want; companies do what they are paid
for. 

If you offer to pay money to ACT to maintain GNAT.Regexp in a way
that is compatible with Rational, they might consider it. If a large
number of people make the same offer, they will definitely consider
it.

On the other hand, if you port GNAT.Regexp once to Rational.Regexp,
you are under no obligation to track future GNAT changes. You are also
free to submit it as a GNU library, so others can use it.

Either way, somebody has to commit time or money.

-- 
-- Stephe

Re: Proving Correctness (was Java Portability)

[Stephen Leake]

"nicolas" <n.brunot@cadwin.com> writes:

> I'm not an Ada vendor, ACT is, Aonix is, Rational is.
> Don't count on us to do compilers vendors job ...

No, but we do count on you to pay them to do it.

> When you buy a TV set, the vendor doesn't ask you to build the
> factory ...

Actually, they do. Part of the cost of the TV set is paying for the
mortgage on the current factory, and investing in the next factory.
Same for Ada libraries; somebody has to pay for them. Part of the fee
we pay to ACT for support goes to building new libraries. 

The difference is in scale. When you sell millions of TVs, the
incremental cost of the next factory for each TV set is very small.
But ACT only sells hundreds (I'm guessing here) of support contracts,
so the incremental cost is higher.

-- 
-- Stephe

Re: Proving Correctness (was Java Portability)

[tmoran]

> If the people developing the library want compiler independence, they
> have to put in some effort.
  Amen!  In my experience with Claw, it takes a lot of time and effort,
most of which has to do with particular compilers' deficiencies or
outright bugs.  And a compiler independent library doesn't particularly
help an individual compiler vendor, so it's not very high on their
priority list.  It's generally better for them to work on things that
distinguish their compiler from the field, rather than things that
diminish its differences.

  It would be interesting to see what fraction of, say, the source code
at www.adapower.com, is compiler independent.

Re: The pace of change (was Proving Correctness (was Java Portability))

[Warren W. Gay VE3WWG]

Pascal Obry wrote:
> "nicolas" <n.brunot@cadwin.com> writes:
> > "Mike@xx" <Mike_member@newsguy.com> a �crit dans le message news:
> > 9klokd0nif@drn.newsguy.com...
> > > The Ada standard library development model is not up to the modern times.
> > >
> > > Look at how Sun manages Java. They have reqular releases of an updated
> > > Java (mostly in the standard libraries). This is how Ada should work. Ada
> > > standard libraries should be updated with more packages all the time.
> 
> I've never seen libraries updating itself right :) ? So who should do
> the job ? This is the real question.
> 
> > >
> > Thanks a lot ... I'd like to hear that kind of things more often
> > I don't understand how Ada could stay up-to-date with one revision of
> > standard libraries in 18 years ...
...snip...
> I agree that Ada is changing slowly compared to Java, but it stays under
> control and well designed. I don't think you can play on both sides.

In fact, if your "libraries" keep undergoing "rapid change" as Nicolas
suggested it should (whatever "keep up to date" suggests here), it then 
becomes a bigger problem to build any "permanance" into
developed applications. This is one reason, I'll not invest time in 
writing for Microsoft -- it just keeps churning and churning, and all 
microsoft based software becomes out of date all too quickly. That is
a bad investment of my premium time.

UNIces at least have some relative form of stability, among other
advantages.

The same (MS problem) holds for Java-- it continues to churn with
change. How can you write Open Source software without being fully 
dedicated to keeping up with the change?

I personally like to "WRAMO" -- that is, "Write Once, And Move On"

I don't want to spend the rest of my life in maintenance mode on a few
given projects. Ada combined with UNIX, provides a good stable combination,
where my "maintenance" will be minimal. This allows _me_ to move onto
new projects, while requiring minimal maintenance on previously 
released projects.

If being up-to-date requires "additions", then fine. They don't necessarily
have to be part of a standard to be useful (though standards are good).
As has been previously mentioned, the GNAT packages are one such example.

Therefore, there is nothing preventing you from using these "up to date" 
packages, which can include XML and anything else you can dream up. Of
course, if these new packages "evolve a lot", you'll be paying the piper
of change again. This mode of operating at least gives you a choice.
-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

Stephen Leake <stephen.a.leake.1@gsfc.nasa.gov> writes:

> "nicolas" <n.brunot@cadwin.com> writes:
> 
> > I'm not an Ada vendor, ACT is, Aonix is, Rational is.
> > Don't count on us to do compilers vendors job ...
> 
> No, but we do count on you to pay them to do it.

Well said. Either way, Nicolas, you just can't get something for free. You
should either comit some time or money.

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Ole-Hjalmar Kristensen]

Pascal Obry <p.obry@wanadoo.fr> writes:

> Stephen Leake <stephen.a.leake.1@gsfc.nasa.gov> writes:
> 
> > "nicolas" <n.brunot@cadwin.com> writes:
> > 
> > > I'm not an Ada vendor, ACT is, Aonix is, Rational is.
> > > Don't count on us to do compilers vendors job ...
> > 
> > No, but we do count on you to pay them to do it.
> 
> Well said. Either way, Nicolas, you just can't get something for free. You
> should either comit some time or money.
> 
> Pascal.
> 
> -- 
> 
> --|------------------------------------------------------
> --| Pascal Obry                           Team-Ada Member
> --| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
> --|------------------------------------------------------
> --|         http://perso.wanadoo.fr/pascal.obry
> --|
> --| "The best way to travel is by means of imagination"

I have to agree with Nicolas here. Please observe that he IS
committing money by buying compilers. To pay for library development
is pretty unheard of in the commercial world. Paying for the finished
library is another thing, of course.

The bottom line is that each language and compiler has its advantages
and disadvantages, including (but not limited to) price, availability,
support, and libraries. Choosing one is a (not always) simple
cost/benefit analysis.

Customers don't usually hang around to argue, the  just go somewhere
else if they cannot find what they're looking for.

-- 
Kabelsalat ist gesund.

Ole-Hj. Kristensen

Re: Proving Correctness (was Java Portability)

[Ole-Hjalmar Kristensen]

Excellent idea. This is essentially what happened with STL.

"Marin David Condic" <dont.bother.mcondic.auntie.spam@acm.org> writes:

> There was an attempt to have a working group via SIGAda do this, but it
> rather fell apart. Maybe a better way of doing it would be to glom onto some
> existing collection of stuff and start enhancing/modifying it & making it
> available to all vendors as easily licensable Ada source. (The license is
> important - many people would not want to build apps with it if it came
> under GPL for fear of infecting their own code.)
> 
> MDC
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/
> 
> 
> "Pascal Obry" <p.obry@wanadoo.fr> wrote in message
> news:uhevluxrf.fsf@wanadoo.fr...
> >
> > Maybe all Ada vendors should put some money on some sort of organization
> to
> > have the Ada libraries updated using standard Ada without vendor-specific
> > additions. This would benefit the whole Ada community. Does such an
> > organization exist ?
> >
> 
> 
> 

-- 
Kabelsalat ist gesund.

Ole-Hj. Kristensen

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1558 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
uzo9cu3o8.fsf@wanadoo.fr...
> Well said. Either way, Nicolas, you just can't get something for free. You
> should either comit some time or money.

I'm fully convinced that ACT would be very happy to get a lot of money from
potential customers :-) :-)
But Stephen and you are reducing the problem to ACT company and ACT
customers.

We were talking about Ada language in a much more general way.
You seem to expect people to be your customer, and pay you before you
satisfy them, while they may find what they need with others Ada compilers
or others languages, which is already satisfying.

When you are in a competitive market, and especially not the dominant actor,
that's a litttle optimistic :-)
Development languages market is a kind of competitive market where Ada is
not especially the dominant one ...

You know, usually, companies first invest to develop a viable product, and
only after that, get an income from their customers, if the product is
satisfying :-)
To rephrase what you said :
"Either way, ACT, you just can't get something for free. You should either
comit some time or money" to develop satisfying libraries before making
money with them ...

I don't understand very well how you can hope I (or anybody else) am going
to pay ACT, Aonix or Rational for an XML library, when I'm using a fully
satisfying, available, and free Java one ......
May be I missed something, but I don't think I'm the only one to have some
problems to understand your way of thinking ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1120 bytes --]

"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> a �crit dans le message
news: uitg1gmzi.fsf@gsfc.nasa.gov...
> All of the code in the GNU project is contributed either by
> volunteers, or by companies being paid to produce GNU code (ACT,
> RedHat). Volunteers do what they want; companies do what they are paid
> for.

Yes, that's a well known situation. I don't see how it is related with what
we were talking about.

> If you offer to pay money to ACT to maintain GNAT.Regexp in a way
> that is compatible with Rational, they might consider it. If a large
> number of people make the same offer, they will definitely consider
> it.

I don't need GNAT.Regexp  ...

> On the other hand, if you port GNAT.Regexp once to Rational.Regexp,
> you are under no obligation to track future GNAT changes. You are also
> free to submit it as a GNU library, so others can use it.

Thanks I was aware of that :-)
But that has nothing to do with what we were talking about ...
Availability of standard libraries, independant of compilers and platforms,
as a requirement for Ada to be a viable solution a few years from now.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2340 bytes --]

"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> a �crit dans le message
news: un15dgn63.fsf@gsfc.nasa.gov...
> ok; you have one example of one library that is not compiler-portable.

The fact is that we were looking standard ones which are ...
This one was proposed, it is not ...

> That is a desirable goal. We are discussing how to get there. Clearly,
> the first version of a library might be compiler dependent. If the
> people developing the library want compiler independence, they have to
> put in some effort.

I don't think so, companies interested in Ada must put some effort to make
the language use viable.
For Ada vendors it's not an option, unless they drop Ada and make another
job.
For Ada users it's a matter of deciding wether using Ada or using more
popular languages. A totally different story.
You see the vendor-customer relationship in a completely false direction.

> Well, maybe its the least _you_ expect :). I have different
> expectations.

my expectations and yours have something in common.
They are not that important, whatever they are.
The question is to know if companies using Ada will go on that way or not.
And moreover if new companies are going to come to Ada.
Actually, the fact that Ada will still exist or not in 5 or 10 years is not
very important.
The only important thing is to have a good answer to this question, in order
to make the right decision about language choice.

> I agree. But that doesn't mean it's a point against Ada, in favor of
> Java or C++.

I think you take all that in a far too much religious way.
This is not the war of languages with good ones and villains.
This is just a matter of software developpers choosing their tools, like
someone choosing a srewdriver or a hammer
The screewdriver is everything but the ultimate goal of your construction.

> > If ACT does everything to force Ada world to be dependant of Gnat
compiler,
> > it is ACT's decision.
> Shades of Microsoft, don't you think? Works for them :).

In a slightly less successful way :-)
I wouldn't exchange 0.001 % of microsoft against 100% of ACT :-)

> Ok. How do we make that happen? ACT isn't going to do it without some
> monetary return; that's what businesses do.

You have several compiler vendors, a RM well defined.
Ada promotion is certainly not the matter of one single company.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1280 bytes --]

"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> a �crit dans le message
news: uelqpgmrx.fsf@gsfc.nasa.gov...
> Actually, they do. Part of the cost of the TV set is paying for the
> mortgage on the current factory, and investing in the next factory.
> Same for Ada libraries; somebody has to pay for them. Part of the fee
> we pay to ACT for support goes to building new libraries.

What you pay a specific company like ACT for what they provide is your
decision.
We were talking about Ada and libraries in a much more global way.

> The difference is in scale. When you sell millions of TVs, the
> incremental cost of the next factory for each TV set is very small.
> But ACT only sells hundreds (I'm guessing here) of support contracts,
> so the incremental cost is higher.

No the difference is that you pay for the product you want to buy, once the
product is finished, available, working, if you like it, and if there is not
a cheaper better one besides.
You don't pay to build the factory, the company invest a lot, and can count
on an income only if the result is good.
I'd really like to see you in a TV store :-)
"Stephen, give me money now, I'm going to develop a great TV set, build a
factory, and I promise you will have a great TV set within a few years ..."

Re: Proving Correctness (was Java Portability)

[Leif Roar Moldskred]

nicolas <n.brunot@cadwin.com> wrote:

> The fact is that we were looking standard ones which are ...
> This one was proposed, it is not ...

Well, yes - but what _other_ language than Java has an _XML library_
that's standard, and portable across all compilers? 

Sure, it would be a Good Thing To Have (tm), but aren't you setting
somewhat high demands? XML is a new technology, after all - and it
takes time to make a hard standard.

And Java does have advantages here that Ada does not - the fact that a
library for Java only have to target one platform: the java virtual
machine. 

And, strictly on gut-feeling since I'm not familiar with Sun's
practices here, I suspect that a standard library for Ada would need
(possibly much) more stringent specifications than what's common for Java.

(And to wouldn't an official Ada standard library have to go through
the ISO standardization process as well?)


Leif Roar Moldskred
wondering

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 820 bytes --]

"Leif Roar Moldskred" <lrm@huldreheim.provida.no> a �crit dans le message
news: kcOb7.4612$e%4.140534@news3.oke.nextra.no...
> Well, yes - but what _other_ language than Java has an _XML library_
> that's standard, and portable across all compilers?

In my opinion, portable libraries which could be only a standard of facts,
as long as you don't violate RM rules, is not a question of fighting against
other languages.
It's just about starting to put in practice what Ada has claimed since the
beginning of its existence.

Ada talks about reuse, software components, portability.
An XML library manipulating files and strings, is compiler dependant, and
the build is not even tested on Windows.

Honestly, a  C++ or a Java programmer seeing that has some reasons not to
take very seriously Ada users lessons ...

Re: Proving Correctness (was Java Portability)

[Leif Roar Moldskred]

nicolas <n.brunot@cadwin.com> wrote:

> In my opinion, portable libraries which could be only a standard of facts,
> as long as you don't violate RM rules, is not a question of fighting against
> other languages.

Well, in that case I've officially lost your point. No, Ada doesn't
have a lot of standard, cross-compiler, cross-platform libraries. So?

> It's just about starting to put in practice what Ada has claimed since the
> beginning of its existence.

> Ada talks about reuse, software components, portability.

Hmmm, I'll admit I'm an Ada-neophyte, but I never thought the Ada
community hyped these aspects of Ada, anymore than other
object-oriented programming languages.

> An XML library manipulating files and strings, is compiler dependent, and
> the build is not even tested on Windows.

> Honestly, a  C++ or a Java programmer seeing that has some reasons not to
> take very seriously Ada users lessons ...

I can see that they might jump to that conclusion, but I'll disagree
that they have a reason to. The lack of a wide selection of XML
libraries is more a question of demand than of the language. 

And as for this particular XML library, it seems to me to be your
typical one-man open source project. To request that a one-man
endeavor should be written and tested for several different
compiler/platform combinations is, in my opinion, unreasonable.

A better test on the reuse / portability of Ada, would be to consider
what additional work needs to be done to make this one-compiler,
one-platform library generally portable. 


Leif Roar Moldskred
confused, but opinionated

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2176 bytes --]

"Leif Roar Moldskred" <lrm@huldreheim.provida.no> a �crit dans le message
news: y%Ob7.737$ep5.11352@news1.oke.nextra.no...
> Well, in that case I've officially lost your point. No, Ada doesn't
> have a lot of standard, cross-compiler, cross-platform libraries. So?

Ada doesn't  have a lot of standard, cross-compiler, cross-platform
libraries, this is in my opinion, one of the main reason why so few people
use Ada.
That's the subject of the dicussion.

> Hmmm, I'll admit I'm an Ada-neophyte, but I never thought the Ada
> community hyped these aspects of Ada, anymore than other
> object-oriented programming languages.

I've seen those aspects extremely highly hyped for about 12 years ...
And if you forget them, I think Ada loose the most part of its reason to
exist.

> I can see that they might jump to that conclusion, but I'll disagree
> that they have a reason to. The lack of a wide selection of XML
> libraries is more a question of demand than of the language.

There is few demand for 2 basic reasons
- There are very few Ada users
- Ada users are often very interested in making their own version of already
existing things, forgetting that goes against all their claims to justify
Ada use.

> And as for this particular XML library, it seems to me to be your
> typical one-man open source project. To request that a one-man
> endeavor should be written and tested for several different
> compiler/platform combinations is, in my opinion, unreasonable.
> A better test on the reuse / portability of Ada, would be to consider
> what additional work needs to be done to make this one-compiler,
> one-platform library generally portable.

From my experience, well written Ada code portability is almost
straightforward as long as you care about not using specific compiler
packages or pragmas.
The additional work is almost nothing, The benefits are enormous.
After all, that's what Ada says about maintenance to justify its use.
When you've tested your code against Windows and Linux, with 2 compilers or
let's say just with Gnat public version, if you care about not using
anything compiler specific, any problem you find is very likely a compiler
bug.

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> I don't understand very well how you can hope I (or anybody else) am going
> to pay ACT, Aonix or Rational for an XML library, when I'm using a fully

To have support.

> satisfying, available, and free Java one ......

There is no support for this one, right ? If you find a bug...

> May be I missed something, but I don't think I'm the only one to have some
> problems to understand your way of thinking ...

This is a good example, you have a free (libre) XML library for Java and
ACT. Just use it. But if you want to use it in a commercial software it could
make sense to have support, here it is possible with ACT I doubt you can get
support for the Java library.

I think the misunderstanding between us is that you talk about product and I'm
talking about support. Or maybe I (one more time) missed the point :)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

"nicolas" <n.brunot@cadwin.com> writes:

> Availability of standard libraries, independant of compilers and platforms,
> as a requirement for Ada to be a viable solution a few years from now.

I agree. The trouble is that again the equation is not fair.

        Java = SUN

        Ada  = ACT, Aonix, Rational...

The Ada market is divided and this will be lot harder than for Java. It is
always easier to be the only driver :)

But don't get me wrong, I'd love to see more standardized libraries. Now to be
constructive, just try to propose something that WILL work! Of course you
can't expect Aonix to drive, build nice library and let others integrate them
for free (gratis) !

I think we need some kind of organization getting money from all Ada vendors
to drive the Ada libraries development (not the language).

Do you see something better ?

Now to be fair, it is quite easy to propose that, but I really don't know if
this is viable solution :) And since many Ada vendors just don't read this
NG...

At least there is nothing to gain to just groan on this news group :)

Anyway I will stop reading this thread as I think we are going nowhere.

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Leif Roar Moldskred]

nicolas <n.brunot@cadwin.com> wrote:

> Ada doesn't  have a lot of standard, cross-compiler, cross-platform
> libraries, this is in my opinion, one of the main reason why so few people
> use Ada.
> That's the subject of the dicussion.

Yes, but then neither does C or C++. I really think Java is the odd
man out here, rather than Ada. (And I wouldn't be at all surprised if
some of the 'standard libraries' for Java have problem with the ":"
file separator on Macintosh, or similar glitches.)

C and C++ certainly have a much wider selection of libraries, but how
many of those can really be said to be cross-compiler, cross-platform?
Sure, you get some that can be used across a couple of different
compilers, and a couple of different platforms - but in the context of
_standard_ libraries, very few would really be wide-reaching enough to
be valid.

> I've seen those aspects extremely highly hyped for about 12 years ...
> And if you forget them, I think Ada loose the most part of its reason to
> exist.

Well, personally I consider robustness, safety and ease of maintenance
to be more central - so I wouldn't agree that it would lose "most part
of its reason to exist."

I've no first-hand experience, so this is an honest question and not
rethoric, but how many Ada projects today have chosen Ada mainly
because of software-reuse or portability reasons, as opposed to those
who've chosen it mainly for other reasons?

> There is few demand for 2 basic reasons
> - There are very few Ada users
> - Ada users are often very interested in making their own version of already
> existing things, forgetting that goes against all their claims to justify
> Ada use.

Hopefully, the open-source movements might alleviate the second reason
some. If Ada software packages becomes easily available on the
Internet, people might get into the habit of looking before
leaping. 

Remember, languages like Java and Perl grew together with the wide
spread of the Internet, and the large, easily available code-base that
makes these languages so handy is probably a direct result of that.

> From my experience, well written Ada code portability is almost
> straightforward as long as you care about not using specific compiler
> packages or pragmas.
> The additional work is almost nothing, The benefits are enormous.
> After all, that's what Ada says about maintenance to justify its use.
> When you've tested your code against Windows and Linux, with 2 compilers or
> let's say just with Gnat public version, if you care about not using
> anything compiler specific, any problem you find is very likely a compiler
> bug.

I think that was my point, really. ;-)


-- 
Leif Roar Moldskred

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1411 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
uvgk0tb03.fsf@wanadoo.fr...
> To have support.
One of the reasons we don't use Gnat is that we find more than 10 bugs in
Gnat for just one (with easy workaround)  in Objectada or Apex.
Good support is nothing if you spend more time with the support than with
your actual work ...
May be that is specific to our code, but we work with our code ...

> There is no support for this one, right ? If you find a bug...
Up to now, no problem ... Don't worry for us, we'll cry for help if needed
:-)
Once again, from the customer side, support doesn't mean paying and working
for your vendor more than for yourself ...

> This is a good example, you have a free (libre) XML library for Java and
> ACT. Just use it. But if you want to use it in a commercial software it
could
> make sense to have support, here it is possible with ACT I doubt you can
get
> support for the Java library.

We have more than 15 years experience in commercial software ...
We have been ACT customers, and know people being currently ACT customers
May be we are stupid, but as we are responsible for what we do, please let
us decide what is good for us :-)

> I think the misunderstanding between us is that you talk about product and
I'm
> talking about support. Or maybe I (one more time) missed the point :)

You talk about ACT, I talk about Ada, that is the point.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2394 bytes --]

"Leif Roar Moldskred" <lrm@huldreheim.provida.no> a �crit dans le message
news: 5OPb7.4643$e%4.140738@news3.oke.nextra.no...
> nicolas <n.brunot@cadwin.com> wrote:
> Yes, but then neither does C or C++. I really think Java is the odd
> man out here, rather than Ada. (And I wouldn't be at all surprised if
> some of the 'standard libraries' for Java have problem with the ":"
> file separator on Macintosh, or similar glitches.)

There are problems. You have to deal at least with IE and Netscapes specific
things
No language is perfect. Otherwise why using Ada ?

> C and C++ certainly have a much wider selection of libraries, but how
> many of those can really be said to be cross-compiler, cross-platform?

The situation is not as dramatic as people think.
We've been using C commercial libraries for years without any bug or problem
for Windows, Linux, Sun, Irix

> Sure, you get some that can be used across a couple of different
> compilers, and a couple of different platforms - but in the context of
> _standard_ libraries, very few would really be wide-reaching enough to
> be valid.

In my personal opinion, Gnat/Objectada/Rational  Windows/Linux is a good
deal if not enough ...

> I've no first-hand experience, so this is an honest question and not
> rethoric, but how many Ada projects today have chosen Ada mainly
> because of software-reuse or portability reasons, as opposed to those
> who've chosen it mainly for other reasons?

Ours and some others we work or have been working with ...
Most Ada projects I know made Ada choice for software-reuse or portability
reasons, which are very important for easy maintenance.

> Hopefully, the open-source movements might alleviate the second reason
> some. If Ada software packages becomes easily available on the
> Internet, people might get into the habit of looking before
> leaping.

I completely agree, if they don't have to worry about compiler portability
when there is no reason to do so.
Due to the current situation of compiler vendors, open-source movements are
likely to be the only chance.

> Remember, languages like Java and Perl grew together with the wide
> spread of the Internet, and the large, easily available code-base that
> makes these languages so handy is probably a direct result of that.

I agree too. So portable libraries related to Internet (like XML for
example) wouldn't harm Ada promotion.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 185 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
upua8tadu.fsf@wanadoo.fr...
> And since many Ada vendors just don't read this
> NG...

I'm quite certain they do

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 855 bytes --]

"Pascal Obry" <p.obry@wanadoo.fr> a �crit dans le message news:
upua8tadu.fsf@wanadoo.fr...
> I agree. The trouble is that again the equation is not fair.
>         Java = SUN
>         Ada  = ACT, Aonix, Rational...

You forget there was a big fight between Microsoft and Sun about Java and
things are not as simple as you say
Anyway there is nothing about being fair or not.
Developpers need tools, they have to choose the right ones, not being kind
with one company or the other, whatever it is, Microsoft, Sun, Aonix,
Rational, ACT, or whatever you want.

> It is always easier to be the only driver :)

One of JCP faq is : Q: What prevents Sun from controlling or dominating the
groups that develop and maintain Java specifications?
well that's not as simple as that, but at least they understood that was
their only chance against Microsoft.

Re: Proving Correctness (was Java Portability)

[Larry Kilgallen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 763 bytes --]

In article <YpOb7.13370$II7.2424964@nnrp4.proxad.net>, "nicolas" <n.brunot@cadwin.com> writes:
> "Leif Roar Moldskred" <lrm@huldreheim.provida.no> a �crit dans le message
> news: kcOb7.4612$e%4.140534@news3.oke.nextra.no...
>> Well, yes - but what _other_ language than Java has an _XML library_
>> that's standard, and portable across all compilers?
> 
> In my opinion, portable libraries which could be only a standard of facts,
> as long as you don't violate RM rules, is not a question of fighting against
> other languages.
> It's just about starting to put in practice what Ada has claimed since the
> beginning of its existence.

So all you need is an economic model for doing this in a multivendor
environment.  I cannot think of any worked examples.

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

nicolas <n.brunot@cadwin.com> wrote:
 
: I don't think you understood what is the problem.
: Don't spend your time trying to convince me, you have the whole rest of the
: world to convince ...

(Note: I'm not writing to convince you, but I found
the style of your postings could easily make later readers
misunderstand. Someone said fear-mongering, and I have to agree
... Gosh why have I started this)

<:-) spoof-mode>
Certainly.  You know, don't try to fool me, Ada isn't exactely
Software Heaven, Ada envs are not as easy to use as Java, and
some libraries are not for free and aren't portable, and out
of the box, which is against Ada rationale. And if this won't
change by introduction of cost-free standard big libraries,
Ada will cease to exist "in 5 or 10 years" and that's "not very
important." Who knows?

Much the same situation as with APL. You know, that dead
language from decades ago that only a few APL fans are still
using. A bit puzzling that these guys are not exactly starving,
and that development has been continuous and stable, despite
the lack of standard workspaces, but ask anyone... I'll stay
with popular languages.

Conclusion: 99%+ of all software projects in the future
will not use any language but one that has free standard DWIM libraries,
and this language is not going to be Ada.
</:-)>


I think I failed in explaining that there is no necessity
to follow each and every of your implicit or explicit criteria
of estimating what matters are like and should be like now.
Your style of argument
suggested to me that there is ("Everyone..., just ask.").
One criterion seems to be polularity. 
You know what? I'm happy in my niche when it allows
me to live, and all I care (a bit) is that this niche isn't
shrinking too much. And I'm writing Java and have written Windows
programs.  No problem.

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 560 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9kolv7$t0m$1@a1-hrz.uni-duisburg.de...
> You know what? I'm happy in my niche when it allows
> me to live, and all I care (a bit) is that this niche isn't
> shrinking too much.

That has nothing to do with Ada users concerns about seeing Ada language
promoted and more widely used.
If your objective is to stay in your niche, which is perfectly
understandable, I don't understand why you try to argue with people having
totally different concerns.
You are not concerned ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 484 bytes --]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> a �crit dans le message
news: 9kolv7$t0m$1@a1-hrz.uni-duisburg.de...
> Note: I'm not writing to convince you,
Who else then :-)

>Someone said fear-mongering, and I have to agree
> ... Gosh why have I started this)

It is often not so bad news not to please everybody :-)
Especially when the someone concerned was replying without having understood
and obvioulsy hardly read what it was about :-)
Don't worry I'll stop here :-)

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Well, one way to get it done would be for a handful of Ada advocates to
build a desirable library under the Ada Developers Cooperative License and
make it available on the net. Compiler vendors could freely bundle the
library(ies) with their compilers and would be encouraged to do so. End
users could use the library(ies) at no cost to develop whatever they want.
The original developers would be doing this as a kind of sweat equity that
would potentially make them some money if/when someone utilizes the
library(ies) in their finished commercial product.

Asking someone to volunteer to make a library in his spare time and make it
available everywhere freely and never offer any compensation to him is going
to require that we find either a saint or a fool. I don't see any big rush
on the part of developers here to do this - so maybe that is evidence that
giving it away free is not sufficient incentive to get the job done. Since
compiler writers are reluctant to do this themselves and can't afford
necessarily to fund a bunch of us to do it for them, I'd think it would be
nothing but up-side for them to have it done under the ADCL.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Pascal Obry" <p.obry@wanadoo.fr> wrote in message
news:upua8tadu.fsf@wanadoo.fr...
>
> But don't get me wrong, I'd love to see more standardized libraries. Now
to be
> constructive, just try to propose something that WILL work! Of course you
> can't expect Aonix to drive, build nice library and let others integrate
them
> for free (gratis) !
>

Re: Proving Correctness (was Java Portability)

[Stephen Leake]

Ole-Hjalmar Kristensen <ohk@clustra.com> writes:

> I have to agree with Nicolas here. Please observe that he IS
> committing money by buying compilers. To pay for library development
> is pretty unheard of in the commercial world. Paying for the finished
> library is another thing, of course.

How is that different? Part of the profit from sales of Library 1 is
used to pay for developing Library 2. At least, it is in any company
that wants to stay in business.

Again, it's a matter of scale. A company that sells 1 million copies
of its library can charge very little per copy for developing the next
one. Unfortunately, no Ada companies have sales in that range.

> The bottom line is that each language and compiler has its
> advantages and disadvantages, including (but not limited to) price,
> availability, support, and libraries. Choosing one is a (not always)
> simple cost/benefit analysis.

True.

> Customers don't usually hang around to argue, the just go somewhere
> else if they cannot find what they're looking for.

Or offer to pay for what they need, when they can't find it.

-- 
-- Stephe

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Yeah, but you will find very few businessmen who are going to fully fund the
development of someone else's product. (Not the smart ones, at least.) Why
should I pay Company X to build me a custom library that I want and then
just give them the full rights to it? Why wouldn't I develop it in-house for
the full cost less Company X's profit and keep the result either a) as a
competitive advantage or b) to resell myself for additional $$$?

Now if 20 companies all want the same thing, they might see some point in
*sharing* the development cost - but again, they'd want to retain some
rights in the end product because of their capital investment. Does Company
X want these 20 customers to be their "partners"? Or is it ultimately going
to degenerate to the case where Company X puts up its risk capital to
develop a product and then sells it for what the market will bear in the
hope that it can sell enough units to recoup the cost and then some?

IOW: The statement "If you really want it, you'll pay for it" may be true,
but the retort will be "...And then *I* own it - not you."

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> wrote in message
news:uitfzg40d.fsf@gsfc.nasa.gov...
>
> Or offer to pay for what they need, when they can't find it.
>

Re: Proving Correctness (was Java Portability)

[David Starner]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9kosp0$dje$1@nh.pace.co.uk...
> Well, one way to get it done would be for a handful of Ada advocates to
> build a desirable library under the Ada Developers Cooperative License and
> make it available on the net. Compiler vendors could freely bundle the
> library(ies) with their compilers and would be encouraged to do so.
[...]
> I'd think it would be
> nothing but up-side for them to have it done under the ADCL.

If it was done under an open source license, ACT might add it to both the
private and public releases of GNAT, and it would be included in at least
one operating system (Debian) as soon as possible. If it's done under the
ADCL, ACT won't add it to the public release, and I seriously doubt it would
add it to the private releases (ACT has shown reluctance to pass costs onto
their customers, as shown by them no longer using cygwin), and I don't know
of any OS vendors who would jump to include it.

> Asking someone to volunteer to make a library in his spare time and make
it
> available everywhere freely and never offer any compensation to him is
going
> to require that we find either a saint or a fool.

I find that somewhat insulting; there are 400 Debian developers (including
me) who are putting together an operating system for no monetary
compensation, and I can tell you that few of us are saints.

Frankly, not everyone values monetary compensation as much as you do. I have
enough money and quasi-monetary resources to keep me feed, clothed and in
college for the next year, with a little money for computer and roleplaying
junk on the side. I don't need money; I'm looking peer respect and
intellectual stimulation. Personally, the money coming from the ADCL would
have too much legal complexity to be worth anything; you'd need to get an
accountant or lawyer involved to handle anything. If thousands were coming
in, it might be worth it, but I rather bet on "Make Money Fast" schemes.

> I don't see any big rush
> on the part of developers here to do this - so maybe that is evidence that
> giving it away free is not sufficient incentive to get the job done

IMO, the ADCL would be an overall negative incentive; it would drive away
any open-source people who would work on it, and attract very few people who
care about money (who probably figure it the same way I do; there's a small
chance of this being worth anything, so going into expecting to get enough
money to pay for my time is stupid.)

Sometimes things just don't come together right; the world of programming is
littered with dead projects. If someone wants to get together another try
(or restart an old one) under a reasonable licenes (XFree or GNAT-modified
GPL - something with few strings attached), I'm willing to work on it. Maybe
this time it will work; maybe it won't. I don't think using the ADCL will
affect that positively.

--
David Starner - dstarner98@aasaa.ofe.org
"The pig -- belongs -- to _all_ mankind!" - Invader Zim

Re: Proving Correctness (was Java Portability)

[Florian Weimer]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> writes:

> Asking someone to volunteer to make a library in his spare time and
> make it available everywhere freely and never offer any compensation
> to him is going to require that we find either a saint or a fool.

Why do you think so?  Perhaps the problem tackled by the library is
interesting and challenging.  Some people climb mountains in their
free time, some write free software libraries.  I don't think these
two things are so much different.

OTOH, coding all the day and coming home just to code more stuff is a
bit sick. ;-)

Re: Proving Correctness (was Java Portability)

[tmoran]

>Frankly, not everyone values monetary compensation as much as you do. I have
>enough money and quasi-monetary resources to keep me feed, clothed and in
>college for the next year, with a little money for computer and roleplaying
>junk on the side. I don't need money; I'm looking peer respect and
>intellectual stimulation.
  Just the way I felt as a student living on a research assistantship -
until I discovered skiing and the costs thereof. ;)
--------
"You can tell the men from the boys by the price of their toys" annonymous

Re: Ada Components (was Java Portability)

[Lao Xiao Hai]

Leif Roar Moldskred wrote:

> Hmmm, I'll admit I'm an Ada-neophyte, but I never thought the Ada
> community hyped these aspects of Ada, anymore than other
> object-oriented programming languages.

Prior to Ada 95, and prior to abrogation of the Dod Ada policy, there was
a lot of effort expended in creating libraries.  Many of those libraries now
languish in obscurity.  For example, the GRACE components, known for
their high reliability, have disappeared from the scene since the company
that created them, EVB Software Engineering has also disappeared.   There
was some volunteer effort by David Weller and others to convert the
BOOCH components to Ada, but little publicity has been forthcoming about
the status of that effort.   The DoD had a large library of components and
even funded some organizations to manage repositories.  Those repositories
seem to have vanished, even though the software still exists somewhere.

When Ivan Stepanov decided to build the STL, he began with Ada 83 but
found the template model (generics) inadequate, and the inability to do
extensible inheritance to be a liability.   He switched to C++ and that is
now where the STL has its most mature representation.    Stepanov has
been variously supported financially by AT&T, HP and Silicon Graphics,
and perhaps by others.    No one is supporting anyone, at present, to build
an equivalent set of libraries for Ada, even though the 1995 Ada standard
addressed and surpassed the concerns originally raised by Stepanov.

Since so many DoD organizations have lost interest in Ada, there is no one
with the motivation to fund an STL-Ada project.   It is unlikely to succeed
as a purely volunteer effort since the other volunteer efforts mentioned have
disappeared.   We need someone like a Randy Brukhardt, designer of CLAW,
who has the tenacity to stick with this kind of project even when the revenues
are disappointing.

At my client sites, I don't see a lot of developers clamoring for reusable
components.
They certainly don't want to pay for components.   This is clear from the EVB
experience where support was good, product excellent, and demand was not
enough to keep the good developers on board.

With the current Ada standard, one could build a set of components that would
be far better than what one finds in C++ STL. However, we still have the
issue of who will pay for it.    An entrepreneur that accepts this challenge will
have to realize that selling the finished product is a monumental task.  It would
have to be sold along with the compilers and that would raise the price-tag of
the compilers.   Unless we can overcome the NIH syndrome, there is little
incentive for anyone to create a business model based on components at this time.

Richard Riehle

Re: Proving Correctness (was Java Portability)

[Larry Elmore]

nicolas wrote:
> 
> "Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> a �crit dans le message
> news: uelqpgmrx.fsf@gsfc.nasa.gov...
> > Actually, they do. Part of the cost of the TV set is paying for the
> > mortgage on the current factory, and investing in the next factory.
> > Same for Ada libraries; somebody has to pay for them. Part of the fee
> > we pay to ACT for support goes to building new libraries.
> 
> What you pay a specific company like ACT for what they provide is your
> decision.
> We were talking about Ada and libraries in a much more global way.

It was a specific example to illustrate a general principle. ("...
_like_ ACT...") You're free to substitute any other company name you
feel like, Nicolas. Unless there's one you know of that's an exception
to that statement?
 
> > The difference is in scale. When you sell millions of TVs, the
> > incremental cost of the next factory for each TV set is very small.
> > But ACT only sells hundreds (I'm guessing here) of support contracts,
> > so the incremental cost is higher.
> 
> No the difference is that you pay for the product you want to buy, once the
> product is finished, available, working, if you like it, and if there is not
> a cheaper better one besides.

Duh. And the price charged is calculated to cover all the costs of
development and doing business (including taxes), plus a certain
percentage for profit (some of which will be reinvested in R&D and other
parts of the company). If you sell fewer units, you need a higher profit
margin per unit to make the equation balance. From your statements
relating to economics, I'd guess you either have nothing to do with the
business side of your company, or else it's customers consist mostly of
government agencies or other companies dealing mostly with government
agencies.

> You don't pay to build the factory, the company invest a lot, and can count
> on an income only if the result is good.

You _don't_??!! _WHO DOES_, then, if not the customer?

> I'd really like to see you in a TV store :-)
> "Stephen, give me money now, I'm going to develop a great TV set, build a
> factory, and I promise you will have a great TV set within a few years ..."

Duh... What do you think investors do every day? How in the world do you
think the vast majority of companies got started in the first place?
Government mandate, or magic?

Larry

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

"David Starner" <dstarner98@aasaa.ofe.org> wrote in message
news:tn0jksmu38rs36@corp.supernews.com...
> If it was done under an open source license, ACT might add it to both the
> private and public releases of GNAT, and it would be included in at least
> one operating system (Debian) as soon as possible. If it's done under the
> ADCL, ACT won't add it to the public release, and I seriously doubt it
would
> add it to the private releases (ACT has shown reluctance to pass costs
onto
> their customers, as shown by them no longer using cygwin), and I don't
know
> of any OS vendors who would jump to include it.
>
Are you speaking for ACT?

I'd suspect that there will likely be a limited range of choices: 1) The
compiler vendors fund some sort of joint development of a semi-standard set
of libraries. 2) Somebody develops a set of libraries as a business venture
that gain some popularity and becomes a de facto standard. 3) Some good
hearted soul(s)develops some libraries and release them at no charge and
they become a de facto standard.

Options 1 & 2 start getting evaluated as business ventures and maybe they
happen and maybe they don't depending on if someone thinks there is profit
to be made. If it looks profitable, they put effort into enhancements,
maintenance, support, advertising, marketing, etc. Option 3 has happened on
a number of occasions - there are dozens of libraries out there filled with
useful or semi-useful Ada code. (And before you pick on me for being a
greedy capitalist, I've contributed some to that body of free stuff.) But
because there is no money in it and we all have day jobs, there isn't much
effort put into improving or promoting the products. Proof? They've been
around for some time now and I don't see them sitting in a distribution of
anyone's compiler. (I'd be glad if you could prove me wrong on that score.)


>
> I find that somewhat insulting; there are 400 Debian developers (including
> me) who are putting together an operating system for no monetary
> compensation, and I can tell you that few of us are saints.
>
Well don't presume I meant to call you a fool. Presume I meant to call you a
saint.


> Frankly, not everyone values monetary compensation as much as you do. I
have
> enough money and quasi-monetary resources to keep me feed, clothed and in
> college for the next year, with a little money for computer and
roleplaying
> junk on the side. I don't need money; I'm looking peer respect and
> intellectual stimulation. Personally, the money coming from the ADCL would
> have too much legal complexity to be worth anything; you'd need to get an
> accountant or lawyer involved to handle anything. If thousands were coming
> in, it might be worth it, but I rather bet on "Make Money Fast" schemes.
>
What you decide to do with your spare time is your business. If you get a
kick out of developing software and putting it out for others to use, fine.
Remember, I've done the same. But I'll bet you wouldn't go buy all the
equipment needed, rent office space, work 40 hours a week (or more) at it,
(day in and day out for a period of ??? years?) fund some advertising and
marketing for it, etc. all out of some sense of charity. Not unless you were
just independently wealthy and liked to tilt at windmills. As a result,
there aren't a huge supply of folks beating down the doors begging ACT,
Aonix, Rational, Averstar, et alia, if they could *please* work for them for
free developing a collection of Ada libraries. (If there are, maybe they
could direct the extras over to my house where they can mow my lawn. :-)


>
> IMO, the ADCL would be an overall negative incentive; it would drive away
> any open-source people who would work on it, and attract very few people
who
> care about money (who probably figure it the same way I do; there's a
small
> chance of this being worth anything, so going into expecting to get enough
> money to pay for my time is stupid.)
>
Well, maybe it makes some money and maybe it doesn't. Maybe it only makes
$20 or maybe it makes $1,000,000. Nobody knows at this juncture since it
hasn't been tried. What I am reasonably sure of is that if I were to put a
bunch of code out under the GPL (or some variant that doesn't retain any
financial remuneration) that it will basically make nothing for me in terms
of money. (O.K. I get that warm fuzzy feeling all over. Thanks.) But there's
a reasonable chance that *someone* is going to make money off of it - it
just won't be me.

If I'm going to put the software out there anyway, why not reserve some
limited right to earn something from it? It would still be out there with
source code available for all the hackers and students and garage-start-ups
and big corporations to use at no charge. They're still getting that same
"free ride" they'd get with GPLed code, right? But if someone goes out and
turns it into a product, the authors aren't left out in the cold without so
much as a thank-you note.


> Sometimes things just don't come together right; the world of programming
is
> littered with dead projects. If someone wants to get together another try
> (or restart an old one) under a reasonable licenes (XFree or GNAT-modified
> GPL - something with few strings attached), I'm willing to work on it.
Maybe
> this time it will work; maybe it won't. I don't think using the ADCL will
> affect that positively.
>
Well, I think one of the reasons that the last effort to build a library
collapsed was that nobody owned the project and nobody saw their own
interest being directly benefited by working on it sufficiently to make it
worth pressing hard to get it done. Maybe I'm wrong on this score, but I
think if someone saw a personal need being fulfilled or some gain to be had,
there would have been more direction and more pressure to produce something.
(Maybe there were just too many people involved as well. Lots of things go
wrong.)

My guess is that the compiler vendors could spur something along because
they have an interest in such a facility. If they formed up some sort of
committee and agreed on some requirements & scope, at least someone (or some
group) would have a clear direction to go in and a ready community of
customers. It might get developed as another open source freebie or it might
get built "on spec" - but it would more likely get built if someone has a
vested interest to persue or protect. Don't forget that this is where a lot
of the "volunteer" labor comes from in groups like SIGAda - some businesses
think they have a vested interest in supporting the language, the standards,
the end products of working groups, the info/knowledge obtained at trade
shows, etc. As a result they send folks to these gigs and pay their salaries
and plane fair and hotels and let them have time to do the work needed. They
aren't doing it because it gives them warm fuzzy feelings! :-)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/

Re: Proving Correctness (was Java Portability)

[Ed Falis]

tmoran@acm.org wrote:
> 
> >Frankly, not everyone values monetary compensation as much as you do. I have
> >enough money and quasi-monetary resources to keep me feed, clothed and in
> >college for the next year, with a little money for computer and roleplaying
> >junk on the side. I don't need money; I'm looking peer respect and
> >intellectual stimulation.
>   Just the way I felt as a student living on a research assistantship -
> until I discovered skiing and the costs thereof. ;)
> --------
> "You can tell the men from the boys by the price of their toys" annonymous

I've been out of school for 20 years, and feel the same way as Starner. 
So much for the wisdom of real life experience.  Now if someone wants to
take over my alimony payments ...

- Ed

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

As I said elsewhere, anybody who wants to write code free of charge and give
it away just because it presents a challenge, etc. is perfectly free to do
so. More power to them. I just don't see the hoards massing at the gates
just swelling with anticipation over the opportunity to work for free so
that someone else will cash in on it.

A lot of the code that ends up out there in "Open Source" (whatever
definition we have for that today...) is something a company may have
developed and made available because it isn't their "core business" and they
benefit from a wider audience, or it is maybe academic work (profs and/or
students) who were building things to gain knowledge and understanding and
didn't see some huge commercial potential in keeping it for themselves. In
both cases, someone did it because they had something to gain by doing so.
You just aren't going to lure tens of thousands of professional programmers
into making rock-solid, commercial grade development libraries unless they
have something they anticipate gaining from it. (Not always cash - but it'll
do as a substitute if you can't get the glory! :-)

Even things like Linux aren't turned into a commercial grade "product"
without someone doing it for the money. Taking something out of the realm of
"hobbyist" software and making it a real product that can compete heads-up
with something like Windows is only going to happen if someone makes a
living by doing that. (Netscape and GNAT would be two more examples of
something becoming a commercial-grade product because people are making a
living off of it.)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Florian Weimer" <fw@deneb.enyo.de> wrote in message
news:873d73twql.fsf@deneb.enyo.de...
> Why do you think so?  Perhaps the problem tackled by the library is
> interesting and challenging.  Some people climb mountains in their
> free time, some write free software libraries.  I don't think these
> two things are so much different.
>
> OTOH, coding all the day and coming home just to code more stuff is a
> bit sick. ;-)

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

"Larry Elmore" <ljelmore@home.com> wrote in message
news:3B706ADC.B4847AC3@home.com...
> nicolas wrote:
>
> > You don't pay to build the factory, the company invest a lot, and can
count
> > on an income only if the result is good.
>
> You _don't_??!! _WHO DOES_, then, if not the customer?
>
Eventually, the customer pays. But that's a lot different than asking the
customer to fund the development cost up-front. That's what the venture
capitalist or entrepeneur is supposed to do.

I think you are misinterpreting the intention here. I believe the point was
about asking the customer to just simply pay some cost-plus development
money in order to get a library. That is definitely not the same sort of
thing as eventually paying off a business's asset acquisition by repeated
sales of a product.

IOW: You go build it - if I like it and find it worth the price you charge,
I buy it. It isn't usually done in the other order - unless you're the
government! :-)

> > I'd really like to see you in a TV store :-)
> > "Stephen, give me money now, I'm going to develop a great TV set, build
a
> > factory, and I promise you will have a great TV set within a few years
..."
>
> Duh... What do you think investors do every day? How in the world do you
> think the vast majority of companies got started in the first place?
> Government mandate, or magic?
>
A bit unfair. An investor is not the same thing as a customer.

--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/

Re: Proving Correctness (was Java Portability)

[Larry Elmore]

nicolas wrote:
> 
> "Leif Roar Moldskred" <lrm@huldreheim.provida.no> a �crit dans le message
> news: y%Ob7.737$ep5.11352@news1.oke.nextra.no...
> > Well, in that case I've officially lost your point. No, Ada doesn't
> > have a lot of standard, cross-compiler, cross-platform libraries. So?
> 
> Ada doesn't  have a lot of standard, cross-compiler, cross-platform
> libraries, this is in my opinion, one of the main reason why so few people
> use Ada.
> That's the subject of the dicussion.

How many standard, cross-compiler, cross-platform libraries exist for
_any_ language? I'd be willing to bet more would be in Fortran than any
other language.

> > Hmmm, I'll admit I'm an Ada-neophyte, but I never thought the Ada
> > community hyped these aspects of Ada, anymore than other
> > object-oriented programming languages.
> 
> I've seen those aspects extremely highly hyped for about 12 years ...
> And if you forget them, I think Ada loose the most part of its reason to
> exist.

It's been no more "hyped" than any other OO language. If this is your
argument, why is C++ still around? There's still not a single compiler,
AFAIK, that implements the C++ standard completely and correctly. So I
rather doubt there's scads of "standard" libraries out there for it that
are truly cross-compiler, cross-platform.
 
> > I can see that they might jump to that conclusion, but I'll disagree
> > that they have a reason to. The lack of a wide selection of XML
> > libraries is more a question of demand than of the language.
> 
> There is few demand for 2 basic reasons
> - There are very few Ada users
> - Ada users are often very interested in making their own version of already
> existing things, forgetting that goes against all their claims to justify
> Ada use.

I think your second point is bogus. Reinventing the wheel goes on
everywhere and in every language (and often because the developer feels
he/she can do a better job than whoever came before).
 
> > And as for this particular XML library, it seems to me to be your
> > typical one-man open source project. To request that a one-man
> > endeavor should be written and tested for several different
> > compiler/platform combinations is, in my opinion, unreasonable.
> > A better test on the reuse / portability of Ada, would be to consider
> > what additional work needs to be done to make this one-compiler,
> > one-platform library generally portable.
> 
> From my experience, well written Ada code portability is almost
> straightforward as long as you care about not using specific compiler
> packages or pragmas.

No kidding. That's more than you can say about C/C++!!!

There were lots of C and C++ programmers long before there were anything
_like_ standard, cross-compiler, cross-platform libraries. How long was
C around before the most _basic_ C library could make that claim?

Larry

Re: Proving Correctness (was Java Portability)

[David Starner]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9kpq82$otf$1@nh.pace.co.uk...
> Are you speaking for ACT?

No. I merely note that their past history does not indicate them using
programs under the ADCL.

However, the public release of GNAT is done by Robert Dewar for the Free
Software Foundation, and the Free Software Foundation has made their opinion
on the matter more than clear. Not only will their version of GNAT not
include these libraries, it should not even mention them, as not to give
publicity to non-free software.

> > I find that somewhat insulting; there are 400 Debian developers
(including
> > me) who are putting together an operating system for no monetary
> > compensation, and I can tell you that few of us are saints.
> >
> Well don't presume I meant to call you a fool. Presume I meant to call you
a
> saint.

The problem is, I'm not a saint, and I'm not really a fool, and neither are
the people I work with. It's just a hobby, like any other, done for various
reasons.

> As a result,
> there aren't a huge supply of folks beating down the doors begging ACT,
> Aonix, Rational, Averstar, et alia, if they could *please* work for them
for
> free developing a collection of Ada libraries. (If there are, maybe they
> could direct the extras over to my house where they can mow my lawn. :-)

But there are a number of people who have developed Ada libraries for free.
If there were a focus - some sort of standard - it could have been
implemented by now. The problem's not really people's time in developing it;
it's getting a standard that the compiler people and the community will put
their support behind.

> Well, maybe it makes some money and maybe it doesn't. Maybe it only makes
> $20 or maybe it makes $1,000,000.

I can see how you could make $1,000,000 (though I find it highly
unlikely) -- I don't see how you could make $20, if more than one person
worked on it. Say John, Joesph, and Mohammad worked on. Who gets the checks?
What happens if (when) a check to someone doesn't show up in time? What is
on time? How does everyone make sure it's getting split right? This can get
hairer - Oklahoman John, Ukrainian Jospeh and Iranian Mohammad. Money
conversions; tariffs? If it's a million, we can hire a lawyer or accountant
part time to handle the money, for a few thousand dollars. If it's a $20
here and there? Things can get hostile when money's involved, too.

> If I'm going to put the software out there anyway, why not reserve some
> limited right to earn something from it? It would still be out there with
> source code available for all the hackers and students and
garage-start-ups
> and big corporations to use at no charge. They're still getting that same
> "free ride" they'd get with GPLed code, right? But if someone goes out and
> turns it into a product, the authors aren't left out in the cold without
so
> much as a thank-you note.

It's all a personal choice. Let me note, however, that there's a lot of
shareware where the authors saw little to nothing from it, while there's a
number of Open Source programmers who now make a living from their
programming.

> Well, I think one of the reasons that the last effort to build a library
> collapsed was that nobody owned the project and nobody saw their own
> interest being directly benefited by working on it sufficiently to make it
> worth pressing hard to get it done.

True. As a general rule, open source projects done by volunteers either have
one big leader (OpenBSD, Linux, most small programs) or have enough inertia
to form a volunteer group (Linux (it's not Linus driven anymore), Debian,
Apache) - and most the later started out as the first. Getting one person
who cares enough to pull the project through and who can command the respect
of the people is important.

> My guess is that the compiler vendors could spur something along because
> they have an interest in such a facility. If they formed up some sort of
> committee and agreed on some requirements & scope, at least someone (or
some
> group) would have a clear direction to go in and a ready community of
> customers.

That would certainly help.

> It might get developed as another open source freebie or it might
> get built "on spec"

I don't understand the contrast here; some open source programs are written
to exact specification, and some proprietary applications ignore whatever
standards whereever they feel like.

> Don't forget that this is where a lot
> of the "volunteer" labor comes from in groups like SIGAda - some
businesses
> think they have a vested interest in supporting the language, the
standards,
> the end products of working groups, the info/knowledge obtained at trade
> shows, etc. As a result they send folks to these gigs and pay their
salaries
> and plane fair and hotels and let them have time to do the work needed.
They
> aren't doing it because it gives them warm fuzzy feelings! :-)

Very true. But on the other hand, note the number of standards that just got
ignored. To the best of their knowledge, the GNU Pascal people are writing
the first widely available compiler to handle Extended Pascal. Likewise,
Unicode could have been some obscure standard, or a standard that Microsoft
"bought" (i.e. did most the work on, and is specialized for Microsoft
systems.) But Unicode managed to get buy-in - W3C, IETF and other non-ISO
standards organizations are including Unicode in their standards in a major
way, as are other ISO standards. A lot of people are pushing for better
Unicode support, including a lot of Open Source people. Not because it's a
standard that a few countries and companies thought they needed - because
it's a standard that people think they need and people like. Without those
people, your standard will go nowhere.

--
David Starner - dstarner98@aasaa.ofe.org
"The pig -- belongs -- to _all_ mankind!" - Invader Zim

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1898 bytes --]

Thanks,
actually, your replies are better than mines :-)

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> a �crit dans
le message news: 9kprk1$pof$1@nh.pace.co.uk...
>
> "Larry Elmore" <ljelmore@home.com> wrote in message
> news:3B706ADC.B4847AC3@home.com...
> > nicolas wrote:
> >
> > > You don't pay to build the factory, the company invest a lot, and can
> count
> > > on an income only if the result is good.
> >
> > You _don't_??!! _WHO DOES_, then, if not the customer?
> >
> Eventually, the customer pays. But that's a lot different than asking the
> customer to fund the development cost up-front. That's what the venture
> capitalist or entrepeneur is supposed to do.
>
> I think you are misinterpreting the intention here. I believe the point
was
> about asking the customer to just simply pay some cost-plus development
> money in order to get a library. That is definitely not the same sort of
> thing as eventually paying off a business's asset acquisition by repeated
> sales of a product.
>
> IOW: You go build it - if I like it and find it worth the price you
charge,
> I buy it. It isn't usually done in the other order - unless you're the
> government! :-)
>
> > > I'd really like to see you in a TV store :-)
> > > "Stephen, give me money now, I'm going to develop a great TV set,
build
> a
> > > factory, and I promise you will have a great TV set within a few years
> ..."
> >
> > Duh... What do you think investors do every day? How in the world do you
> > think the vast majority of companies got started in the first place?
> > Government mandate, or magic?
> >
> A bit unfair. An investor is not the same thing as a customer.
>
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/
>
>
>

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1026 bytes --]

"Larry Elmore" <ljelmore@home.com> a �crit dans le message news:
3B706ADC.B4847AC3@home.com...
>From your statements
> relating to economics, I'd guess you either have nothing to do with the
> business side of your company, or else it's customers consist mostly of
> government agencies or other companies dealing mostly with government
> agencies.

Sorry to disapoint you ...
We are a company small enough so that no side of the company is really far
from anybody.
Almost all our customers are private companies ready to jump to your
competitor if he is better or cheaper than you.
Or sometimes just because he feels like it, even if your product is far
better
I've learned about all that visiting customers and factories throughout the
world to understand prospects and customers needs.
I can assure you that investing unsuccessfully, in order to try to turn a
prospect in a customer, is not that uncommon.

I'll be curious to know what's your job, and how close you are of the
business side, to think otherwise ...

Re: Proving Correctness (was Java Portability)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 904 bytes --]

"Larry Elmore" <ljelmore@home.com> a �crit dans le message news:
3B707688.340B8A70@home.com...
> How many standard, cross-compiler, cross-platform libraries exist for
> _any_ language? I'd be willing to bet more would be in Fortran than any
> other language.

You are right, but I doubt that helps Ada a lot.

> It's been no more "hyped" than any other OO language. If this is your
> argument, why is C++ still around?

Microsoft ?

> I think your second point is bogus. Reinventing the wheel goes on
> everywhere and in every language (and often because the developer feels
> he/she can do a better job than whoever came before).

It's obvious to me that this is a counterproductive attitude.
People fight or at least should fight with very good reasons against this
kind of developer's feelings.

> No kidding. That's more than you can say about C/C++!!!

I agree. Why do you think we use Ada ?

Re: Proving Correctness (was Java Portability)

[Ole-Hjalmar Kristensen]

Stephen Leake <stephen.a.leake.1@gsfc.nasa.gov> writes:

> Ole-Hjalmar Kristensen <ohk@clustra.com> writes:
> 
> > I have to agree with Nicolas here. Please observe that he IS
> > committing money by buying compilers. To pay for library development
> > is pretty unheard of in the commercial world. Paying for the finished
> > library is another thing, of course.
> 
> How is that different? Part of the profit from sales of Library 1 is
> used to pay for developing Library 2. At least, it is in any company
> that wants to stay in business.
> 

Yes, but that's OK. I pay for something now, GET THE GOODS, and you
use the revenue to develop something better for the future. This is
substantially different from paying you now, and you promise to deliver
me something in the future.

> Again, it's a matter of scale. A company that sells 1 million copies
> of its library can charge very little per copy for developing the next
> one. Unfortunately, no Ada companies have sales in that range.
> 
> > The bottom line is that each language and compiler has its
> > advantages and disadvantages, including (but not limited to) price,
> > availability, support, and libraries. Choosing one is a (not always)
> > simple cost/benefit analysis.
> 
> True.
> 
> > Customers don't usually hang around to argue, the just go somewhere
> > else if they cannot find what they're looking for.
> 
> Or offer to pay for what they need, when they can't find it.
> 
> -- 
> -- Stephe

In which case the vendor is merely a contractor, and the customer owns
the product, if they have any business sense at all.

-- 
Kabelsalat ist gesund.

Ole-Hj. Kristensen

Re: Proving Correctness (was Java Portability)

[Ole-Hjalmar Kristensen]

Larry Elmore <ljelmore@home.com> writes:

> nicolas wrote:
> > 

<stuff deleted>

> > I'd really like to see you in a TV store :-)
> > "Stephen, give me money now, I'm going to develop a great TV set, build a
> > factory, and I promise you will have a great TV set within a few years ..."
> 
> Duh... What do you think investors do every day? How in the world do you
> think the vast majority of companies got started in the first place?
> Government mandate, or magic?
> 
> Larry

But he's not an investor, he's potential customer in the mass
market. There is a difference you know, and compilers and libraries
are commodities these days, not custom-made.

-- 
Kabelsalat ist gesund.

Ole-Hj. Kristensen

More Uniform Ada libraries (was: Proving Correctness)

[Larry Kilgallen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1309 bytes --]

In article <DT6c7.13785$SG4.2689164@nnrp4.proxad.net>, "nicolas" <n.brunot@cadwin.com> writes:
> "Larry Elmore" <ljelmore@home.com> a �crit dans le message news:
> 3B707688.340B8A70@home.com...
>> How many standard, cross-compiler, cross-platform libraries exist for
>> _any_ language? I'd be willing to bet more would be in Fortran than any
>> other language.
> 
> You are right, but I doubt that helps Ada a lot.

The point is not that people would not like to have uniform libraries
distributed with all compilers that are easy to use and not subject to
becoming obsolete.  The point is that nobody (including you, it would
seem) has a solid business model to achieve that goal.

When I say "business model", I mean to include "Nicolas write them all
and convinces all Ada compiler vendors to distribute them".  But even that
model has trouble at the "convinces all Ada compiler vendors" stage.

My understanding is that the Ada95 standard represents a compromise
based on what existing Ada83 compiler vendors were willing to accept.
At least one vendor dropped out, declining to accept the level of
change.  At least one other vendor joined in.  The point is that even
if your persuasive power is the ability to require something in the
standard, it is difficult to convince all Ada compiler vendors.

Re: More Uniform Ada libraries (was: Proving Correctness)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1074 bytes --]

"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> a �crit dans le
message news: 0TDoe8bALz3g@eisner.encompasserve.org...
>The point is that nobody (including you, it would
> seem) has a solid business model to achieve that goal.
>
> When I say "business model", I mean to include "Nicolas write them all
> and convinces all Ada compiler vendors to distribute them".  But even that
> model has trouble at the "convinces all Ada compiler vendors" stage.

May be I'm not very clear about my point, but I'll try to be some more :
Ada is one of my current tools.
I'm an Ada user and an Ada compiler vendor customer.

I don't have to define, and don't care about defining, the business model of
Ada solution providers.
My problem is to know if Ada will still be the right tool for my purpose in
a few years from now.
Or if I have to be prepared to drop this specific tool.

If I need a TV set, and if one TV set vendor doesn't have a product which
satisfies me,
I'm not going to teach him which business model he should adopt.
I'm simply going to chose another vendor.

Re: More Uniform Ada libraries (was: Proving Correctness)

[Larry Kilgallen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1083 bytes --]

In article <7U9c7.69$Ll6.174170@nnrp4.proxad.net>, "nicolas" <n.brunot@cadwin.com> writes:
> "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> a �crit dans le
> message news: 0TDoe8bALz3g@eisner.encompasserve.org...
>>The point is that nobody (including you, it would
>> seem) has a solid business model to achieve that goal.
>>
>> When I say "business model", I mean to include "Nicolas write them all
>> and convinces all Ada compiler vendors to distribute them".  But even that
>> model has trouble at the "convinces all Ada compiler vendors" stage.
> 
> May be I'm not very clear about my point, but I'll try to be some more :
> Ada is one of my current tools.
> I'm an Ada user and an Ada compiler vendor customer.
> 
> I don't have to define, and don't care about defining, the business model of
> Ada solution providers.

But you seem to feel there is a workable business model to achieve
uniformity across all Ada compilers, while the rest of us are less confident.

In the abstract sense, I want world peace, but I am not able to take
my business to another planet.

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:

:  What I am reasonably sure of is that if I were to put a
: bunch of code out under the GPL (or some variant that doesn't retain any
: financial remuneration) that it will basically make nothing for me in terms
: of money.

My small up-front contribution might only buy you a lawn
mower, but if I can make good use of a GPLed library,
I'm quite ready to invest some money to support the developer(s).
Maybe some others are interested, too.

And, now speaking for my company, we have actually once made
a somewhat more substantial offer for the development
of open code that we wouldn't own, and everyone could use.
Could that be an option for you, in principal?

 

Re: More Uniform Ada libraries (was: Proving Correctness)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1095 bytes --]

"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> a �crit dans le
message news: uiGL0WHzXluf@eisner.encompasserve.org...
> But you seem to feel there is a workable business model to achieve
> uniformity across all Ada compilers, while the rest of us are less
confident.

The business model is not my problem, and I don't have feelings about it.
The only feelings I have is that there are some customer side requirements,
coming from the reasons why we, and most Ada projects I know, made the Ada
choice about 12 years ago.
Our projects, and most projects I know, have nothing to do with embedded
systems and 'DOD similar' projects.
I don't believe that satisfying (for me) Ada solutions will be available in
the future, if those requirements are not fulfilled.
We have to know if we will be the last Ada users to have concerns, which are
not embedded systems and 'DOD similar' projects concerns.
If this is the case, we don't feel like fighting with Ada solution providers
which don't want to understand anything about what we need, and deliberately
ignore any other market's rules.

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Wellllll....... You *might* pay to get a custom library built rather than
shop for an off-the-shelf part. Sometimes you have a problem domain for
which no suitable library exists or there are licensing issues, etc. But the
one thing you can count on here is that any (non-brain-dead) businessman is
*not* going to pay "Ada Library Writers Inc." the full development cost,
plus profit and not walk away with 100% ownership of the end product.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Ole-Hjalmar Kristensen" <ohk@clustra.com> wrote in message
news:umqr8umyhp0.fsf@maestro.clustra.com...
> Larry Elmore <ljelmore@home.com> writes:
>
> But he's not an investor, he's potential customer in the mass
> market. There is a difference you know, and compilers and libraries
> are commodities these days, not custom-made.
>

Re: Proving Correctness (was Java Portability)

[Leif Roar Moldskred]

Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:
> Wellllll....... You *might* pay to get a custom library built rather than
> shop for an off-the-shelf part. Sometimes you have a problem domain for
> which no suitable library exists or there are licensing issues, etc. But the
> one thing you can count on here is that any (non-brain-dead) businessman is
> *not* going to pay "Ada Library Writers Inc." the full development cost,
> plus profit and not walk away with 100% ownership of the end product.

Actually, there might be good business-sense in doing just that in a
few cases. If the library is outside of the paying company's focus, it
might be better for the company to leave the ownership of the library
in the hands of the creators, and let them continue to develop and
support it; rather than demanding the rights for themselves, and then
never having the time or inclination to do anything more with the
library.

A second reason is even simpler: if nobody will write the library
unless you pay them the development cost, and lets them retain the
rights. For more specialized problem domains, it might be seller's
market.

-- 
Leif Roar Moldskred

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

"David Starner" <dstarner98@aasaa.ofe.org> wrote in message
news:tn1n46nn3deedb@corp.supernews.com...
>
> However, the public release of GNAT is done by Robert Dewar for the Free
> Software Foundation, and the Free Software Foundation has made their
opinion
> on the matter more than clear. Not only will their version of GNAT not
> include these libraries, it should not even mention them, as not to give
> publicity to non-free software.
>
What FSF decides to do is their own business. What ACT decides to do is
*their* own business. Right now ACT includes stuff (like AdaGIDE) in the
public distribution that aren't part of the "formal" release of GNAT. If you
were to write some useful library of stuff and ACT put it out on the disk
right next to GNAT how does that affect FSF? We're talking about a whole
separate body of code that can be included in any number of manners in
someone's compiler delivery. All they have to do is believe that it is to
their advantage to include some .ZIP file or directory on the CD they ship
you or the FTP site the make accessdable and there you have it. The FSF
doesn't exactly have a small garrison of storm troopers to leave at your
offices to keep you from doing things they don't like.


> The problem is, I'm not a saint, and I'm not really a fool, and neither
are
> the people I work with. It's just a hobby, like any other, done for
various
> reasons.
>
Hobbies are fine. However, I'd believe there is a better likelihood that
products will get brought to market at a high quality if someone is doing it
for something a little more tangible than the satisfaction someone gets from
their hobby.


>
> But there are a number of people who have developed Ada libraries for
free.

Never said there weren't. As a matter of fact, I've done that myself. (O.K.
I didn't put it out there under GPL, but I left a really strong hint about
just how easy it would be to get permission from me to use the libraries.)


> If there were a focus - some sort of standard - it could have been
> implemented by now. The problem's not really people's time in developing
it;
> it's getting a standard that the compiler people and the community will
put
> their support behind.
>
I'd agree that there are a number of people who might devote time to writing
some library. I was one of them and might be willing to do so again. I'd
agree that it would help if the compiler vendors said something to the
effect of "If someone writes something that looks like X and works half-way
well, I'll distribute it with my compiler." I just think that it would get
done a lot quicker and at a higher level of quality if the vendors said
"...And I'll write you a check for $X.XX if you do it..." Probably they
won't go for that, but if there were *some* way that the "volunteers" get
something out of it (besides the warm fuzzy feeling) I bet the odds of
getting it done and done well would go up. (Remember, there is a kind of
business benefit to be had by some potential participants - so it doesn't
necessarily have to be some financial transaction that creates the
incentive.)


>
> I can see how you could make $1,000,000 (though I find it highly
> unlikely) -- I don't see how you could make $20, if more than one person
> worked on it. Say John, Joesph, and Mohammad worked on. Who gets the
checks?
> What happens if (when) a check to someone doesn't show up in time? What is
> on time? How does everyone make sure it's getting split right? This can
get
> hairer - Oklahoman John, Ukrainian Jospeh and Iranian Mohammad. Money
> conversions; tariffs? If it's a million, we can hire a lawyer or
accountant
> part time to handle the money, for a few thousand dollars. If it's a $20
> here and there? Things can get hostile when money's involved, too.
>
No, I don't have illusions that somehow I'd make millions overnight because
I released a square-root subroutine under the ADCL. :-) My point has
consistently been that the ADCL holds out *some* hope of financial gain *if*
the software released under it is good enough to be useful to someone who
then wants to incorporate it in a product for sale.

All of those "what if" questions you ask are dealt with to some extent in
the articles by Dr. Leif. To the extent that they aren't dealt with - I
agree that they need to be dealt with. The ADCL is not a finished product -
it is a concept that is being kicked around and may ultimately emerge in
some form different from what I've seen so far. The fact that there are
questions and "what if" scenarios doesn't make it A Bad Thing. Go look back
at all the questions and "what ifs" that are discussed here and elsewhere
concerning the GPL.

>
> It's all a personal choice. Let me note, however, that there's a lot of
> shareware where the authors saw little to nothing from it, while there's a
> number of Open Source programmers who now make a living from their
> programming.
>
Sure an Open Source programmer can make a living from his work. No doubt. Go
look at the ACT programmers who are doing just that. And yes, there are lots
of people who have written some kind of softrware with the intention of
selling it for money (under a variety of license terms) and seen $0.00 from
their effort. It rather amounts to a big "So What?" Profit and loss are
going to be decided more by your business plan and your business skills than
by the license you use.

For some products, I can see how Open Source will make you a buck. If the
product is large enough and/or requires some significant level of support
and/or has some kind of value added products bundled with it that aren't
Open Source (manuals, packaging, telephone support, etc.) and/or priovides
an opportunity for consulting/education, then the business model may best be
served by putting the code under an Open Source license in order to get it
into the hands of as many people as is possible. However, I'd suspect that
some sort of semi-standard collection of utilities in the form of Ada source
libraries is not going to meet one or more of these criteria. If I put out a
megabyte of data structure code that is highly reliable and doesn't need
much explanation (and is maybe well documented?) it doesn't necessarily
afford the opportunity to take advantage of the other means of generating
revenue.

Would you hire a consultant to maintain a stable library of data structure
code or provide you with telephone support? (Unless I write really bad code!
:-) Maybe you'd hire a consultant to conduct some training in its proper
use, and/or *maybe* you'd shell out some money for a textbook/manual about
it, but probably not enough to generate a reasonable revenue stream that
would keep the writer interested in enhancing the product. In the mean time,
you might just take that megabyte of code and add your own megabyte of code
and suddenly have the next "Ms. Pacman" game craze that makes you a
billionaire and you might just send me a Thank You card for having provided
50% of your product development for you at no cost. :-)

>
> > It might get developed as another open source freebie or it might
> > get built "on spec"
>
> I don't understand the contrast here; some open source programs are
written
> to exact specification, and some proprietary applications ignore whatever
> standards whereever they feel like.
>
"On Spec" as in "On Speculation". I'm suggesting that someone might put
together a library of Ada stuff on speculation that they might be able to
find a way of making some money from its sale or support in some manner.


> > Don't forget that this is where a lot
> > of the "volunteer" labor comes from in groups like SIGAda - some
> businesses
> > think they have a vested interest in supporting the language, the
> standards,
> > the end products of working groups, the info/knowledge obtained at trade
> > shows, etc. As a result they send folks to these gigs and pay their
> salaries
> > and plane fair and hotels and let them have time to do the work needed.
> They
> > aren't doing it because it gives them warm fuzzy feelings! :-)
>
> Very true. But on the other hand, note the number of standards that just
got
> ignored. To the best of their knowledge, the GNU Pascal people are writing
> the first widely available compiler to handle Extended Pascal. Likewise,
> Unicode could have been some obscure standard, or a standard that
Microsoft
> "bought" (i.e. did most the work on, and is specialized for Microsoft
> systems.) But Unicode managed to get buy-in - W3C, IETF and other non-ISO
> standards organizations are including Unicode in their standards in a
major
> way, as are other ISO standards. A lot of people are pushing for better
> Unicode support, including a lot of Open Source people. Not because it's a
> standard that a few countries and companies thought they needed - because
> it's a standard that people think they need and people like. Without those
> people, your standard will go nowhere.
>

Not sure what standards have to do with it. I was suggesting that the people
who typically sign up for SIG committees and so forth have some modicum of
corporate support behind them. The corporate support is there because the
corporation perceives some business advantage to supporting a SIG.

If you are suggesting that a SIG may not be the best way to produce a
standard library of Ada tools because they'd have trouble standardizing it,
maybe you're right. I recall that the last attempt to produce a standard
library drifted because there wasn't a collective vision of what it should
be. Everybody had their own ideas of what it should include and how it
should look and it just never got to the point where enough of it gelled to
start moving towards an end product. (There were three of us who had the
notion that a statistics package would be a good place to start & I produced
one {with help!} that was tossed into the arena as a strawman, but it just
never grabbed enough interest to really get rolling. Maybe it is difficult
to do this sort of thing by committee and it should be left to a very small
handful of individuals with an end product coming out of the process that
can sink or swim based on how well it addresses everyone's needs?)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

I'm not sure what you are suggesting here. If what you mean is that your
company would pay to have some product developed (cost plus) and that it
would Open Source (some flavor of GPL?) the end result, I'd have no problem
with that. That's a work made for hire. Take the end result and do whatever
you like with it.

The rest is a matter of what and for how much.

If you are suggesting that perhaps a consortium of companies could/would
fund the development of some software product that they all wanted and would
pay salaries to the programmers and would eventually Open Source the end
result, I'd again have no problem with it.

It comes down to the Golden Rule. He who has the Gold makes the Rules. :-)

If you're suggesting that someone build you what you want and you'll give
them some modicum of cash and they keep the end result with the hope that
they can find some additional buyers, and one of the conditions is that the
source be provided to you with minimal restriction? I certainly don't
object. But it again comes down to what and how much? If the "what" is a
million SLOC of specialized code suitable only to your particular company
and the "how much" is $0.50, I don't think you'll find many takers.

My only objection to the Open Source thing is the notion that I as a
software developer should do all this work and give it away with no thought
as to remuneration. (That somehow it is "dirty" or "impure" to expect to
earn a living from making {Sacred Music, please} *Software!*) This is
especially onerous in my mind if someone else is going to take the software
I write and sell it in some way and never give me a nickel. I'm not opposed
to Open Source - in some cases that may make very good sense. (BTW,
depending on what you mean by "Open Source" - something done under the ADCL
might just qualify. You get the source. You don't have to pay for it. You
can use it and modify it freely. You don't have to ADCL your code by using
ADCL code. You just can't sell it without giving the original author
something in return. Maybe this isn't "Open Source(tm)" - but it certainly
is visable, usable, modifyable source at no cost - until you decide to
resell it.)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:9krak1$3f4$1@a1-hrz.uni-duisburg.de...
> And, now speaking for my company, we have actually once made
> a somewhat more substantial offer for the development
> of open code that we wouldn't own, and everyone could use.
> Could that be an option for you, in principal?
>
>

Re: More Uniform Ada libraries (was: Proving Correctness)

[Marin David Condic]

If I understand you correctly, you are basically saying "I think having some
big collection of utility code tacked on to a language compiler is important
for my development business. I further think it is important that this
library of utility code be (at least *mostly*) standard so that I can move
from one vendor to another and not lose my investment in this library. If
some substantial portion of Ada vendors do *not* provide some collection of
utilities and some substantial portion of C++ venders *do* provide some
collection of utilities, I am more apt to select C++ for my next development
effort."

If that is close to reflecting your opinion, I would certainly understand
the sentiment. How to standardize a set of utilities and where they should
come from and so on isn't necessarily the issue. If a language provides more
utility code with (most) distributions, then I gain a lot of leverage by
choosing that language. If that is what other languages are providing and it
is gaining them adherents, perhaps Ada (as a collective effort) needs to
consider how to do the same.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/



"nicolas" <n.brunot@cadwin.com> wrote in message
news:DMac7.86$E96.95326@nnrp4.proxad.net...
>
> The business model is not my problem, and I don't have feelings about it.
> The only feelings I have is that there are some customer side
requirements,
> coming from the reasons why we, and most Ada projects I know, made the Ada
> choice about 12 years ago.
> Our projects, and most projects I know, have nothing to do with embedded
> systems and 'DOD similar' projects.
> I don't believe that satisfying (for me) Ada solutions will be available
in
> the future, if those requirements are not fulfilled.
> We have to know if we will be the last Ada users to have concerns, which
are
> not embedded systems and 'DOD similar' projects concerns.
> If this is the case, we don't feel like fighting with Ada solution
providers
> which don't want to understand anything about what we need, and
deliberately
> ignore any other market's rules.
>
>
>
>

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

There are always unusual situations, so maybe I can't say "always" or
"never", but I doubt you will find many examples where someone pays
cost-plus and doesn't retain ownership - or some other highly exclusive
arrangement at least. I know of one situation where Company X paid Company Y
to develop a product and Company Y retained ownership with Company X having
few rights. (Names changed to protect the stupid! :-) However, I don't think
Company Y has yet found a customer for the product outside of Company X, so
ownership may be a moot point. (Besides, Company X was known to be
brain-dead)

Your second case is presuming that I have some specialized knowledge or
capability that you lack and hence I can charge you whatever I like and
you'll just accept it because you have no choice. Those situations would
again be extremely rare.

Why would Company X pay Company Y the full cost of development *plus profit*
to build them something unless they are going to in the end own it? Why not
just hire the staff to do the job, keep the product to yourself and retain
the profit margin that Company Y would make? I can imagine why I'd hire
Company Y - maybe I don't have the skills in house - and I could imagine
cases where it is in my interest to let Company Y have some rights to the
product, but why would I give up something I don't have to and get nothing
in exchange?

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Leif Roar Moldskred" <lrm@huldreheim.provida.no> wrote in message
news:WZbc7.5072$e%4.154183@news3.oke.nextra.no...
>
> Actually, there might be good business-sense in doing just that in a
> few cases. If the library is outside of the paying company's focus, it
> might be better for the company to leave the ownership of the library
> in the hands of the creators, and let them continue to develop and
> support it; rather than demanding the rights for themselves, and then
> never having the time or inclination to do anything more with the
> library.
>
> A second reason is even simpler: if nobody will write the library
> unless you pay them the development cost, and lets them retain the
> rights. For more specialized problem domains, it might be seller's
> market.
>
> --
> Leif Roar Moldskred

Re: More Uniform Ada libraries (was: Proving Correctness)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3149 bytes --]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> a �crit dans
le message news: 9krkfa$g12$1@nh.pace.co.uk...
> If I understand you correctly, you are basically saying "I think having
some
> big collection of utility code tacked on to a language compiler is
important
> for my development business. I further think it is important that this
> library of utility code be (at least *mostly*) standard so that I can move
> from one vendor to another and not lose my investment in this library. If
> some substantial portion of Ada vendors do *not* provide some collection
of
> utilities and some substantial portion of C++ venders *do* provide some
> collection of utilities, I am more apt to select C++ for my next
development
> effort."
>
> If that is close to reflecting your opinion, I would certainly understand
> the sentiment. How to standardize a set of utilities and where they should
> come from and so on isn't necessarily the issue. If a language provides
more
> utility code with (most) distributions, then I gain a lot of leverage by
> choosing that language. If that is what other languages are providing and
it
> is gaining them adherents, perhaps Ada (as a collective effort) needs to
> consider how to do the same.
>

Shortly said :
- We need Ada compilers suitable for us
- We will have them only if the number of Ada users with our concerns is
high enough.
- I don't think this number will be high enough with the current situation
concerning portable and standard Ada libraries.

More precisely, I think you got it right for companies starting today with
Ada language on our market.
This is slightly different for our specific situation.
We are used to develop our own Ada libraries
At the very beginning we worked on graphical applications for DOS platform
and almost no Ada library was available for our needs.
Now, when we need external libraries, we usually buy C commercial libraries
and import them in Ada.
The reason is that the libraries we need simply don't exist in Ada, while
excellent ones are available in C.

As far as we are concerned, we don't really need the kind of Ada standard
libraries we are talking about.
Because we know how to develop them, or import them from other languages.
But this has a cost, and anyway we are still dependant of the availability
of Ada compilers suitable for what we do.

The question is to know if compilers suitable for us will be available in
the future.
Nowadays, I don't believe that software companies will use Ada to develop
the kind of software we develop, if Ada standard libraries are not
available.
Newcomers to Ada won't accept what we accept, because having no Ada
background, it will be easier and cheaper for them to develop in C, C++ or
Java, than to be tied to either ACT or Aonix or Rational, or spend their
time writing endless pragmas import.
More than that, when you want to use subcontractors, you'd much better be
looking for subcontrators with C, C++ or Java knowledge, than Ada knowledge.

If in the end, we are too few companies with the kind of requirements we
have, no Ada compiler will be suitable for us.
That's our main concern.

Re: Ada Components (was Java Portability)

[Brian Rogoff]

On Tue, 7 Aug 2001, Lao Xiao Hai wrote:
> When Ivan Stepanov decided to build the STL, 

You've already been corrected on the name, it is Alexander. 

> he began with Ada 83 but
> found the template model (generics) inadequate, and the inability to do
> extensible inheritance to be a liability.

Actually, he and David Musser started with Scheme. They later produced the 
Ada version. I don't believe inheritance was an issue; if anything
Stepanov believes that OO is OOverhyped. 

> He switched to C++ and that is now where the STL has its most mature
> representation. 

While he was at ATT with Stroustrup is when templates were being designed. 
Many of his library desires fed right in to the C++ template design. 

> Stepanov has
> been variously supported financially by AT&T, HP and Silicon Graphics,
> and perhaps by others.    No one is supporting anyone, at present, to build
> an equivalent set of libraries for Ada, even though the 1995 Ada standard
> addressed and surpassed the concerns originally raised by Stepanov.

That's both true and false. Ada 95 is great in some ways, lousy in others, 
from the standpoint of building C++s STL. C++ has a crude form of
parametric polymorphism, and Ada generics, which must be explicitly
instantiated and named, are just *not* the same thing as parametric
polymorphism. Given the differences, I think copying the STL too closely
will just make Ada look bad. 

Ada does have a half decent approach to signatures now (which looks 
really primitive and clunky to any ML user) which is nice when building an
STL because you get compile time checking. 

There's some work on "generic programing" in the FP community, which has a
more mathematical flavor, and has some of the same goals as the C++ work, 
but is almost entirely different, and is, like most work from the FP 
community (and unlike C++ :), beautiful. 
 
As far as the rest goes: why do you think anyone working on an Ada project
would pay for an Ada STL? They'd either use free Ada libraries or switch
to C++ and not pay for the C++ STL. 

-- Brian

Re: Proving Correctness (was Java Portability)

[tmoran]

>someone's compiler delivery. All they have to do is believe that it is to
>their advantage to include some .ZIP file or directory on the CD they ship
  As, I believe, Rational does with CLAW on their Windows compiler.

Re: Proving Correctness (was Java Portability)

[tmoran]

> I know of one situation where Company X paid Company Y
>to develop a product and Company Y retained ownership with Company X having
>few rights. (Names changed to protect the stupid! :-) However, I don't think
>Company Y has yet found a customer for the product outside of Company X,
  And then there's the case where X was IBM, Y was Microsoft, and the
product was MSDOS (sometimes, but less often, known as PCDOS).

Re: More Uniform Ada libraries (was: Proving Correctness)

[tmoran]

>The reason is that the libraries we need simply don't exist in Ada, while
>excellent ones are available in C.
  On the Windows platform, at least, Claw is an example of an extensive
library that does exist in Ada, and Gnatcom is a tool to make it easy
to use things available in C (well, in DLL binary).  So it's not as if
there's a complete wasteland for the Ada programmer.

Re: More Uniform Ada libraries (was: Proving Correctness)

[Marin David Condic]

Very true and I often cite Claw as an example of what is available in the
way of development tools for Ada. I think it would be an excelent and major
part of a development kit for Windows. With a compiler, a debugger, an
editor, maybe some library management and configuration management and an
overall "How To" book, you'd have an excellent kit if all the pieces were
smoothly integrated.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


<tmoran@acm.org> wrote in message
news:y9fc7.42123$Kd7.25933469@news1.rdc1.sfba.home.com...
> >The reason is that the libraries we need simply don't exist in Ada, while
> >excellent ones are available in C.
>   On the Windows platform, at least, Claw is an example of an extensive
> library that does exist in Ada, and Gnatcom is a tool to make it easy
> to use things available in C (well, in DLL binary).  So it's not as if
> there's a complete wasteland for the Ada programmer.

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:
: I'm not sure what you are suggesting here.

Yes, something like the first two cases.
It pays programmers as long as they are developing
a much needed library. If their is sufficient demand
for a GMGPL XML library, say, some skilled Ada programmer
could write one. If this person is know to those who are
interested, they might be ready to "sponsor" the development
of that library.

Maybe then someone could join Simon Wright in developing
the 95 Booch components, wouldn't there be quite some
opportunities, to make good use of them?

The basic issue for me is then, how can we make more
people understand that the development of _any_ software
costs time and money, be that Free Software, or Don't-Disassemble-Ware.
Some already do, but I won't guess how many.

: company would pay to have some product developed (cost plus) and that it
: would Open Source (some flavor of GPL?) the end result, I'd have no problem
: with that. That's a work made for hire.

: If you are suggesting that perhaps a consortium of companies could/would
: fund the development of some software product that they all wanted and would
: pay salaries to the programmers and would eventually Open Source the end
: result, I'd again have no problem with it.
: 
: It comes down to the Golden Rule. He who has the Gold makes the Rules. :-)

Yes, within the restrictions that rule-makers have to live
with :-)

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Well, if you mean that, for example, you "sponsor" me to write a square root
routine and put it out under some flavor of the GPL on the net where you can
download it and use it and your "sponsorship" consists of writing me a check
for $50.00 US, then obviously, I have received some remuneration for doing
this and (under GPL) this is about all I could reasonably expect. Possibly,
I could make some additional money from providing customer driven updates to
the square root routine and maybe get some fees for lecturing about the
design of the square root routine and that would have to figure into my
decision to undertake the work.

If someone makes a deal like this, more power to them! They are getting some
kind of remuneration for their work rather than just simply giving it away
and hence they gain in doing it and have incentive to continue to follow up
on it. Its not at all a bad thing. However, I'm not aware of very many of
these sorts of arrangements being made. Obviously, the willingness of people
to develop under these terms is going to depend on how large the check is
that you send them. The more money, the more likely you'll find willing
developers.

As to the equity of this situation - that's for us both to determine in the
process. Is $50 to me worth the time I'll spend developing the square root
code? No? Can I see enough back-end business coming out of it that I might
still develop it? This is essentially a work made for hire and I've got no
problems with whatever license terms you want to put on it - except insofar
as it may influence the amount of money I expect to do the job.

But if Ada's success is going to depend on lots of people developing
software with no thought as to financial gain, then I doubt it will see much
success. That's why I favor something like the ADCL as a means of holding
out *some* financial incentive that might encourage people to develop
software on speculation that they might get some back-end reward. If someone
doesn't like the ADCL, I'd like to hear an alternative that promotes wide
circulation and use of Ada source and holds out some financial reward for
the guy who develops it. We might find lots of entrepeneurs in this
newsgroup alone who would undertake significant projects if they thought
they might get some financial reward out of it.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:9ks3e9$69v$1@a1-hrz.uni-duisburg.de...
> Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:
> : I'm not sure what you are suggesting here.
>
> Yes, something like the first two cases.
> It pays programmers as long as they are developing
> a much needed library. If their is sufficient demand
> for a GMGPL XML library, say, some skilled Ada programmer
> could write one. If this person is know to those who are
> interested, they might be ready to "sponsor" the development
> of that library.
>
> Maybe then someone could join Simon Wright in developing
> the 95 Booch components, wouldn't there be quite some
> opportunities, to make good use of them?
>
> The basic issue for me is then, how can we make more
> people understand that the development of _any_ software
> costs time and money, be that Free Software, or Don't-Disassemble-Ware.
> Some already do, but I won't guess how many.
>

Re: More Uniform Ada libraries (was: Proving Correctness)

[Stephen Leake]

"nicolas" <n.brunot@cadwin.com> writes:

> Shortly said :
> - We need Ada compilers suitable for us

Which you appear to have today. You have given no reason to suggest
that might change in the future.

> - We will have them only if the number of Ada users with our
> concerns is high enough. 

Apparently it is today; why will it change for the worse? The trend is
toward more Ada use, not less.

> - I don't think this number will be high enough with the current
> situation concerning portable and standard Ada libraries.
> 
> <snip> We are used to develop our own Ada
> libraries At the very beginning we worked on graphical applications
> for DOS platform and almost no Ada library was available for our
> needs. Now, when we need external libraries, we usually buy C
> commercial libraries and import them in Ada. The reason is that the
> libraries we need simply don't exist in Ada, while excellent ones
> are available in C.

Great! I'll assume they will continue to be avialable, independent of
the Ada market.

> As far as we are concerned, we don't really need the kind of Ada
> standard libraries we are talking about. 

Then why are we talking about them!?

> Because we know how to develop them, or import them from other
> languages. But this has a cost, and anyway we are still dependant of
> the availability of Ada compilers suitable for what we do.
> 
> The question is to know if compilers suitable for us will be available in
> the future.
> Nowadays, I don't believe that software companies will use Ada to develop
> the kind of software we develop, if Ada standard libraries are not
> available.
> Newcomers to Ada won't accept what we accept, because having no Ada
> background, it will be easier and cheaper for them to develop in C, C++ or
> Java, than to be tied to either ACT or Aonix or Rational, or spend their
> time writing endless pragmas import.

Being able to take advantage of all the software engineering features
of Ada is _well worth_  writing pragmas! That is the point! I guess
you feel "Newcomers to Ada" won't agree with that assesment.

> More than that, when you want to use subcontractors, you'd much
> better be looking for subcontrators with C, C++ or Java knowledge,
> than Ada knowledge.

Ok, (assuming there are fewer Ada contractors available) but that is a
totally separate argument from "we need standard Ada libraries".

> If in the end, we are too few companies with the kind of
> requirements we have, no Ada compiler will be suitable for us.
> That's our main concern.

I don't see how you reach this conclusion from the above statements;
you've said nothing about why the Ada compilers you actually use, or
the features they provide, might disappear.

I also don't see how this relates to "standard Ada libraries". You are
_not_ using Ada libraries; you are using the _standard_ mechanisms in
Ada to import available C libraries. Sounds like a good situation to
me!

-- 
-- Stephe

Re: Proving Correctness (was Java Portability)

[David Starner]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9krils$fcb$1@nh.pace.co.uk...
> What FSF decides to do is their own business. What ACT decides to do is
> *their* own business. Right now ACT includes stuff (like AdaGIDE) in the
> public distribution that aren't part of the "formal" release of GNAT. If
you
> were to write some useful library of stuff and ACT put it out on the disk
> right next to GNAT how does that affect FSF?

Okay, but we were talking about stuff that travels with the compiler. During
my periods of Internet via modem and sneaker, I wouldn't download those
files unless I had to have them to run GNAT, or I later found I needed
them - and later I'd be as likely to run across some other package on a web
search. If it gets tossed on a shovelware CD, I'm likely to overlook them
completely.

> Hobbies are fine. However, I'd believe there is a better likelihood that
> products will get brought to market at a high quality if someone is doing
it
> for something a little more tangible than the satisfaction someone gets
from
> their hobby.

I'm not sure that's true. A lot of the good Open Source products compete not
necessarily on bells and whistles, but on quality and reliability - see NT
versus Linux, or IIS versus Apache.

> > But there are a number of people who have developed Ada libraries for
> > free.
>
> Never said there weren't. As a matter of fact, I've done that myself.
(O.K.
> I didn't put it out there under GPL, but I left a really strong hint about
> just how easy it would be to get permission from me to use the libraries.)

There's something to be said for no-hassle licensing, whether commerical or
free. I'd hesitate to use a library, or spend the time discussing using a
library unless it was unique, if I thought I might get a vague "go ahead"
and latter "that's not what I meant by go ahead."

> Probably they
> won't go for that, but if there were *some* way that the "volunteers" get
> something out of it (besides the warm fuzzy feeling) I bet the odds of
> getting it done and done well would go up. (Remember, there is a kind of
> business benefit to be had by some potential participants - so it doesn't
> necessarily have to be some financial transaction that creates the
> incentive.)

Okay. But I don't see the ADCL doing that. It's just too much of a crap
shoot making money of off it. Also, a lot of potential participants gain
from it being open source (me, ACT).

> All of those "what if" questions you ask are dealt with to some extent in
> the articles by Dr. Leif.

Do you have a link?

> To the extent that they aren't dealt with - I
> agree that they need to be dealt with. The ADCL is not a finished
product -
> it is a concept that is being kicked around and may ultimately emerge in
> some form different from what I've seen so far. The fact that there are
> questions and "what if" scenarios doesn't make it A Bad Thing. Go look
back
> at all the questions and "what ifs" that are discussed here and elsewhere
> concerning the GPL.

But there's a difference. The GPL very clearly stops you from distributing
copies without providing source in some way. The "what if"s - sure, you
might be able to evade the GPL if crosswire the COM ports and inject C#
particles, but there are very few parties who have tried, and if someone
gets away with it, you don't lose much. Money, on the other hand, has this
tendency to end up in the wrong pocket and get arguments over whose pocket
it belongs in. If it's just you, than great, but most programs are a lot of
work for a single programmer. If you're willing to form a buisness with your
fellow contributers, you've got the money issues cleared up some, but then
why the ADCL? If you've got 5 major contributers for places around the
world, with 12 minor contributers and a couple dozen bug fixers (like a
mid-sized Open Source project), you've got a problem you really need an
accountant to handle (but accountants get expensive.)

> (There were three of us who had the
> notion that a statistics package would be a good place to start

I'm curious - why? I'd start with the stuff that several others have put
into their standard libraries - look at the C++ standard libraries, the Java
standard libraries, what's ended up in libc, and take what's common that Ada
doesn't do well. I'd also look at the Ada libraries already written, and the
common points from them. I don't think a statistics package would come up #1
in that list.

Re: Proving Correctness (was Java Portability)

[Ole-Hjalmar Kristensen]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> writes:

> Wellllll....... You *might* pay to get a custom library built rather than
> shop for an off-the-shelf part. Sometimes you have a problem domain for
> which no suitable library exists or there are licensing issues, etc. But the
> one thing you can count on here is that any (non-brain-dead) businessman is
> *not* going to pay "Ada Library Writers Inc." the full development cost,
> plus profit and not walk away with 100% ownership of the end product.
> 
> MDC
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/
> 

Yes, I agree completely. The problem with such libraries is that they
usually not are released on the open market, although there is of
course nothing preventing the customer to do so if it can make some
extra business.

> 
> "Ole-Hjalmar Kristensen" <ohk@clustra.com> wrote in message
> news:umqr8umyhp0.fsf@maestro.clustra.com...
> > Larry Elmore <ljelmore@home.com> writes:
> >
> > But he's not an investor, he's potential customer in the mass
> > market. There is a difference you know, and compilers and libraries
> > are commodities these days, not custom-made.
> >
> 
> 
> 

-- 
Kabelsalat ist gesund.

Ole-Hj. Kristensen

Re: More Uniform Ada libraries (was: Proving Correctness)

[nicolas]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2928 bytes --]

"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> a �crit dans le message
news: u66byfd4u.fsf@gsfc.nasa.gov...
> "nicolas" <n.brunot@cadwin.com> writes:
>
> > Shortly said :
> > - We need Ada compilers suitable for us
>
> Which you appear to have today. You have given no reason to suggest
> that might change in the future.

Alsys Ada83 is excellent, but is now somewhat out of date (about 6 years
old), will it run for WinXP ?
We never had any need of support in 6 years, but it's no longer supported
nor evolving,
It will have to be replaced someday, at least to take full advantage of
Ada95 features.

Aonix Objectada is excellent for us, except that link time is far too long
for our applications and cannot be used today for dayly development
When the link time is 12mns instead of 20s, and has to be done 10 or 20
times a day, it is a problem.
Will Aonix go on with Ada ? some have doubts

Rational Apex is not bad, but still have some little problems
Will they go on with Ada ? some have doubts too

Gnat (professional or public version) is too far from being suitable for us

So compiler availibility could become critical sooner than you think

> > - We will have them only if the number of Ada users with our
> > concerns is high enough.
>
> Apparently it is today; why will it change for the worse? The trend is
> toward more Ada use, not less.

From we hear on comp.lang.ada the number of people having our concerns is
everything but high enough ...
The trends towards Ada use depends of who you listen and what kind of
application you develop.

> > <snip> We are used to develop our own Ada
> > libraries At the very beginning we worked on graphical applications
> > for DOS platform and almost no Ada library was available for our
> > needs. Now, when we need external libraries, we usually buy C
> > commercial libraries and import them in Ada. The reason is that the
> > libraries we need simply don't exist in Ada, while excellent ones
> > are available in C.
>
> Great! I'll assume they will continue to be avialable, independent of
> the Ada market.

They have been for years and continuously improving.
We have absolutely no reason to worry about that.
If only we had the same situation for Ada compilers, we would have nothing
to worry about.

> > As far as we are concerned, we don't really need the kind of Ada
> > standard libraries we are talking about.
>
> Then why are we talking about them!?

Have you read my post ?

> I don't see how you reach this conclusion from the above statements;
> you've said nothing about why the Ada compilers you actually use, or
> the features they provide, might disappear.

Now that's done :-)

> I also don't see how this relates to "standard Ada libraries". You are
> _not_ using Ada libraries; you are using the _standard_ mechanisms in
> Ada to import available C libraries. Sounds like a good situation to
> me!

Have you really read the post entirely ?

Re: Proving Correctness (was Java Portability)

[Leif Roar Moldskred]

Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:
> There are always unusual situations, so maybe I can't say "always" or
> "never", but I doubt you will find many examples where someone pays
> cost-plus and doesn't retain ownership - or some other highly exclusive
> arrangement at least. 

Well, that's why I said "in a few cases" :-) I just wanted to point
out that it make be good business sense to do this, sometimes.

[SNIP]

> Your second case is presuming that I have some specialized knowledge or
> capability that you lack and hence I can charge you whatever I like and
> you'll just accept it because you have no choice. Those situations would
> again be extremely rare.

Well, I don't think it would have to be quite that extreme - as long
as the rights to the software is worth little to you, but much to the
company that writes it, and the writing company is not in a
competition-fierce niche; this way might be advantageous.

To take a fictional example - Cool Cats, a graphical design shop does
a lot of Photoshop work. They've estimated that a particular,
non-existent add-on tool to Photoshop might save them $100,000 a
year. So they call up PhotoPlugins inc., a company that makes and
sells a popular set of Photoshop plug-ins.

PhotpPlugins aren't too keen on the assignment, as they're not sure if
there's really much of a market for this particular plug-in, so they
say "Yes, we can make that - but you will have to cover the
development costs - $50,000 - rather than the price of $15,000 we
would normally charge for such a plug-in."

To Cool Cats it makes sense to accept this deal - they'll get the
product, and save $50,000.

> Why would Company X pay Company Y the full cost of development *plus profit*
> to build them something unless they are going to in the end own it? Why not
> just hire the staff to do the job, keep the product to yourself and retain
> the profit margin that Company Y would make? 

Well, for the same reasons that you would out-source in other ways - it
might be cheaper, faster or otherwise better to pay someone else to do
it, than to do it yourself. You might not have the same profit margin
on this product as Company Y does - and Company Y already has the
infrastructure in place to make it, so they don't have the additional
costs and time to build it from scratch.

And if this difference between your cost of making this, and Company
Y's cost of making this, is greater than the value of the software
rights _to you_; it pays to do it this way. Plus profit is probably
stretching it, but that depends on the profit margins of Company Y.

> I can imagine why I'd hire
> Company Y - maybe I don't have the skills in house - and I could imagine
> cases where it is in my interest to let Company Y have some rights to the
> product, but why would I give up something I don't have to and get nothing
> in exchange?

But you _do_ get something in return - the finished product, which
would otherwise not exist; and that's what you're really interested
in, after all. The value of the software-rights might be worth next to
nothing to the buying company - they might not be in the business of
selling software.

Also, if the development goes over budget or falls flat on the face in
some other way, _Company Y bear the costs_. You pay for the finished
product - you are not hiring consultants to do the work for you; so you
don't pay for the hours - only the final result.

And you don't have to pay in advance either, although for large
project there would normally be some sort of incremental
payment.


-- 
Leif Roar Moldskred

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:

: doesn't like the ADCL, I'd like to hear an alternative that promotes wide
: circulation and use of Ada source and holds out some financial reward for
: the guy who develops it. We might find lots of entrepeneurs in this
: newsgroup alone who would undertake significant projects if they thought
: they might get some financial reward out of it.

This morning, having tee, a button crossed my mind, to be
found on adapower, above it some text asking visitors to say
whether they want to contribute money to some library project
described there :-) Somwhat like IBM did with some Java
product, I don't remember exactly which one, asking people to
confirm willingness bo actually buy.

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

O.K. Let me deal with just this one point.

I need some plug-in for Photoshop that will save me a bunch of money and
plug-ins to photoshop are not my core business (Software in general isn't.
Let's say I'm a graphic arts company) It makes sense for me to outsource the
development of this plug-in because I don't have the expertise in house and
don't want to acquire it either with regular employees or consultants to
whom I've got to provide infrastructure. So your company has expertise in
developing plug-ins for photoshop and you don't really think there is a
market beyond me. You say you'll develop it if I pay cost-plus. I say O.K.,
I'll pay cost-plus, but at the end of the day there is this work product
which I have totally financed. What are we going to do about ownership?

I'm not a software vendor, so I don't want to try to sell it myself.
However, I've paid for it much like I've paid for my plant and equipment and
so on - it is an investment in assets. You propose that *you* retain
ownership, perhaps to sell it elsewhere and perhaps to create a bigger user
base and perhaps I gain some additional benefits from a bigger user base.
However, the whole reason I paid cost-plus was because you didn't think
there was much of a market out there. At this point, it gets down to an
interesting negotiation. I've invested in an asset and I don't want to just
give that away to someone else if it has earning potential. You've got the
infrastructure to actually market it and perhaps generate those earnings. We
both have something the other guy wants. Sooooo......

I propose that we have some sort of joint ownership. I finance the
development because I have the capital. You go out and sell it because you
have the marketing infrastructure. As the money rolls in, we split it in
some manner. Or you give me revenue until it pays for the development and
then keep the rest. Or we agree that if you find another buyer, you'll cut
my price in half. Or any other number of possible scenarios wherein I don't
simply make a gift to you of one of my assets.

You say you don't like the deal and you want me to pay the full cost and you
get to retain ownership? I say "fine!" and go hire myself some geeks to do
it for me and keep it all for myself - forcing my competitors to spend their
own money to develop the tool for themselves if they want to keep up with my
competitive advantage.

If the plug-in doesn't either reduce costs or increase revenue in some way,
then why did I want to buy it in the first place? If it *does* contribute to
the bottom line, then I don't want my competitors to get it free of charge
or at a discount while I had to pay full-boat for the asset. I might agree
to let my competitors buy it because I calculate that the revenue from
selling it will exceed whatever loss I have in competitive advantage. But I
don't *give* it away. See my point?

I guess I just don't see many circumstances where a non-brain-dead
businessman is going to give up an asset to another business venture without
negotiating something in return. I certainly know *I* wouldn't give it up if
I could make some bucks for it. That's why I'm in business - not for my
health. :-)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Leif Roar Moldskred" <lrm@huldreheim.provida.no> wrote in message
news:1nvc7.5408$e%4.165724@news3.oke.nextra.no...
>
> But you _do_ get something in return - the finished product, which
> would otherwise not exist; and that's what you're really interested
> in, after all. The value of the software-rights might be worth next to
> nothing to the buying company - they might not be in the business of
> selling software.
>
> Also, if the development goes over budget or falls flat on the face in
> some other way, _Company Y bear the costs_. You pay for the finished
> product - you are not hiring consultants to do the work for you; so you
> don't pay for the hours - only the final result.
>
> And you don't have to pay in advance either, although for large
> project there would normally be some sort of incremental
> payment.
>

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

That is at least one alternative. It kind of seems a little like begging.
"Please won't you donate to our out-of-work Ada programmers fund so we can
keep these poor kids off the streets and out of delinquency..." (Midnight
Programming, anybody?)

But as a kind of market research tool, it might be useful. If there were
some requirements spec sitting there that potential buyers could look over
and fill in a market response card (Would you buy this product for $1000 if
it were available today?) It would give you some idea as to where you could
hawk your wares and for how much if/when you completed the project.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:9ktvqi$cbb$1@a1-hrz.uni-duisburg.de...
>
> This morning, having tee, a button crossed my mind, to be
> found on adapower, above it some text asking visitors to say
> whether they want to contribute money to some library project
> described there :-) Somwhat like IBM did with some Java
> product, I don't remember exactly which one, asking people to
> confirm willingness bo actually buy.

Re: Proving Correctness (was Java Portability)

[Stefan Skoglund]

Pascal Obry wrote:
> Maybe all Ada vendors should put some money on some sort of organization to
> have the Ada libraries updated using standard Ada without vendor-specific
> additions. This would benefit the whole Ada community. Does such an
> organization exist ?

I don't think an XML library would be that useful in an compiler
targetting embedded applications.

Ow, nice we can send midcourse updates to the navigation
computer as an XML stream. Really nice :-)

ARGGGGHHHHHHH

Having a lot of libraries added to the Ada standard shoul be done
as annexes which means that not every compiler needs to have them.

After all we still have to think about what a change to standards
means with regards to compiler validation.

Claw could be added as a annex so could GtkAda if we didn't have that
small problem called creeping featuritis (in gtk+ in this case)

Re: Proving Correctness (was Java Portability)

[David Starner]

[As noted, this was written by Martin Cordic and sent to me for the reasons
below.]

I couldn't get my newsreader to post a reply to the particular message on
the newsgroup. Here's my reply - if you feel like posting it to C.L.A. go
ahead.

MDC

"David Starner" <dstarner98@aasaa.ofe.org> wrote in message
news:tn48ce7fdodvbb@corp.supernews.com...
>
> But there's a difference. The GPL very clearly stops you from distributing
> copies without providing source in some way. The "what if"s - sure, you
> might be able to evade the GPL if crosswire the COM ports and inject C#
> particles, but there are very few parties who have tried, and if someone
> gets away with it, you don't lose much. Money, on the other hand, has this
> tendency to end up in the wrong pocket and get arguments over whose pocket
> it belongs in. If it's just you, than great, but most programs are a lot
of
> work for a single programmer. If you're willing to form a buisness with
your
> fellow contributers, you've got the money issues cleared up some, but then
> why the ADCL? If you've got 5 major contributers for places around the
> world, with 12 minor contributers and a couple dozen bug fixers (like a
> mid-sized Open Source project), you've got a problem you really need an
> accountant to handle (but accountants get expensive.)
>

Again, I'm not a lawyer nor do I play one on newsgroups. However, my reading
of Dr. Leif's articles & the license seem to indicate to me that if you used
ADCL source and *gave* the code to one of your customers, you'd have to
*give* them the ADCL source as well. You just wouldn't be forced to give
them anything that *you* own - unless you felt like it. You would also be
required to give them the ADCL source if you *sold* the product - but
wouldn't be required to give them your own part unless you wanted to. In
that sense, I think the ADCL is less restrictive than the GPL.

Here's another dirty little secret about software licenses: People ignore
them *all the time* and use/sell software for which they don't have
permission. They steal GPL'ed code, include it in their products, ship it
off to their customers without including source, etc. etc. etc. The license
doesn't create a garrison of stormtroopers to whup-up on the guys that do
this. More or less, its an "honor system". So if both of us put code out
there - me under ADCL, you under GPL - we can both bet our rent checks that
*someone* is going to use it in violation of the license.

In both cases you only want to look for the large, high profile cases. It
isn't worth going after the small timers because it costs too much or
becomes too complicated to try to stop them. The trick here is to create a
scenario where some big concern with some large scale usage is going to look
at it and say: "Yeah, we could cheat, but it isn't worth the liability. If
we really want to use that GPLed code, we'll GPL our end product as well. If
we really want to use that ADCL code, we'll pay the license fee and avoid
the lawsuit." In both cases, the minute the terms and conditions of the
license become too onerous, the integrator goes "I'm better off building the
whole thing in-house". There are projects that don't go with GPLed code for
that reason. There would be projects that would avoid ADCLed code for that
reason. I think the key to success would be to make it known that the ADCL
applies *unless* you cut a separate deal with the author(s). Then its up to
the author(s) to decide if its worth it to go with an alternate license for
this one particular customer.

As for the complexity of collecting fees, etc? I agree. That is a problem.
Some level of infrastructure would have to exist that made this as painless
as possible. But remember that there are existing models for doing this such
as ASCAP in the recording industry. I'd like to see the mechanisms for
determining royalty percentages, etc. and the mechanisms for
collecting/disbursing fees. I think there are a lot of questions that need
to be answered and resolved before the ADCL is something practical. I just
think that the *concept* is worth investigating and seeing if something
could be made to work.


> > (There were three of us who had the
> > notion that a statistics package would be a good place to start
>
> I'm curious - why? I'd start with the stuff that several others have put
> into their standard libraries - look at the C++ standard libraries, the
Java
> standard libraries, what's ended up in libc, and take what's common that
Ada
> doesn't do well. I'd also look at the Ada libraries already written, and
the
> common points from them. I don't think a statistics package would come up
#1
> in that list.
>
Lots of reasons:

1) The three of us that worked on this all had an interest in statistics and
could have made use of the statistics code that resulted.

2) I rather felt that the whole group was spending lots of time arguing over
the data structure-ish stuff and didn't want to get bogged down in that. My
personal reaction was to think: "O.K. I'll go take on this niche that the
rest of the group doesn't have as much interest in. We will come up with
something that we agree looks like "A Good Start" The rest of the group will
be presented with it and its a done-deal. No point arguing over it - there
it is and it works. Why not accept it?

3) A lot of the library stuff you'll see with other languages (C/C++ being
big examples) are either to interface to OS features (not suitable for a
cross-platform Ada thing) or to make up for the crippling deficiencies of
the other languages (Hey? What do you say? Let's cross-post this to C/C++
newsgroups and start another flame war! :-) Ada didn't need de-crippling
features. Any library should seek to extend the usefulness of Ada and offer
things that didn't exist for other languages. Statistics seemed to be a
natural outgrowth from the Ada.Numerics stuff. (Maybe not as generally
useful as a square root routine, but why not have it?)

4) BTW: If you look at what sort of "standard libraries" come with C or C++
you'll discover that there really isn't much. What you get is some major
implementer providing a bunch of libraries that may or may not be common to
other implementations. So almost *anything* that Ada did to provide a
"Standard" set of libraries (at least something that all significant
implementations for general purpose computers supported) would be ahead of
the game. (Java might be a notable exception - but that starts the debate
about "Standard Java" not being anything but "Sun Java" - one implementer,
one library.)

I agree that stats wouldn't be the #1 item on the list - that ought to be
some tree of data structures code. Other possible branches would be commonly
used data types that Ada may support but doesn't necessarily provide all the
useful functionality - things like dates/times with different string
formatting, temperatures, various scientific units and conversions, etc. The
branch that I had an interest in (and thought would at least rank highly
with a large segment of programmers) would be mathematics. Why not support
statistics, probability, vector & matrix math, etc? All this stuff exists -
but there isn't a "common" interface. (Same problem arose with Ada83 - no
math functions under the argument that everyone can roll their own. The
complaint was lack of portability because each vendor supplied a slightly
different interface.)

Don't you think there might be a lot of people who might look at Ada and say
"You know, I've heard all sorts of bad things about this language and I
don't like (insert complaint here) about it, but gee! Look at all that
statistics code I get with the compiler and can just go off and use. It sure
buys me lots of leverage in my problem domain... and I don't get it
automatically with my favorite single-character-language++"
>


MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/

Re: Proving Correctness (was Java Portability)

[David Starner]

Martin Cordic wrote in message news:tn6225iggb0u78@corp.supernews.com...
> 1) The three of us that worked on this all had an interest in statistics
and
> could have made use of the statistics code that resulted.
>
> 2) I rather felt that the whole group was spending lots of time arguing
over
> the data structure-ish stuff and didn't want to get bogged down in that.
My
> personal reaction was to think: "O.K. I'll go take on this niche that the
> rest of the group doesn't have as much interest in. We will come up with
> something that we agree looks like "A Good Start" The rest of the group
will
> be presented with it and its a done-deal. No point arguing over it - there
> it is and it works. Why not accept it?

I understand. I started working on a Unicode library, Ngeadal, for similar
reasons. It's moribund now; I'm planning on officially calling it dead once
I get one piece of algorithimic skill done that may impress and be useful to
other Unicode hackers.

--
David Starner - dstarner98@aasaa.ofe.org
"The pig -- belongs -- to _all_ mankind!" - Invader Zim

Re: Proving Correctness (was Java Portability)

[Pascal Obry]

Ed Falis <efalis@mediaone.net> writes:

> I've been out of school for 20 years, and feel the same way as Starner. 
> So much for the wisdom of real life experience.  Now if someone wants to
> take over my alimony payments ...

I feel the same too. So it seems that this is not an isolated case or are we
all fool :)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--|
--| "The best way to travel is by means of imagination"

Re: Proving Correctness (was Java Portability)

[Simon Wright]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:

> Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:
> : I'm not sure what you are suggesting here.
> 
> Yes, something like the first two cases.
> It pays programmers as long as they are developing
> a much needed library. If their is sufficient demand
> for a GMGPL XML library, say, some skilled Ada programmer
> could write one. If this person is know to those who are
> interested, they might be ready to "sponsor" the development
> of that library.
> 
> Maybe then someone could join Simon Wright in developing
> the 95 Booch components, wouldn't there be quite some
> opportunities, to make good use of them?

I can't parse that, sorry ..


I would have a _lot_ of trouble convincing my employer that I should
accept payment from a third party for working on the BCs. (Mind you,
our commercial manager thinks I'm mad to work on them for free!
Personally I'd make that "sad")

Re: More Uniform Ada libraries (was: Proving Correctness)

[Stephen Leake]

"nicolas" <n.brunot@cadwin.com> writes:

> "Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> a �crit dans le message
> news: u66byfd4u.fsf@gsfc.nasa.gov...
> > "nicolas" <n.brunot@cadwin.com> writes:
> >
> > > Shortly said :
> > > - We need Ada compilers suitable for us
> >
> > Which you appear to have today. You have given no reason to suggest
> > that might change in the future.
> 
> Alsys Ada83 is excellent, but is now somewhat out of date (about 6 years
> old), will it run for WinXP ?
>
> We never had any need of support in 6 years, but it's no longer
> supported nor evolving, It will have to be replaced someday, at
> least to take full advantage of Ada95 features.

Ok. So you need a new Ada compiler.

> Aonix Objectada is excellent for us, except that link time is far
> too long for our applications and cannot be used today for dayly
> development When the link time is 12mns instead of 20s, and has to
> be done 10 or 20 times a day, it is a problem. Will Aonix go on with
> Ada ? some have doubts

How much money are you paying them to improve the situation?

> Rational Apex is not bad, but still have some little problems Will
> they go on with Ada ? some have doubts too

How much money are you paying them to improve the situation? Or, how much
feedback have you given them about why their product is unsuitable for
you? They won't fix it if they don't know it's broke.

> Gnat (professional or public version) is too far from being suitable
> for us

Have you done a formal evaluation process with ACT? They will give you
free support for a limited time, while you use GNAT for your system.
They will help you fix or workaround any problems. In my experience,
they offer far better support than any other Ada company.

> So compiler availibility could become critical sooner than you think

Ok. Good reason to be investing money in some compiler company.

> > > - We will have them only if the number of Ada users with our
> > > concerns is high enough.
> >
> > Apparently it is today; why will it change for the worse? The trend is
> > toward more Ada use, not less.
> 
> From we hear on comp.lang.ada the number of people having our concerns is
> everything but high enough ...

comp.lang.ada is hardly an authoritative source of information on how
many people use Ada, and what the trend is. The set of all Ada
compiler vendor's customer lists would be the real data.

> <snip>
> > > As far as we are concerned, we don't really need the kind of Ada
> > > standard libraries we are talking about.
> >
> > Then why are we talking about them!?
> 
> Have you read my post ?

Yes, and I have not understood your rationale. Which is why I am
trying to ask very specific questions; that tends to clarify things.

> > I don't see how you reach this conclusion from the above
> > statements; you've said nothing about why the Ada compilers you
> > actually use, or the features they provide, might disappear.
> 
> Now that's done :-)

Well, actually, no. All you have done is say you have doubts, and
quote vague experience from comp.lang.ada. Hardly convincing.

> > I also don't see how this relates to "standard Ada libraries". You
> > are _not_ using Ada libraries; you are using the _standard_
> > mechanisms in Ada to import available C libraries. Sounds like a
> > good situation to me!
> 
> Have you really read the post entirely ?

Yes. It would be more friendly if you actually answered the question,
but I'll cope.

-- 
-- Stephe

Re: Proving Correctness (was Java Portability)

[Larry Elmore]

Marin David Condic wrote:
> 
> "Larry Elmore" <ljelmore@home.com> wrote in message
> news:3B706ADC.B4847AC3@home.com...
> > nicolas wrote:
> >
> > > You don't pay to build the factory, the company invest a lot, and can
> count
> > > on an income only if the result is good.
> >
> > You _don't_??!! _WHO DOES_, then, if not the customer?
> >
> Eventually, the customer pays. But that's a lot different than asking the
> customer to fund the development cost up-front. That's what the venture
> capitalist or entrepeneur is supposed to do.

Actually, I thought that sort of thing isn't uncommon. Of course,
there's a lot of contractual complications and financial details
involved, but developing a library is hardly on the same level of risk
as commercializing a new technology, let alone a new discovery in
physics, for example.

> I think you are misinterpreting the intention here. I believe the point was
> about asking the customer to just simply pay some cost-plus development
> money in order to get a library. That is definitely not the same sort of
> thing as eventually paying off a business's asset acquisition by repeated
> sales of a product.
> 
> IOW: You go build it - if I like it and find it worth the price you charge,
> I buy it. It isn't usually done in the other order - unless you're the
> government! :-)

I don't think any halfway sane company would even attempt to sell
anything on those terms (full payment up-front, before anything is
provided), but I'm sure a lot of custom software is developed with
partial payments paid upon achievement of intermediate goals, and even
with some money paid up-front. Certainly a lot of construction jobs are
done that way.
 
> > > I'd really like to see you in a TV store :-)
> > > "Stephen, give me money now, I'm going to develop a great TV set, build
> a
> > > factory, and I promise you will have a great TV set within a few years
> ..."
> >
> > Duh... What do you think investors do every day? How in the world do you
> > think the vast majority of companies got started in the first place?
> > Government mandate, or magic?
> >
> A bit unfair. An investor is not the same thing as a customer.

Sometimes they are. :)
I take your point, though. I was a bit unfair, and also rudely
sarcastic. I was feeling particularly irritable when I wrote that, and I
should've had the sense to review and edit before posting. :(

Larry

Re: Proving Correctness (was Java Portability)

[Larry Elmore]

nicolas wrote:
> 
> "Larry Elmore" <ljelmore@home.com> a �crit dans le message news:
> 3B706ADC.B4847AC3@home.com...
> >From your statements
> > relating to economics, I'd guess you either have nothing to do with the
> > business side of your company, or else it's customers consist mostly of
> > government agencies or other companies dealing mostly with government
> > agencies.
> 
> Sorry to disapoint you ...
> We are a company small enough so that no side of the company is really far
> from anybody.
> Almost all our customers are private companies ready to jump to your
> competitor if he is better or cheaper than you.
> Or sometimes just because he feels like it, even if your product is far
> better
> I've learned about all that visiting customers and factories throughout the
> world to understand prospects and customers needs.
> I can assure you that investing unsuccessfully, in order to try to turn a
> prospect in a customer, is not that uncommon.
> 
> I'll be curious to know what's your job, and how close you are of the
> business side, to think otherwise ...

I'm a programmer for a small (~500 employees) hardware/software company
that makes diagnostic and network monitoring equipment for major telecom
companies. I don't have personal experience with the business side.

I (of course) agree with everything you say here. I'm sorry for the tone
of my previous post -- it was kind of rude and uncalled for, no matter
how irritated I was by what I was replying to (and I normally wouldn't
be that irritable, but something totally unrelated to your post already
had me going).

I don't think any halfway sane company would demand up-front payment in
full before even starting work, though. A negotiated system of partial
payments upon fulfillment of intermediate goals is reasonable in some
conditions, though, even with a portion paid up-front. I know it's
common in other segments of the economy.

I think your TV example was flawed because developing a software library
has nowhere near the risk if the company you're dealing with has a good
track record, and nothing has happened to it recently that might affect
its ability to carry out the job. Software libraries can also usually be
used in a restricted fashion long before they're complete, which is a
far different situation than a factory, which must be functionally
complete before it does anything useful.

Larry

Re: Proving Correctness (was Java Portability)

[Larry Elmore]

Ole-Hjalmar Kristensen wrote:
> 
> Larry Elmore <ljelmore@home.com> writes:
> 
> > nicolas wrote:
> > >
> 
> > > I'd really like to see you in a TV store :-)
> > > "Stephen, give me money now, I'm going to develop a great TV set, build a
> > > factory, and I promise you will have a great TV set within a few years ..."
> >
> > Duh... What do you think investors do every day? How in the world do you
> > think the vast majority of companies got started in the first place?
> > Government mandate, or magic?
> >
> > Larry
> 
> But he's not an investor, he's potential customer in the mass
> market. There is a difference you know, and compilers and libraries
> are commodities these days, not custom-made.

Not if the library doesn't yet exist and isn't available from _any_
supplier. In that case, you either write it yourself, you contract with
someone else to write it for you, or you wait for someone to develop it
(which might not be an option depending on when you really need it, and
on how specialized it might be). Not all compilers are commodity items,
either. The vast majority of custom compilers and interpreters are
developed in-house, I'm sure, but I doubt all of them are. There are
probably more of these than you might think.

Larry

Re: Proving Correctness (was Java Portability)

[Larry Elmore]

nicolas wrote:
> 
> "Larry Elmore" <ljelmore@home.com> a �crit dans le message news:
> 3B707688.340B8A70@home.com...
> > How many standard, cross-compiler, cross-platform libraries exist for
> > _any_ language? I'd be willing to bet more would be in Fortran than any
> > other language.
> 
> You are right, but I doubt that helps Ada a lot.

No it doesn't, unfortunately.
 
> > It's been no more "hyped" than any other OO language. If this is your
> > argument, why is C++ still around?
> 
> Microsoft ?

I'd say inertia, myself. C++ got popular largely, IMO, because C was
popular and it built on that base. It _appeared_ to minimize the risk
and the learning curve for an OO language. I think one of the reasons
Java got popular was because it came along just as many companies and
individuals were looking for an alternative to C++ after learning the
hard way what kinds of problems it has as a language. Partly due to
Java's superfical resemblance to C/C++, its (still not-quite-realized)
promise of easy portability, and its fortunate timing, Java took off.
 
> > I think your second point is bogus. Reinventing the wheel goes on
> > everywhere and in every language (and often because the developer feels
> > he/she can do a better job than whoever came before).
> 
> It's obvious to me that this is a counterproductive attitude.
> People fight or at least should fight with very good reasons against this
> kind of developer's feelings.

Yes, but managing programmers has frequently been compared to herding
cats, and with good reason! I think the fact that there is no officially
sanctioned standard for many commonly used libraries contributes to
this. I think this is even more important for a standards committee to
do than any changes to the language itself.

Larry

Re: Proving Correctness (was Java Portability)

[Simon Wright]

Stefan Skoglund <stetson@ebox.tninet.se> writes:

> I don't think an XML library would be that useful in an compiler
> targetting embedded applications.
> 
> Ow, nice we can send midcourse updates to the navigation
> computer as an XML stream. Really nice :-)

I don't quite see the problem here, I must say! after all, as an
example of using a high-level protocol in an embedded app, often
routers can be configured using an in-built http server.

My view is that if there is an (industry) standard {whatever} the
first approach considered should be to (re)use it.

Clearly this is less appropriate in the case where the unit cost
overwhelms the development effort.

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Well, a sane company might *try* to sell something on those terms (pay me
everything now and I'll give you a product later and/or pay me the full
development cost plus profit). However I doubt that ery many sane
*customers* would go for it.

Its not that these sort of arrangements don't ever happen. Note that the
giovernment often engages in exactly this sort of arrangement (Well, maybe
with progress payments rather than up front, but the risk is minimal to the
company. As long as they deliver on time, they get paid - often cost plus,
but not as much as it used to be.) The reason is quite simple: The
government is often buying products where a) the technology is totally
unproven, b) the usage is limited to some very small application domain, c)
they aren't going to let you sell the product anywhere else. The contractor
is in effect in an impossible risk position if the government doesn't fund
the development cost.

There may be times when there are similar situations in the private sector.
(unproven technology, limited usage, proprietary capabilities). You may see
some companies funding a development under these conditions. However, it is
a lot more rare than most developments and as I've pointed out elsewhere,
the customer is in a position to apply a lot of leverage on the suplier and
is not likely to give up the end result without some good and valuable
consideration.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Larry Elmore" <ljelmore@home.com> wrote in message
news:3B75C020.E00D149C@home.com...
> >
> > IOW: You go build it - if I like it and find it worth the price you
charge,
> > I buy it. It isn't usually done in the other order - unless you're the
> > government! :-)
>
> I don't think any halfway sane company would even attempt to sell
> anything on those terms (full payment up-front, before anything is
> provided), but I'm sure a lot of custom software is developed with
> partial payments paid upon achievement of intermediate goals, and even
> with some money paid up-front. Certainly a lot of construction jobs are
> done that way.
>

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Well, the situation probably lies somewhere between the two extremes. If the
software library we are talking about is fairly general-purpose in nature,
this is one thing. If it addresses some narrow niche, then its another.

I get the impression that we were talking about some semi-general-purpose
library that would be distributed with a compiler. Something similar to the
MFC - interfaces to the OS, graphics, data structures, etc. Notice that
Microsoft didn't go to one or more customers and say "Please fund the
development of this end product". (Of course at the size they are, there is
a lot more capability to develop things like this on speculation.)

If "Ada-Libraries-R-US" wanted to build a similar library of development
tools - one that has a market beyond one company or even one compiler
company - I don't think you'd find very many companies that would contract
for its development and expect to pay the full development cost. Someone
might partially fund it, realizing that a) they need it and b) there has to
be some seed money to get it going and c) this ends up cheaper than doing it
themselves. Even this, I think is doubtful and as evidence, I would observe
that there are companies that would benefit from having such a library and
there are any number of eager entrepeneurs, start-ups or existing hungry
companies that would want to build such a thing. Yet no such relationship
exists. Why?

I think it is because most customers are not willing to totally or even
partially fund a development of this kind. Its too risky & they have
alternatives. Why pay all or some of the cost of developing a library in Ada
if they can go to C++Compilers-R-Us and buy the whole thing for a fraction
of what it costs to get it built from bottom-dead-center in Ada? Because
then it would be in Ada? Why does a business care if it is in Ada, C++ or
Lisp? They care about how it impacts the bottom line & (say) $50,000 to
develop it in Ada vs $500 to buy it COTS in C++ has to be justified in those
terms.

Put yourself in the shoes of the customer (as I'm sure you are from time to
time) and ask why it is you buy or don't buy some software product. When
does your company out-source the development of any software? Why do they do
it? What conditions are placed on it? What advantages do they get and what
do they give up when they do this? These are the important questions to ask
if we want to understand why/why-not some company would want to fund the
development of an Ada library of some sort.

BTW: My experience has been that very few companies ever outsource custom
software development of any kind. (By this I mean contracting with an
outside company to build something they want that they can't get
off-the-shelf. Not the business of hiring consultants or other indirect
labor to work on it in-house.) Why is that? I can speculate, but perhaps its
more valuable to get the speculation of others.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Larry Elmore" <ljelmore@home.com> wrote in message
news:3B75C79C.38C670F2@home.com...
>
> I think your TV example was flawed because developing a software library
> has nowhere near the risk if the company you're dealing with has a good
> track record, and nothing has happened to it recently that might affect
> its ability to carry out the job. Software libraries can also usually be
> used in a restricted fashion long before they're complete, which is a
> far different situation than a factory, which must be functionally
> complete before it does anything useful.
>

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Simon Wright <simon@pushface.org> wrote:
 
: I can't parse that, sorry ..

what I wanted to say is that these components appear to
be darn useful (if actually used). Also, and not by accident,
cost-effective.

: I would have a _lot_ of trouble convincing my employer that I should
: accept payment from a third party for working on the BCs.

Even if your employer receives the money?

Re: Proving Correctness (was Java Portability)

[Marin David Condic]

Yeah. The model shouldn't likely be "I want to work on my employer's time
and receive financial remuneration from other parties as a result". It is
more likely to be successful as "I'll work on my spare time in the garage
and make some money as a result" or "I'll work on my employer's nickel and
they get any revenue as a result."

One problem is if the employer has already been paying for development of
the library in question and it has been given away for the most part so far.
What happens if you now expect to charge for it?

The reason a company probably won't go for it is that it isn't their core
business & they're not equipped to handle it. (Stick to your knitting!)
Hence, my best guess is that it will come from garage operations in some
form.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:9l8n2l$jml$1@a1-hrz.uni-duisburg.de...
> Simon Wright <simon@pushface.org> wrote:
> : I would have a _lot_ of trouble convincing my employer that I should
> : accept payment from a third party for working on the BCs.
>
> Even if your employer receives the money?
>

Re: Proving Correctness (was Java Portability)

[Simon Wright]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:

> Simon Wright <simon@pushface.org> wrote:

> : I would have a _lot_ of trouble convincing my employer that I should
> : accept payment from a third party for working on the BCs.
> 
> Even if your employer receives the money?

They wouldn't have much problem with that, but that is _not_ what I said!
".. that *I* should accept payment .."

Re: Proving Correctness (was Java Portability)

[Georg Bauhaus]

Marin David Condic <dont.bother.mcondic.auntie.spam@[acm.org> wrote:
 
: The reason a company probably won't go for it is that it isn't their core
: business & they're not equipped to handle it. (Stick to your knitting!)

Yes, and it is both sad and known that these guys very often overlook
what the long term effect of this is going to be. I'll have
to dig out that book by Mancur Olson again (Logic of Collective Action).

: Hence, my best guess is that it will come from garage operations in some
: form.

Well, quite a few things have started in garages :-)

Re: More Uniform Ada libraries (was: Proving Correctness)

[brentcarnellis]

On Thu, 09 Aug 2001 07:47:46 GMT, "nicolas" <n.brunot@cadwin.com>
wrote:
>
>Aonix Objectada is excellent for us, except that link time is far too long
>for our applications and cannot be used today for dayly development
>When the link time is 12mns instead of 20s, and has to be done 10 or 20
>times a day, it is a problem.

You have something royally screwed up in your environment.
Meaty-sized 50 Megabyte executables link in about 10 seconds with GNAT
(Sun, SGI, Linux, etc).

>Will Aonix go on with Ada ? some have doubts
>
>Rational Apex is not bad, but still have some little problems

No Apex has big problems for student programs, like not doing
Text_IO.Get_Line correctly.

Re: More Uniform Ada libraries (was: Proving Correctness)

[Ed Falis]

brentcarnellis@hotmail.com wrote:
> 
> On Thu, 09 Aug 2001 07:47:46 GMT, "nicolas" <n.brunot@cadwin.com>
> wrote:
> >
> >Aonix Objectada is excellent for us, except that link time is far too long
> >for our applications and cannot be used today for dayly development
> >When the link time is 12mns instead of 20s, and has to be done 10 or 20
> >times a day, it is a problem.
> 
> You have something royally screwed up in your environment.
> Meaty-sized 50 Megabyte executables link in about 10 seconds with GNAT
> (Sun, SGI, Linux, etc).

ObjectAda uses the Microsoft linker, which is what's taking up all the
time.

- Ed

Re: More Uniform Ada libraries (was: Proving Correctness)

[Jeff Creem]

<brentcarnellis@hotmail.com> wrote in message
news:3b9456ca.15085013@news.tc.umn.edu...
> On Thu, 09 Aug 2001 07:47:46 GMT, "nicolas" <n.brunot@cadwin.com>
> wrote:
> >
> >Aonix Objectada is excellent for us, except that link time is far too
long
> >for our applications and cannot be used today for dayly development
> >When the link time is 12mns instead of 20s, and has to be done 10 or 20
> >times a day, it is a problem.
>
> You have something royally screwed up in your environment.
> Meaty-sized 50 Megabyte executables link in about 10 seconds with GNAT
> (Sun, SGI, Linux, etc).
>
> >Will Aonix go on with Ada ? some have doubts
> >
> >Rational Apex is not bad, but still have some little problems
>
> No Apex has big problems for student programs, like not doing
> Text_IO.Get_Line correctly.
>
>

Hmm.. I have had my share of complaints over the years with Apex but not
doing Text_IO.Get_Line correctly was never one of them...There are probably
literally > 100 million lines of code of Ada compiled with Apex deployed in
the world so to
assume that it is that broken on a regular basis is.....odd..

Re: More Uniform Ada libraries (was: Proving Correctness)

[brentcarnellis]

On Wed, 05 Sep 2001 00:01:17 GMT, "Jeff Creem" <jeff@thecreems.com>
wrote:
><brentcarnellis@hotmail.com> wrote in message
>> On Thu, 09 Aug 2001 07:47:46 GMT, "nicolas" <n.brunot@cadwin.com>
>> wrote:
>> >
>> >Aonix Objectada is excellent for us, except that link time is far too
>long
>> >for our applications and cannot be used today for dayly development
>> >When the link time is 12mns instead of 20s, and has to be done 10 or 20
>> >times a day, it is a problem.
>>
>> You have something royally screwed up in your environment.
>> Meaty-sized 50 Megabyte executables link in about 10 seconds with GNAT
>> (Sun, SGI, Linux, etc).
>>
>> >Will Aonix go on with Ada ? some have doubts
>> >
>> >Rational Apex is not bad, but still have some little problems
>>
>> No Apex has big problems for student programs, like not doing
>> Text_IO.Get_Line correctly.
>>
>
>Hmm.. I have had my share of complaints over the years with Apex but not
>doing Text_IO.Get_Line correctly was never one of them...There are probably
>literally > 100 million lines of code of Ada compiled with Apex deployed in
>the world so to
>assume that it is that broken on a regular basis is.....odd..

Yes, once fixed it will be a regression test nightmare for Rational.
For example, if you do the following, don't expect it to work over a
remote pipe.

with Text_IO;
procedure Hello_World is
   Text : String(1..100);
   Len : Integer;
begin
   loop
      Text_IO.Get_Line(Text, Len);
      Text_IO.Put_Line("[Hello World] " & Text(1..Len));
   end loop;
end;

$ rsh ilab-02 hello_world
eat me
where is the response?
[Hello World] eat me
what?
[Hello World] where is the response?
what?
[Hello World] what?
thanks
[Hello World] what?
you suck
[Hello World] thanks
bye
[Hello World] you suck

Re: More Uniform Ada libraries (was: Proving Correctness)

[Samuel T. Harris]

brentcarnellis@hotmail.com wrote:
> 
> On Wed, 05 Sep 2001 00:01:17 GMT, "Jeff Creem" <jeff@thecreems.com>
> wrote:
> >
> >Hmm.. I have had my share of complaints over the years with Apex but not
> >doing Text_IO.Get_Line correctly was never one of them...There are probably
> >literally > 100 million lines of code of Ada compiled with Apex deployed in
> >the world so to
> >assume that it is that broken on a regular basis is.....odd..
> 
> Yes, once fixed it will be a regression test nightmare for Rational.
> For example, if you do the following, don't expect it to work over a
> remote pipe.
> 
> with Text_IO;
> procedure Hello_World is
>    Text : String(1..100);
>    Len : Integer;
> begin
>    loop
>       Text_IO.Get_Line(Text, Len);
>       Text_IO.Put_Line("[Hello World] " & Text(1..Len));
>    end loop;
> end;
> 
> $ rsh ilab-02 hello_world
> eat me
> where is the response?
> [Hello World] eat me
> what?
> [Hello World] where is the response?
> what?
> [Hello World] what?
> thanks
> [Hello World] what?
> you suck
> [Hello World] thanks
> bye
> [Hello World] you suck

Funny, my rather dated Apex 2.0.8D for SGI IRIX 5
works with no problems (and that Apex is, what, over
4 years old).

And my VADS 621.52 works as well (and that is a
pre-Rational true-blue Verdix compiler).

Anyway, you are confusing expections with required
LRM behavior. "Broken" is a word used when the compiler
does not do as it should and what it should do is defined
in the LRM. Many things about the LRM are implementation
defined and many things about the model of files defined
for Text_IO are incompatible with many aspects of the
real-world.

Consider the Honeywell Ada compiler for the DPS-6.
Because of the way the DPS-6 buffered terminal input
and output, a standard put("prompt "); get_line(response,last);
would wait for the input and then output the prompt and
the response. This was a conforming and certified implementation.
My expectations concerning reasonable behavior remained
unsatisfied because of the brain-dead operating system.

Now consider that the Text_IO model for files requires
an end_of_line followed by an end_of_page followed by
an end_of_file at the end of a file. Most UNIX text files
will only have the end_of_line. Because of this, most
UNIX files should produce something like end_error
and the function call end_of_file will never yield true.
This would be conforming behavior and the compiler
cannot said to be broken. It can be said that the compiler
does not meet my expectation and I would like an enhancement
to handle this exceptional, but common, situation.
Indeed every compiler I've every used takes some pains
to infer the end_of_page and end_of_file by using
some form of buffering.

Now consider what impact that buffering will have
on interactive streams such as standard_input,
especially through remote pipes. An implementation
of Text_IO which meets most of our expectations
will be rather complicated indeed since most of
the real-world "files" do not fit the definition
of a Text_IO file.

-- 
Samuel T. Harris, Senior Software Engineer II
Raytheon, Aerospace Engineering Services
"If you can make it, We can fake it!"

Re: Proving Correctness (was Java Portability)

[Ted Dennison]

In article <9kosp0$dje$1@nh.pace.co.uk>, Marin David Condic says...
>Asking someone to volunteer to make a library in his spare time and make it
>available everywhere freely and never offer any compensation to him is going
>to require that we find either a saint or a fool. I don't see any big rush

Since I'm not Catholic, I guess that make me a fool. Oh well, I've been called
far worse...

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html
          home email - mailto:dennison@telepath.com

Re: More Uniform Ada libraries (was: Proving Correctness)

[brentcarnellis]

On Wed, 05 Sep 2001 08:13:31 -0500, "Samuel T. Harris"
<u61783@gsde.hou.us.ray.com> wrote:

>brentcarnellis@hotmail.com wrote:
>> 
>> On Wed, 05 Sep 2001 00:01:17 GMT, "Jeff Creem" <jeff@thecreems.com>
>> wrote:
>> >
>> >Hmm.. I have had my share of complaints over the years with Apex but not
>> >doing Text_IO.Get_Line correctly was never one of them...There are probably
>> >literally > 100 million lines of code of Ada compiled with Apex deployed in
>> >the world so to
>> >assume that it is that broken on a regular basis is.....odd..
>> 
>> Yes, once fixed it will be a regression test nightmare for Rational.
>> For example, if you do the following, don't expect it to work over a
>> remote pipe.
>> 
>> with Text_IO;
>> procedure Hello_World is
>>    Text : String(1..100);
>>    Len : Integer;
>> begin
>>    loop
>>       Text_IO.Get_Line(Text, Len);
>>       Text_IO.Put_Line("[Hello World] " & Text(1..Len));
>>    end loop;
>> end;
>> 
>> $ rsh ilab-02 hello_world
>> eat me
>> where is the response?
>> [Hello World] eat me
>> what?
>> [Hello World] where is the response?
>> what?
>> [Hello World] what?
>> thanks
>> [Hello World] what?
>> you suck
>> [Hello World] thanks
>> bye
>> [Hello World] you suck
>
>Funny, my rather dated Apex 2.0.8D for SGI IRIX 5
>works with no problems (and that Apex is, what, over
>4 years old).
>
>And my VADS 621.52 works as well (and that is a
>pre-Rational true-blue Verdix compiler).

That's why I mentioned the regression test scenario. 
In your particular case, Apex Get_Line apparently provided utility
at one time but current versions are unusable over a remote pipe. 

Re: More Uniform Ada libraries (was: Proving Correctness)

[Samuel T. Harris]

brentcarnellis@hotmail.com wrote:
> 
> On Wed, 05 Sep 2001 08:13:31 -0500, "Samuel T. Harris"
> <u61783@gsde.hou.us.ray.com> wrote:
> 
> >brentcarnellis@hotmail.com wrote:
> >>
> >
> >Funny, my rather dated Apex 2.0.8D for SGI IRIX 5
> >works with no problems (and that Apex is, what, over
> >4 years old).
> >
> >And my VADS 621.52 works as well (and that is a
> >pre-Rational true-blue Verdix compiler).
> 
> That's why I mentioned the regression test scenario.
> In your particular case, Apex Get_Line apparently provided utility
> at one time but current versions are unusable over a remote pipe.

Still funny. My recent Apex 3.2.0 installation also
performs as you expect. Just what version and platform
is mis-behaving for you and have you reported it
to Rational support?

-- 
Samuel T. Harris, Senior Software Engineer II
Raytheon, Aerospace Engineering Services
"If you can make it, We can fake it!"

Re: More Uniform Ada libraries (was: Proving Correctness)

[brentcarnellis]

On Thu, 06 Sep 2001 08:29:48 -0500, "Samuel T. Harris"
<u61783@gsde.hou.us.ray.com> wrote:

>brentcarnellis@hotmail.com wrote:
>> 
>> On Wed, 05 Sep 2001 08:13:31 -0500, "Samuel T. Harris"
>> <u61783@gsde.hou.us.ray.com> wrote:
>> 
>> >brentcarnellis@hotmail.com wrote:
>> >>
>> >
>> >Funny, my rather dated Apex 2.0.8D for SGI IRIX 5
>> >works with no problems (and that Apex is, what, over
>> >4 years old).
>> >
>> >And my VADS 621.52 works as well (and that is a
>> >pre-Rational true-blue Verdix compiler).
>> 
>> That's why I mentioned the regression test scenario.
>> In your particular case, Apex Get_Line apparently provided utility
>> at one time but current versions are unusable over a remote pipe.
>
>Still funny. My recent Apex 3.2.0 installation also
>performs as you expect. Just what version and platform
>is mis-behaving for you and have you reported it
>to Rational support?
>
>-- 
>Samuel T. Harris, Senior Software Engineer II
>Raytheon, Aerospace Engineering Services
>"If you can make it, We can fake it!"

Apex 4.0 and No I won't report it because we use GNAT and it's
Text_IO.Get_Line works as expected over a remote pipe.

I decided to experiment with the aid of the Solaris "truss" 
filter by rerunning the program with input AAA followed by BBB.
Notice that "Hello World: AAA" is only echoed to stdout after 
BBB is read. Apex Text_IO.Get_Line is definitely reading the first AAA
(check the read call) but refuses to echo it out until the second line
is entered! 

write(1, " P l e a s e   e n t e r".., 22)	= 22
write(1, "\n", 1)				= 1
door_return(0x00000000, 0, 0x00000000, 0)	= 0
open("/dev/zero", O_RDWR)			= 3
mmap(0x00000000, 24576, PROT_READ|PROT_WRITE|PROT_EXEC, 
MAP_PRIVATE|MAP_NORESERVE, 3, 0) = 0xEF4D0000
close(3)					= 0
mprotect(0xEF4D0000, 8192, PROT_NONE)		= 0
lwp_create(0xEF5A5A40, 0x00C0, 0xEF4D5E0C)	= 5
lwp_continue(5)					= 0
lwp_create(0x00000001, 0x6000, 0xEF5A5BDC)	= 0
lwp_schedctl(SC_STATE|SC_BLOCK, -1, 0xEF4D5D14)	= 0
lwp_schedctl(SC_DOOR, 0, 0x00000000)		= 3
door_bind(3)					= 0
close(3)					= 0
lwp_self()					= 3
lwp_create(0xEF4E5840, 0x00C0, 0xEF20BE0C)	= 6
lwp_continue(6)					= 0
lwp_create(0x00000000, 0, 0x00000000)		= 0
lwp_self()					= 6
lwp_schedctl(SC_STATE|SC_BLOCK, -1, 0xEF20BD74)	= 0
lwp_sema_post(0xEF7884D8)			= 0
sigprocmask(SIG_BLOCK, 0xEF20BE2C, 0x00000000)	= 0
sigprocmask(SIG_UNBLOCK, 0xEF7884C8, 0x00000000) = 0
lwp_sema_wait(0xEF7884D8)			= 0
sigprocmask(SIG_BLOCK, 0xEF7884C8, 0x00000000)	= 0
setitimer(ITIMER_REAL, 0xEF20BCD0, 0x00000000)	= 0
sigprocmask(SIG_UNBLOCK, 0xEF7884C8, 0x00000000) = 0
door_return(0x00000000, 0, 0x00000000, 0)	= 0
open("/dev/zero", O_RDWR)			= 3
mmap(0x00000000, 24576, PROT_READ|PROT_WRITE|PROT_EXEC, 
MAP_PRIVATE|MAP_NORESERVE, 3, 0) = 0xEF4C0000
close(3)					= 0
mprotect(0xEF4C0000, 8192, PROT_NONE)		= 0
lwp_create(0xEF4D5A40, 0x00C0, 0xEF4C5E0C)	= 7
lwp_continue(7)					= 0
lwp_create(0x00000001, 0x6000, 0xEF4D5BDC)	= 0
lwp_schedctl(SC_STATE|SC_BLOCK, -1, 0xEF4C5D14)	= 0
lwp_schedctl(SC_DOOR, 0, 0x00000000)		= 3
door_bind(3)					= 0
close(3)					= 0
sigaction(SIGWAITING, 0xEF787B78, 0x00000000)	= 0
time()						= 999823649


read(0, " A A A\n", 400)			= 4
BB    Received signal #14, SIGALRM, in lwp_sema_wait() [caught]
lwp_sema_wait(0xEF7884D8)			Err#91 ERESTART
sigprocmask(SIG_SETMASK, 0xEF20BE2C, 0x00000000) = 0
lwp_sema_post(0xEF4E5E78)			= 0
lwp_sema_wait(0xEF4E5E78)			= 0
setcontext(0xEF20B728)
sigprocmask(SIG_BLOCK, 0xEF7884C8, 0x00000000)	= 0
sigprocmask(SIG_UNBLOCK, 0xEF7884C8, 0x00000000) = 0
lwp_sema_post(0xEF7884D8)			= 0
lwp_sema_wait(0xEF7884D8)			= 0
sigprocmask(SIG_BLOCK, 0xEF7884C8, 0x00000000)	= 0
setitimer(ITIMER_REAL, 0xEF20BCD0, 0x00000000)	= 0
sigprocmask(SIG_UNBLOCK, 0xEF7884C8, 0x00000000) = 0
B
read(0, " B B B\n", 397)			= 4
Hello world! AAAwrite(1, " H e l l o   w o r l d !".., 16)	= 16
write(1, "\n", 1)				= 1

    Received signal #14, SIGALRM, in lwp_sema_wait() [caught]

Re: More Uniform Ada libraries (was: Proving Correctness)

[Samuel T. Harris]

brentcarnellis@hotmail.com wrote:
> 
> Apex 4.0 and No I won't report it because we use GNAT and it's
> Text_IO.Get_Line works as expected over a remote pipe.
> 

I see. You are willing to take time and effort to
discuss it in this forum yet you refuse to report
the problem to the vendor. Hmmmmm ....

-- 
Samuel T. Harris, Senior Software Engineer II
Raytheon, Aerospace Engineering Services
"If you can make it, We can fake it!"