Re: Java momentum slowing ?

Re: Java momentum slowing ?

[Corey Ashford]

Jean-Marten Marchi wrote:
> 
> It's just a question, and i want to know what you feel about this
> subject.
> 
> I feel personally that the momentum behind Java is slowing. Just
> remember what was said about it 2 years from now, Network Computing,
> etc etc...
> 
> Today, where are those Java apps? Where are Network Computing Devices?
> Thin clients ? Furthermore, i don't see too much of Java applets on
> the Web pages, where Java has its strength.
> 
> Thanks to you.

Thin clients are available now and have been for many months
(check out www.ncd.com).

Java lost some of its momentum due, I think, to:

1) over-hype
2) pollution of the standard by Microsoft and others
3) Sun trying to keep control of it.

I think it will still be popular, but it seems to have split
into two camps now - J++ / COM  and Java / JavaBeans.

Re: Java momentum slowing ?

[x]

In article <37289A2C.4B78EB60@rocketmail.com>, Corey Ashford <yeroca@rocketmail.com> wrote:
>Jean-Marten Marchi wrote:
>
>Thin clients are available now and have been for many months
>(check out www.ncd.com).

I work with Oracle Applications which now has a Java thin client front end.  
Actually, it runs entirely from appletviewer.  That makes client machine set 
up a breeze.

One question I have is how does one define "thin".  It still takes a lot of 
software to achieve that.

>Java lost some of its momentum due, I think, to:
>
>1) over-hype
>2) pollution of the standard by Microsoft and others
>3) Sun trying to keep control of it.

4) It's SLOW (I don't care what anyone says about "Java is almost as fast as 
C++".  BS.)

I think the momentum for Java has slowed, but I also think it's starting to 
settle in and find its niche.  It *will* eventually find it's place, but I 
don't think the world will be rewritten in Java as Sun originally dreamed.

In the end, it will be just another language.

Java momentum slowing ?

[Jean-Marten Marchi]

It's just a question, and i want to know what you feel about this
subject. 

I feel personally that the momentum behind Java is slowing. Just
remember what was said about it 2 years from now, Network Computing,
etc etc... 

Today, where are those Java apps? Where are Network Computing Devices?
Thin clients ? Furthermore, i don't see too much of Java applets on
the Web pages, where Java has its strength. 

Thanks to you. 

Re: Java momentum slowing ?

[Tony Huynh]

IMHO I think that java apps and javascripting are too slow to load, as a
webdeveloper I use java to do things but it does hinder the efficiency of
the website.  What Java really needs is fast loading (along with most
things that look good on the web).  There are many clients who simply do
not want java in their website because of the slow loading rate.  I don't
know how much of what I said is relevant to your concern, but that is my
opinion of Java.

Jean-Marten Marchi wrote:

> It's just a question, and i want to know what you feel about this
> subject.
>
> I feel personally that the momentum behind Java is slowing. Just
> remember what was said about it 2 years from now, Network Computing,
> etc etc...
>
> Today, where are those Java apps? Where are Network Computing Devices?
> Thin clients ? Furthermore, i don't see too much of Java applets on
> the Web pages, where Java has its strength.
>
> Thanks to you.

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On Fri, 30 Apr 1999 09:06:55 +1000, Tony Huynh <wdragon@uq.net.au>
wrote:

>IMHO I think that java apps and javascripting are too slow to load, as a
>webdeveloper I use java to do things but it does hinder the efficiency of
>the website.  What Java really needs is fast loading (along with most
>things that look good on the web).  There are many clients who simply do
>not want java in their website because of the slow loading rate.  I don't
>know how much of what I said is relevant to your concern, but that is my
>opinion of Java.

In fact, it's a feeling i have (Java momentum slowing) but want to see
if i'm the only one. Apparently not.

I think that the hype around Java has done serious damage to Ada and
its community, and it's sad. As Java doesn't lives up to its
expectations, there are some reasons that the Ada language can get a
boost. 

3 years ago, i invested some time working with Ada95 (with the GNAT),
trying to build a binding for OS/2. It was fun, but stopped when the
Java hype was at the Maxx and the DOD pulled the plug. Why continue to
invest time and energy in (apparently) a dead end. Remember, it was
quite impossible to talk about something else than Java. It's Linux
time now ! (not saying it won't succeed).

Too, i did'nt have time (children, etc etc). But, as children grow and
get more educated... , i can have some time to hack some code. I took
a look  to this newsgroup and found it was still busy. So, i'm asking
myself, why not Ada again? 

Re: Java momentum slowing ?

[Mike]

In article <372BB93F.7DCFD2E1@rocketmail.com>, Corey says...
 
>
>I think you're burying your head in the sand if you don't recognize that
>M$ is and will continue to be a force to reckon with in the software
>world.

If MS shoddy software is a force in the software 
world, I feel now more sorry for the software world in
what it has put itself into.

I can recongnize that M$ software is used in many places,
but I don't have to follow the crowds blindly as many seem
to do, and I don't have to be one of those windows generation
programmers.

I've have developed software for few companies in
the last 3 years, and I am proud to say I have not 
needed to write one single line of code on windows.
(That does not mean I do not know windows, I do, and
I have programmed on it before, and that is why
I decided to get away from it, I did not like it 
at all).
 
All my work in the last 3 years has been on Unix. And 
thanks to Java now, I'll never need to touch windows 
for any reason again.

I really have no idea what do people use windows for (other
than games), as Unix and Linux has everything anyone wants, 
and in much better quality, and with GNU/linux, I get
the source code also, and now with Java being used more
and more, it even removes the need for windows all together.
 
At number of places where I worked, we use Solaris and now 
Linux starting to show up more, at home I use Linux. I write 
everything from database applications, to GUI, to 
transaction processing to network management software, 
all on Unix and Linux and with Java and C/C++, and even 
some Ada (thanks to GNAT being on Unix).

Why why would any one choose windows, a flaky and closed OS, 
with licenses restrictions, and higher costs, over an open system
like Unix for developing applications on is beyond my 
comprehension.
 
I can see the general public choosing windows out of
ignorance of the alternatives, but I am amazed when I see
technical pepole, managers, and software engineers who 
should know better choosing windows to develop software on.

May be some of us have their head buried, but we love 
where it is buried and for me, it does not seem to 
have affected me in any way. I tell any new hiring 
manager that I do not do windows, only Unix, and I 
am more busy than ever, last year alone I made $400,000 
in just contract programming fees :) (I have been programming
for 16 years now) and again, never touched one windoz box,
(well, I did touch them, but I do not program on them). 
and I will never do. 
 
regards, 
Mike. 
a windows liberated programmer and loving it.

Re: Java momentum slowing ?

[Corey Ashford]

David Botton wrote:
> 
> Check out DCOM Free for Linux:
> 
> http://www.softwareag.com/corporat/solutions/entirex/entirex.htm
> 
> COM/DCOM is alive and well on non-MS platforms.
> 
> COM+ can best be described to the Ada world as a MS defined thick
> binding to COM.
> 
> David Botton

Very cool!

Thanks for the reference!

- Corey

Re: Java momentum slowing ?

[bob]

In article <372CBBDA.F0EDD79D@Botton.com>, David says...
>
>Check out DCOM Free for Linux:
>
>http://www.softwareag.com/corporat/solutions/entirex/entirex.htm
>
>COM/DCOM is alive and well on non-MS platforms.
>
 
Sorry. No source code provided. Only binary libraries. When they release
it with source code, then I'll use it.

I will not commit my software to be integrated abd be dependent on 
something that is closed and with no source code. If you are wise, 
you will do the same. 

I have no problem paying $$ for 3rd party libraries to use and for 
support, but I will not do it if the source code does not come with it. 

In my applications on Linux, every thing I use is open source (GPL'ed etc..)

btw, I have not seen a single programmer so far use COM/DCOM on non-windows
platform. And even in the companies that use windows machines, the ones
I know of, are using Java or C++ with Corba (on windows). 

I have no problem using DCOM, but when it opens up, and not be controlled 
by just one company, and not before.
 
Bob

Re: Java momentum slowing ?

[Mitch]

In article <7git59$1085@drn.newsguy.com>, bob@nospam says...
 
>Sorry. No source code provided. Only binary libraries. When they release
>it with source code, then I'll use it.
>
>I will not commit my software to be integrated abd be dependent on 
>something that is closed and with no source code. If you are wise, 
>you will do the same. 
>
 
Not only is there is source code present, COM/DCOM specification are
closed. Meaning no one can create an open source free DCOM product,
becuase MS will not make such complete specifications public, 
and MS can control the direction and any new features in COM.

 
One can argue than Sun did not also supply the source code with their
Java JVM. but the difference is the the JVM specifications are open,
and many have now implemented their own JVM's based on those 
specs (IBM, oracle, Linux folks http://www.blackdown.org, etc..

(Still many people are not happy with Sun not making Java 
open source) and many are not happy that Sun having the only control
on Java direction now (even though Sun is more open with Java
than MS is with MS own products).

Any way, I agree that it is not wise to link your application to
a binary only libraries, with no open specification or open source.

If I do so, and the above company goes under tommorrow, I am stuck.

I thought people are smarter these days than to get themselves in this
predicement again.

Mitch
 

Re: Java momentum slowing ?

[Tom]

In article <7giv8n$13fo@drn.newsguy.com>, Mitch@nospam@ says...
>

>In article <7git59$1085@drn.newsguy.com>, bob@nospam says...
> 
>>Sorry. No source code provided. Only binary libraries. When they release
>>it with source code, then I'll use it.
>>
>>I will not commit my software to be integrated abd be dependent on 
>>something that is closed and with no source code. If you are wise, 
>>you will do the same. 
>>
> 

I agree.

See below, this is from 5/99 interview in Linux Journal with Larry Wall:
 
".... There is no such thing as a perfect organism, biologically
speaking. About the most you can say is an organism is more or less
suitable for the environment in which it finds itself. In fact,
biologists are just now realizing that any organism which seems to be
"perfect" for one environment is likely to be in danger of extinction
as soon as the evnvironment changes. Over- specialization is only as
good as your ecological niche. We're not just talking about dinosaurs
here, but also snail darters and cheetahs and a bazillion beetles in
Brazil---not to mention Visual Basic.

     We've already seen the deaths of many over-specialized organisms
in computing: Lisp machines, Ada chips and many so- called fourth
generation languages. Any program ever written in assembly language
for an obsolete architechure is now obsolete. Likewiise, any program
that ties its fortunes to a single operating system is likely to go
down with the ship. I don't know how many more torpedoes Windows can
take before it sinks, but if and when it does, a whole batch of
specialized programs are going down with it. Obviously, for reasons
relating to the open source movement, Linux does'nt have this
particular problem...."

 
The good thing is that Ada is open source (thanks to Gnat), so, if you
use Gnat, you'r safe when M$ goes billy up :)

Tom

Re: Java momentum slowing ?

[Corey Ashford]

mitch@nospam wrote:
> 
> In article <37289A2C.4B78EB60@rocketmail.com>, Corey says...
> >
> 
> >I think it will still be popular, but it seems to have split
> >into two camps now - J++ / COM  and Java / JavaBeans.
> 
> J++ is dead. you have not been keep up with the new. MS killed J++.
> no body was using it.
> 
> COM (or it it COM+ or is it DCOM these daya?) only runs on windows.

Where are you getting your information?

COM (DCOM is really just COM with a some distribution features, COM+
is the new thing) runs on other platforms besides Microsoft
(Sun and HP, if I recall correctly).  Granted, though, it's primary
use is on WinX platforms.

Microsoft didn't kill J++ as far as I can tell.  About six months ago
they released
VisualJ++ 6.0.  Check out this Q&A at Microsoft's web site for details
about the court ruling: http://msdn.microsoft.com/visualj/vjfaq.asp


> only an idiot will commit their software to run on only one platform
> using closed technology, and we know the world is full idiots, else
> windows will not be used as much as it is.

If your market was PC's, and you intended to run your code only on
PC's (for whatever reason), does that make you an idiot fo using a technology
which is well supported on the PC platform?

> 
> Java and Corba are getting more popular, offcourse, there will always be
> some managers, those bean counters types, with no clue about software, who
> will insist on ordering M$ products for they know no better, I've worked
> for one such moron.

I'm not saying M$ is great stuff, but a lot of developers are using it, and
that does give a developer the benefit of a large base of coders banging
away at problems and finding work-arounds for M$ weirdities.

I think you're burying your head in the sand if you don't recognize that
M$ is and will continue to be a force to reckon with in the software
world.

Re: Java momentum slowing ?

[Corey Ashford]

Thanks for your thoughts, Mike.

I agree with all of your points, and if you can avoid M$, do so.

For the time being, there are a huge amount of WinX PC's out there
and so there is a strong financial incentive to continue to
develop apps for that platform.

I suppose it's partially because of people's ignorance of other available
technologies that we're in this bind.  Unfortunately that's just
the way it is.  Kind of a chicken-and-egg scenario.  Also, there's
a real lack of consumer level apps for Unix (i.e. Linux).  The one
I hear most often is, "can I run quicken on it?  If not, forget it."

Myself, I've never developed WinX apps either (been in the embedded
Ada and Unix Ada world for many years now), but my wife does and
I hear quite a lot from her about it (both grumbling and success
stories).  I'm also taking a class in
building software components.  The technology used in the class
is J++/COM, but we're also looking at Java Beans and Corba.
(As a side note, I think it's interesting that Mozilla has chosen
COM for its component technology, and apparently has it running
on Unix platforms).

Good luck to ya.

Re: Java momentum slowing ?

[Tarjei Tj�stheim Jensen]

Mike@spam_free wrote:

> Why why would any one choose windows, a flaky and closed OS,
> with licenses restrictions, and higher costs, over an open system
> like Unix for developing applications on is beyond my
> comprehension.

A reason is that people want to do real work. Only recently have good office
support application made their appearances on anything but PCs (Windows,
Mac).  DDE, OLE (COM) and later ODBC are very good concepts which effectively
terminated the usefulness of any competing platform not doing the same. DDE,
OLE and ODBC made Windows _the_ integration platforms (crashes? yes, but it
_works_ ). This is what makes Windows an effective work platform even if the
implementation is not neccessarily efficient and stable.

There is also the "minor" point of  the X/GUI people deciding that printing
was not important and had nothing to do with graphical user interfaces.
Calling those who decided that idiots would be cheritable.

I have talked to hardened Unix veterans and we pretty much agreed that for
real life office work Windows 3.0 on a 286 was more useful than any Unix
system. The key words are word prosessing, spreadsheet, presentation
graphics  and printer drivers (not neccessarily in order of importance). For
later versions of Windows you can add access and its work alikes.


Greetings,

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On 30 Apr 1999 18:28:42 -0700, Mike@world.nospam wrote:

Thanks for your input. 

But i'm still thinking that there's something wrong with Java. I
particularly don't understand why there are not so much Java applets
on the Web pages, where the portability of Java is the most efficient.
There were more applets 2 years ago. 

I'm just telling that there are signs that the momentum is slowing,
not that Java suddenly disappeared.I had many relatives trying to hack
some Java code 2 years ago. But they all stopped. Perhaps for
different reasons. 

 Instead, Linux is now under the spotlight and i can tell you that it
sells like hell. Around me, they are all trying to install a Linux
distrib (at home and at work) and trying to master the GNU tools.
Intranet servers based on Linux are growing like mushrooms. 

On another front, it seems that M$ has stopped all Java developments
and is pushing VC++ 6.0 very hard. 

So, like Ada, Java will have to find its niche. 

> 
>Java is getting used more and more than ever.
>
>I work in silicon valley as software contractor, 2 years ago, only
>few positions required Java, now, I can't find something that is NOT
>in java.  my last contract was Java, and next one is Java. Got 3 calls
>today from agencies, and the 3 positions were Java.
>
>Not C, Not C++ (and of course not Ada as many here have even heared of it, 
>never mind using).
> 
>And these are not small projects, these are large, networking, databases,
>transaction processing, etc.. applications, being done by well known 
>companies around here.
> 
>If this is slowing down, I don't know what being popular is.
>
>One good thing about all of this, is that I don't have to use C and C++
>anymore. If I can't use Ada, at least I can use Java. The one good thing
>Java did is free us from C and C++.
>
>
>Mike.

Re: Java momentum slowing ?

[Bob]

In article <372f0ed9.104817048@news2.ibm.net>, jmarten@ibm.net says...
 
>But i'm still thinking that there's something wrong with Java. I
>particularly don't understand why there are not so much Java applets
>on the Web pages, where the portability of Java is the most efficient.
>There were more applets 2 years ago. 
>

Thinking that there is something wrong with Java, does not mean there
is something wrong with it :)

As for applets, I think it becuase applets take more time to load, and
the more logic you add to the, the more btye code, and the longer it
takes to download. 

And also there are other technologies to do something similar to 
applets (ie. logic that execute inside the browser) such as javaScript 
(which have nothing to do with Java). Even plain HTML can do 
the job just fine for data entry from the client. Also browsers a
re not yet standardized in term of JVM support.

>I'm just telling that there are signs that the momentum is slowing,
>not that Java suddenly disappeared.I had many relatives trying to hack
>some Java code 2 years ago. But they all stopped. Perhaps for
>different reasons. 
>

Ok, I don't know what your relatives do for living and why they stopped :) 
but where I am, there is plenty of demand for Java, and it is increasing, 
and alot of it is on the server side, but also alot of GUI is now being 
written in swing, and on windows also. I know of a large company in this
area where they even forbid anyone at the company to write in C or C++ any
new applications! The manager there is in love with Java.

If you need to write multithreaded applications on the server side, the 
choices are to either use C++ and some external threads library 
with all the problems of porting issues from one platform to another, 
or use Java. (Ada would be a great choice also for server side applications, 
but for some reason it is not used here). Java is better integrated with
Corba also, and many who write corba apps in Java say it is easier than
C++ with corba.

> Instead, Linux is now under the spotlight and i can tell you that it
>sells like hell. Around me, they are all trying to install a Linux
>distrib (at home and at work) and trying to master the GNU tools.
>Intranet servers based on Linux are growing like mushrooms. 
>

Sure. Linux is also spreading very fast. GNU/Linux will be the 
dominant OS in few short year, and windows will be something of the
past, a big buggy bloated legacy complicated C++ code for someone 
to maintain for many years to come with service pack after service pack.

Java+Linux are a powerfull combination too. A Solid open free OS to 
run your JVM's on. what more can one ask for? plenty of development tools 
also.  Java on Linux is still behnind that on windows, but so is 
every other platform. Even Solaris is behind WIndows in this area. 
HotSPot was just released by SUN, and it is for windows only. Not 
even for Solaris. 

(Actually I also think Gnat+Linux are a very powerfull combination, and
I would like to see more Ada application on Linux. There is a window here
for Ada as Linux becomes more used. I do not think Ada on windows has much
of chance as it would have on an open system like Unix.

>On another front, it seems that M$ has stopped all Java developments
>and is pushing VC++ 6.0 very hard. 
>

Yes. MS has killed J++, (I read that few people only used it) and 
it is moving away from Java. They decided they can't win the Java 
fight. They will come up with something similar to Java, but not 
call it Java, and whatever it is, it will only work on windows as 
usuall, and some fish will take bait, but by then Linux and Java would
have had a solid standing in the industry.

>So, like Ada, Java will have to find its niche. 
>

Java is niche? you must be joking. right? tell that to IBM who is investing
billions on Java, tell that to Cisco who is writing all of their network
management SNMP code in Java, tell that to any of the large companies who
are now writing most of their new applications in Java. Try to look at the
programming jobs out there and see for yourself.

I personally think Ada is a better language than Java, but Java is still
a very good language, and is an improvment over C and C++ for sure. I have
programmed in all languages for many years. I am much more productive in
Java than in C or C++, and my applications have less bugs (same effect I have
when I program in Ada compared to C/C++).  This is not just my experience,
but also many other programmers have seen the same effect. 
 
cheers,
Bob

Re: Java momentum slowing ?

[David Botton]

Check out DCOM Free for Linux:

http://www.softwareag.com/corporat/solutions/entirex/entirex.htm

COM/DCOM is alive and well on non-MS platforms.

COM+ can best be described to the Ada world as a MS defined thick
binding to COM.

David Botton


> COM (DCOM is really just COM with a some distribution features, COM+
> is the new thing) runs on other platforms besides Microsoft
> (Sun and HP, if I recall correctly).  Granted, though, it's primary
> use is on WinX platforms.
>

Re: Java momentum slowing ?

[David Botton]

You can do as much, ok more, with Ada and COM then J++! Take the course,
but don't retire your Ada hat. (http://www.botton.com/ada/com - Its
coming....)

COM if looked at as a frame work (which is what it is) and a method of
programming is not restricted to any technology, platform, etc. There is
no reason not to implement code on a UNIX platform as COM style objects.

I can not say I am an MS fan, but COM was done right and it has a long
bright future on MS and NON MS platforms.

David Botton


> I'm also taking a class in
> building software components.  The technology used in the class
> is J++/COM, but we're also looking at Java Beans and Corba.
> (As a side note, I think it's interesting that Mozilla has chosen
> COM for its component technology, and apparently has it running
> on Unix platforms).
> 
> Good luck to ya.

Re: Java momentum slowing ?

[dvdeug]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 966 bytes --]

In article <372C00C6.D9AA6D72@online.no>,
  "Tarjei Tj�stheim Jensen" <tarjei@online.no> wrote:
> There is also the "minor" point of  the X/GUI people deciding that printing
> was not important and had nothing to do with graphical user interfaces.
> Calling those who decided that idiots would be cheritable.

Why? It has nothing to do with GUI's. I might want to print from the CLI at
anytime, and would want it to work the same way. Part of Unix's advantage is
modularity. If something breaks, it and only it can be fixed or replaced. This
would mix two totally unrelated things.

Not saying that printing doesn't need some work in Unix (I still haven't got
my printer working under Linux), but it's not X's job to do printing, and
there is no reason it should be.

--
David Starner - dstarner98@aasaa.ofe.org

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On 1 May 1999 20:50:23 -0700, Mike@spam_free wrote:

>In article <372BB93F.7DCFD2E1@rocketmail.com>, Corey says...
> 
>>
> 
>All my work in the last 3 years has been on Unix. And 
>thanks to Java now, I'll never need to touch windows 
>for any reason again.
>
>I really have no idea what do people use windows for (other
>than games), as Unix and Linux has everything anyone wants, 
>and in much better quality, and with GNU/linux, I get
>the source code also, and now with Java being used more
>and more, it even removes the need for windows all together.
> 

In fact, it appears clearly that (3 years ago) Java was a marketing
answer from Sun to M$. It didn't succed too much. 

But, now, as Linux grows exponentially, Java will have to compete not
as a platform, but as a language. It's a field where Ada95 has key
advantages. 

Linux and Gnat : the perfect match ! 

 




 

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On 2 May 1999 03:45:57 -0700, Bob@nospam wrote:

>In article <372f0ed9.104817048@news2.ibm.net>, jmarten@ibm.net says...

>
>As for applets, I think it becuase applets take more time to load, and
>the more logic you add to the, the more btye code, and the longer it
>takes to download. 
>
>And also there are other technologies to do something similar to 
>applets (ie. logic that execute inside the browser) such as javaScript 
>(which have nothing to do with Java). Even plain HTML can do 
>the job just fine for data entry from the client. Also browsers a
>re not yet standardized in term of JVM support.

This is problem one, with Java. It has not succeeded where it was the
most easy. 

>If you need to write multithreaded applications on the server side, the 
>choices are to either use C++ and some external threads library 
>with all the problems of porting issues from one platform to another, 
>or use Java. (Ada would be a great choice also for server side applications, 
>but for some reason it is not used here). Java is better integrated with
>Corba also, and many who write corba apps in Java say it is easier than
>C++ with corba.

On the server side, i don't understand why Java would be so useful.
There's no GUI to drive, it has to be fast But, maybe i'm wrong ?  

>
>(Actually I also think Gnat+Linux are a very powerfull combination, and
>I would like to see more Ada application on Linux. There is a window here
>for Ada as Linux becomes more used. I do not think Ada on windows has much
>of chance as it would have on an open system like Unix.

We are totally in sync on this point. Effectively, i see that there's
a huge opportunity there for Ada. (let's me modest,  i imagine that
i'm not alone) 

>
>Java is niche? you must be joking. right? tell that to IBM who is investing
>billions on Java, tell that to Cisco who is writing all of their network
>management SNMP code in Java, tell that to any of the large companies who
>are now writing most of their new applications in Java. Try to look at the
>programming jobs out there and see for yourself.

I'm sometime a bit excessive. But it was just to illustrate that the
momentum is so slowing that it will be niche soon. We have not to be
too serious ! 

As for IBM, they can jump in the Linux bandwagon anytime soon, just
putting their java in maintenance mode for their customers who bought
the Java hype too fast... Doing so, they wouldn't have to deal with a
- partner - like Sun ... 
 
Cordially,

Re: Java momentum slowing ?

[Tarjei Tj�stheim Jensen]

dvdeug@my-dejanews.com  wrote:

>Why? It has nothing to do with GUI's. I might want to print from the CLI at
>anytime, and would want it to work the same way. Part of Unix's advantage is
>modularity. If something breaks, it and only it can be fixed or replaced. This
>would mix two totally unrelated things.

How do you think MS word users would react if Microsoft told them that they
could not print their documents. Or that they would have to pay for a third
party accessory to print.

>Not saying that printing doesn't need some work in Unix (I still haven't got
>my printer working under Linux), but it's not X's job to do printing, and
>there is no reason it should be.


Which is why Microsoft still makes a lot of money.


Greetings,

Re: Java momentum slowing ?

[David Starner]

"Tarjei Tj�stheim Jensen" wrote:
> 
> dvdeug@my-dejanews.com  wrote:
> 
> >Why? It has nothing to do with GUI's. I might want to print from the CLI at
> >anytime, and would want it to work the same way. Part of Unix's advantage is
> >modularity. If something breaks, it and only it can be fixed or replaced. This
> >would mix two totally unrelated things.
> 
> How do you think MS word users would react if Microsoft told them that they
> could not print their documents. Or that they would have to pay for a third
> party accessory to print.

What does that have to do with anything? How do you think MS word users
would react if Microsoft told them that printing wasn't in the GUI
modules anymore, it was now in the printing modules?  Does the word
"Huh?" come to mind? It's not an end-user effect, it's a internal thing.

And what if MS word users were told that they can only print at 640X480
resolution? That's the equivelent of telling Unix users they can only
print under X.
 
> >Not saying that printing doesn't need some work in Unix (I still haven't got
> >my printer working under Linux), but it's not X's job to do printing, and
> >there is no reason it should be.
> 
> Which is why Microsoft still makes a lot of money.

Which is Windows major technical problem - lack of appropriate
seperation and modularity. It's nice to have a system where, if the
printing demons die, nothing but printing is affected. 

I admit there's a problem, but you haven't shown any reasons why there
should be a connection between X and printing.

Re: Java momentum slowing ?

[Andrzej Lewandowski]

mitch@nospam wrote:
> 
> In article <37289A2C.4B78EB60@rocketmail.com>, Corey says...
> >
> 
> >I think it will still be popular, but it seems to have split
> >into two camps now - J++ / COM  and Java / JavaBeans.
> 
> J++ is dead. you have not been keep up with the new. MS killed J++.
> no body was using it.
> 
> COM (or it it COM+ or is it DCOM these daya?) only runs on windows.
> only an idiot will commit their software to run on only one platform
> using closed technology, and we know the world is full idiots, else
> windows will not be used as much as it is.
> 
> Java and Corba are getting more popular, offcourse, there will always be
> some managers, those bean counters types, with no clue about software, who
> will insist on ordering M$ products for they know no better, I've worked
> for one such moron.
> 


I am this idiot. I run my software on NT, I dont care about UNIX and
other stuff. Not because I like MS that much, but this is what 
customers want.

A.L.

Re: Java momentum slowing ?

[Tarjei Tj�stheim Jensen]

David Starner wrote

>And what if MS word users were told that they can only print at 640X480
>resolution? That's the equivelent of telling Unix users they can only
>print under X.


Me thinks you are confused. Try removing the word "only" from the last sentence
above.

>I admit there's a problem, but you haven't shown any reasons why there
>should be a connection between X and printing.


I rest my case.


Greetings,

Re: Java momentum slowing ?

[bob]

In article <7gpiri$1dt2@ftp.kvaerner.com>, "Tarjei says...
 
>
>>I admit there's a problem, but you haven't shown any reasons why there
>>should be a connection between X and printing.
>

>
>I rest my case.
>

There are now GUI way's to config printing on Linux. It is as simple 
as pointing and clicking.

But there is a more fundemental issues here when it comes to Unix vs. Windows,
which this news group is not the right place to discuess, so I'll stop.

As far as Java momentum slowing: NO. It is not. It is increasing. More and
more new software is being done in Java every day. The new Oracle 8i db many
consider to be just a huge Java engine, with a new filesystem written
entirly in Java based one what I read.

Thanks God, C and C++ are dying and going away (at the application level). 

At least the world in moving to a new language, too bad it is not Ada, 
but at least, for the software world who is interested in a better 
quality software, we are lucky that the new language the world has 
decided to love, is really a better quality one than the earlier ones.

And now, since the world seems to have awaken that there something out there
other than C and C++, may be Ada now can come in and say, "Look at me too!".

Bob 

Re: Java momentum slowing ?

[David Starner]

"Tarjei Tj�stheim Jensen" wrote:
> 
> David Starner wrote
> 
> >And what if MS word users were told that they can only print at 640X480
> >resolution? That's the equivelent of telling Unix users they can only
> >print under X.
> 
> Me thinks you are confused. Try removing the word "only" from the last sentence
> above.
You can print from X, the same way you print anywhere else. Most good
programs have a print option that converts it to postscript and sends it
to lpr. If you made X handle printing, then at best printing from the
console and from X would be a mess. At worst, you couldn't mix them at
all.

On some boxes, like my school's Sun, a.cs.okstate.edu, most people
aren't running X. So adding printing to X would help few at best, and
possibly hurt those people as they would have to worry about the
interactions of their system with the console printing system.
> 
> >I admit there's a problem, but you haven't shown any reasons why there
> >should be a connection between X and printing.
> 
> I rest my case.

Without ever showing that X should handle printing.

Re: Java momentum slowing ?

[Matthew Whiting]

Richard D Riehle wrote:
> 
> The Java momemtum is certainly not slowing.  I just returned
> from the STC conference in Salt Lake City.  This conference once
> had an Ada track.  No more.  The Ada track has been replaced by
> a Java track.

Maybe Ada should be renamed ... Java++  :-)

Matt

Re: Renaming Ada (Re: Java momentum slowing ?)

[bob]

In article <37323E13.A5A2AFA7@Botton.com>, David says...
>
 
>Java++ would just get you an entrance to the court room where you would
>have Sun on one side for the Java name and MS on the other for the ++.
>
 
good point. someone told me also that MS has a pending patent on "++".

How about we rename Ada to "COOL" before MS graps that name also? I
hear MS is working on this new language and they want to call it
COOL, or COOL++.

It seems that all the good and cool and zappy and modern names for computer
languages has been taken allready by different languages.

I really think that the biggest blunder the Ada95 team did was not 
to rename Ada to something completely different and wild than the
generic name 'Ada' (for crying out loud, 'Ada' is a name of a person
who lived more than 130 years ago!, it is not a modern name! what is so
cool about a name such as 'Ada'? it has the letter 'a' repated 2 times
in a 3 letters name!! not only that, there is no 'X' or 'Z' or '+' or 'C' or
'J' or any of the eye catching letters in the name.

They had the chance to do it then, and they blew it, and we are now
all paying for that positional mistake and will continue to do so.

Bob

Re: Java momentum slowing ?

[Richard D Riehle]

The Java momemtum is certainly not slowing.  I just returned
from the STC conference in Salt Lake City.  This conference once
had an Ada track.  No more.  The Ada track has been replaced by
a Java track.  

Richard Riehle
richard@adaworks.com
http://www.adaworks.com

Renaming Ada (Re: Java momentum slowing ?)

[David Botton]

If there was a Delphi style system put together with Ada in its belly, I
think that a name change like Pascal -> Delphi, might work.

Java++ would just get you an entrance to the court room where you would
have Sun on one side for the Java name and MS on the other for the ++.

David Botton


Matthew Whiting wrote:
> 
> Maybe Ada should be renamed ... Java++  :-)

Re: Renaming Ada (Re: Java momentum slowing ?)

[carlislemc]

In article <37323E13.A5A2AFA7@Botton.com>,
  David Botton <David@Botton.com> wrote:
> Java++ would just get you an entrance to the court room where you would
> have Sun on one side for the Java name and MS on the other for the ++.

Wouldn't Bjarne Stroustrop have a better claim on ++?  After all, MS surely
didn't invent C++.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Renaming Ada (Re: Java momentum slowing ?)

[P.S. Norby]

bob@nospam wrote:
> 
> In article <37323E13.A5A2AFA7@Botton.com>, David says...
> >
> 
> >Java++ would just get you an entrance to the court room where you would
> >have Sun on one side for the Java name and MS on the other for the ++.
> >
> 
> good point. someone told me also that MS has a pending patent on "++".
> 

Does that mean all the C programmers will have to go back and remove the
++ operator from all their code?  ;-)

> How about we rename Ada to "COOL" before MS graps that name also? I
> hear MS is working on this new language and they want to call it
> COOL, or COOL++.
> 
> It seems that all the good and cool and zappy and modern names for computer
> languages has been taken allready by different languages.
> 
> I really think that the biggest blunder the Ada95 team did was not
> to rename Ada to something completely different and wild than the
> generic name 'Ada' (for crying out loud, 'Ada' is a name of a person
> who lived more than 130 years ago!, it is not a modern name! what is so
> cool about a name such as 'Ada'? it has the letter 'a' repated 2 times
> in a 3 letters name!! not only that, there is no 'X' or 'Z' or '+' or 'C' or
> 'J' or any of the eye catching letters in the name.
> 
> They had the chance to do it then, and they blew it, and we are now
> all paying for that positional mistake and will continue to do so.
> 
> Bob

If you do a search on the net for Ada software jobs, you will also turn
up
  1) jobs involving radar
  2) jobs in Ada, OK
  3) jobs where the employer complies with the Americans with
Disabilities Act

How about "Zyzyx"?  "Dweezl"?  Some one suggested "Tucker".  I like
that.  (Saw the movie.)

-- 
P.S. Norby

 "Software engineers are, in many ways, similar to normal people"

        --  Scott Adams

"No excuses.  No embarrassment.  No apologies...
 Ada -- the most trusted and powerful programming language
 on earth, or in space." -- S. Tucker Taft
 
\\\    \\\    \\\    \\\    \\\    \\\    \\\    \\\    \\\ 
( :)   ( :)   ( :)   ( :)   ( :)   ( :)   ( :)   ( :)   ( :)
///    ///    ///    ///    ///    ///    ///    ///    /// 
(Speaking only for myself)

Re: Renaming Ada (Re: Java momentum slowing ?)

[Ada2001]

I named my four year old daughter Ada in honor of my favorite programming
language and because Ada is a nice, traditional name. She likes her name very
much and would be very much opposed to being renamed :-).

F. Britt Snodgrass

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Renaming Ada (Re: Java momentum slowing ?)

[dennison]

In article <7gv4ve$gfk$1@nnrp1.deja.com>,
  Ada2001 <ada2001@my-dejanews.com> wrote:
> I named my four year old daughter Ada in honor of my favorite programming
> language and because Ada is a nice, traditional name. She likes her name very
> much and would be very much opposed to being renamed :-).
>
> F. Britt Snodgrass

I don't think I'd want to have my name changed to "Dweezl Snodgrass" either.


(sorry for the name-based jest. I had to, uh, they were pointing a gun at me,
yeah,  that's it...)

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Renaming Ada (Re: Java momentum slowing ?)

[dennison]

In article <7gtvbd$1qo5@drn.newsguy.com>,
  bob@nospam wrote:
> In article <37323E13.A5A2AFA7@Botton.com>, David says...
> >
>
> >Java++ would just get you an entrance to the court room where you would
> >have Sun on one side for the Java name and MS on the other for the ++.
> >
>
> good point. someone told me also that MS has a pending patent on "++".
>
> How about we rename Ada to "COOL" before MS graps that name also? I
> hear MS is working on this new language and they want to call it
> COOL, or COOL++.

"COOL" is copyrighted by Sterling software for their line of software tools.
(The old ObjectTeam is now COOL:JEX) I suspect Microsoft could afford to buy
them or destroy them in court if they want the name, but I don't think the
entire Ada community could do that.

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Renaming Ada (Re: Java momentum slowing ?)

[Corey Ashford]

bob@nospam wrote:
[snip]
> I really think that the biggest blunder the Ada95 team did was not
> to rename Ada to something completely different and wild than the
> generic name 'Ada' (for crying out loud, 'Ada' is a name of a person
> who lived more than 130 years ago!, it is not a modern name! what is so
> cool about a name such as 'Ada'? it has the letter 'a' repated 2 times
> in a 3 letters name!! not only that, there is no 'X' or 'Z' or '+' or 'C' or
> 'J' or any of the eye catching letters in the name.
> 
> They had the chance to do it then, and they blew it, and we are now
> all paying for that positional mistake and will continue to do so.


The problem is that you'd get this sort of comment: "Oh, Z_Edge,  yeah...
that's just a new, more bloated version of that bloated, designed-by-commitee,
military language Ada, hiding under a Gen-X name.  Forget about it."

I think the real problem is perception.  I've found a number of programmers who
can't stand Ada.  They think it's huge and cumbersome and hard to program.
Have they ever programmed in it? No.  Have they ever looked at it seriously?
No.  How do they "know" what they know about it?  Hearsay.

- Corey

Re: Java momentum slowing ?

[Martin C. Carlisle]

In article <37333a43.181207462@news.mindspring.com>,
Bob Munck <munck@Mill-Creek-Systems.com> wrote:
>On Thu, 06 May 1999 17:42:37 GMT, Richard D Riehle
><laoxhai@ix.netcom.com> wrote:
>I've heard that DISA is mandating the use of Java for all new code.

This sounds preposterous.  After all the grief the DoD got over the last
language mandate, I doubt they'd try again any time soon.  In any event, you
can rest assured no such rumor has reached the Air Force Academy, where we
do most of our work and almost all of our teaching in Ada.

--Martin

-- 
Martin C. Carlisle, Asst Prof of Computer Science, US Air Force Academy
carlislem@acm.org, http://www.usafa.af.mil/dfcs/bios/carlisle.html
DISCLAIMER:  This content in no way reflects the opinions, standards or 
policy of the US Air Force Academy or the United States Government.

Re: Renaming Ada (Re: Java momentum slowing ?)

[Ronald Cole]

carlislemc@my-dejanews.com writes:
> In article <37323E13.A5A2AFA7@Botton.com>,
>   David Botton <David@Botton.com> wrote:
> > Java++ would just get you an entrance to the court room where you would
> > have Sun on one side for the Java name and MS on the other for the ++.
> 
> Wouldn't Bjarne Stroustrop have a better claim on ++?  After all, MS surely
> didn't invent C++.

Now that Bill Gates has boarded the Deathstar, all bets are off...  ;)

-- 
Forte International, P.O. Box 1412, Ridgecrest, CA  93556-1412
Ronald Cole <ronald@forte-intl.com>      Phone: (760) 499-9142
President, CEO                             Fax: (760) 499-9152
My PGP fingerprint: 15 6E C7 91 5F AF 17 C4  24 93 CB 6B EB 38 B5 E5

Re: Renaming Ada (Re: Java momentum slowing ?)

[Matthew Whiting]

David Botton wrote:
> 
> If there was a Delphi style system put together with Ada in its belly, I
> think that a name change like Pascal -> Delphi, might work.
> 
> Java++ would just get you an entrance to the court room where you would
> have Sun on one side for the Java name and MS on the other for the ++.

True ... and you couldn't afford to buy that kind of publicity and free
advertising!  :-)

Matt

Re: Java momentum slowing ?

[Bob Munck]

On Thu, 06 May 1999 17:42:37 GMT, Richard D Riehle
<laoxhai@ix.netcom.com> wrote:

>... the STC conference in Salt Lake City ... once
>had an Ada track.  No more.  The Ada track has been replaced by
>a Java track. 

I've heard that DISA is mandating the use of Java for all new code.

Bob Munck

Re: Java momentum slowing ?

[steve]

I am finding that Ada is actually more portable than Java.

just go to http://www.alphaworks.ibm.com

see all the Java stuff to download. most run on windoz only.
and this is by IBM itself, who is the biggest supporter for
Java these days.

example:

http://www.alphaworks.ibm.com/tech/cfparse

You'll see the platform is windoz. but it is written in Java.

I've worked on a place where coding was done in Java, but
we all were required to only use windows for development,
since the application was supposed to run on windows only, even
though we used only Java code. 

Java portability is a big spam. Ada is much much more portable
than Java will ever be. 

Steve

 

Re: Java momentum slowing ?

[bill]

In article <7h399g$4nb@drn.newsguy.com>, steve says...
>
>
>I am finding that Ada is actually more portable than Java.
>
>just go to http://www.alphaworks.ibm.com
>
>see all the Java stuff to download. most run on windoz only.
>and this is by IBM itself, who is the biggest supporter for
>Java these days.
>
>example:
>
>http://www.alphaworks.ibm.com/tech/cfparse
>
>You'll see the platform is windoz. but it is written in Java.
  
I think this is just a packaging issue. IBM 'alpha works' programmers
are too lazy to use generic packaging, they seem to use windows specific
ways to install the software which will not work on Unix systems.

It does seem kind'a stupied of IBM to write something in Java, and then
to have an installer that only works on windows. (but what else 
would you expect from big companies).

Bill.

Re: Java momentum slowing ?

[Pascal F. Martin]

In article <7h399g$4nb@drn.newsguy.com>,
	steve  <steve@newsguy.com> writes:
> 
> I am finding that Ada is actually more portable than Java.
> [...]
> example:
> 
> http://www.alphaworks.ibm.com/tech/cfparse
> 
> You'll see the platform is windoz. but it is written in Java.
>  [...]
> Java portability is a big spam. Ada is much much more portable
> than Java will ever be. 

I would not start a flame war here, but do you know _any_ Ada
standard for Graphical User Interface ? 8-{

Most of the time, people use bindings to X Windows (usually custom),
which does not work on 95 or NT, or binding to win32, which does not
work on UNIX.

I have used a few Java (and written several Tcl/Tk) programs with
a GUI which worked on both systems. I have, so far, never seen such
a program from the Ada side.

Portability is always a touchy issue: beyond the language, there
is the environment, and it seems people confuse both. If a library
exists in one environment only, using this library makes your program
non portable. It does not matter how portable your language is.

BTW, Ada is very poor when it comes to user interfaces tools:
either you redo it yourself (hardware dependencies ?) or you use
bindings for a specific OS (or OS family). At least Java is trying
to solve this issue.

-- 

Pascal F. Martin.

Re: Java momentum slowing ?

[Pascal F. Martin]

Thanks for the information.

Most of the links you refer to are listed as "unstable", "usable" or
"in development": too bad.. In addition, none of these bindings is part of
the Ada standard: there is simply _no_ Ada standard here, while Java does
have a standard (albeit quite unstable: AWT, then SWING, then what else ?).

BTW, Java is not a scripting language: you need to use a compiler. Maybe
you confused with javascript ?

The main points I wanted to make is that the argument "the language X is
more portable than Y" is invalid without a distinction between language
and environment, and that people are comparing the Ada language with another
language + environment: apple with orange. If a valid comparison is to be
made, it should compare similar features.

As a conclusion, having used Ada for 7 years, I am still surprised how
immature are some language zealots. The fact that Ada is a good language
does not mean it is problem-free, or a silver bullet. I still remember the
silly thread "The Ariane 5 rocket would have not crashed if it had been
programmed in PL/1". A language is just a tool. No program is going to be
more robust or portable than its design allows it to be (intentionally
or not: I have seen easily portable designs that was not intended to be).

PS: Is anyone knows any language where I can open "foo" and "Foo" with
the same result on any OS ? The file name syntax & rules is still one of
the sticky portability problem. I have had a few bugs because of problems
like that.

In article <3735DEB0.5D2FDB05@botton.com>,
	David Botton <David@Botton.com> writes:
> I think you need to see:
> 
> TASH - TCL/TK bindings for Ada
> 
> http://tash.calspan.com/
> 
> [...]

-- 

Pascal F. Martin.

Re: Java momentum slowing ?

[bob]

In article <CrmZ2.2549$2j3.4026@clnws01.we.mediaone.net>,
pmartin@mail.earthlink.net says...

>
>Thanks for the information.
>
>Most of the links you refer to are listed as "unstable", "usable" or
>"in development": too bad.. In addition, none of these bindings is part of
>the Ada standard: there is simply _no_ Ada standard here, while Java does
>have a standard (albeit quite unstable: AWT, then SWING, then what else ?).
>

I have no idea what you mean but 'quite unstable' in the above.

Pepole are writing now very advanced GUI's in Java, more advanced
in some cases than what you get with your platform specific, cryptic,
tie me down to one platform forever, MFC and VB type of applications.
 
And you are confusing things. SWING use AWT for the native components,
but swing introduces many light-weight components that are written all
in java, with no native counterpart to them. think of it as Swing 
adding more things on top of AWT, The Java AWT and SWING API's are fixed 
and people are using them all the time and are well documented.

And I expect in Next JDK release to see more GUI stuff added.

There is nothing new here, people want to improve things, and new technologies
will show up building on what was done before. If you call this unstability,
then I am not sure what you are doing in the software business, this sort
of thing happens all the time. A technology that does not change with
time, and continue to evolve to meet the changing needs that arise, 
will soon die.

>BTW, Java is not a scripting language: you need to use a compiler. Maybe
>you confused with javascript ?
>

Yes, he must have been, alot of people confuse the 2 things.

>The main points I wanted to make is that the argument "the language X is
>more portable than Y" is invalid without a distinction between language
>and environment, and that people are comparing the Ada language with another
>language + environment: apple with orange. 

But you are missing something. The whole point of having a standard library
that comes with the language, is to be the interface between the application
written in the language and the environemnt.

For example, in Ada, you use the Ada IO package to open a file in the 
'environemnt'.

in Java, you use the Java IO package to open the same file.

no difference. 

Each library for different platforms is implemented for that specific
platform, sheilding the application from the environment changes.

The difference is that in Java, the shield is much more advanced.

The Java standard packages are more rich and exhtensive and are
growing all the time, while the Ada ones are small, limited, and frozen.

For example, Sun is adding a new package to manage XML documents using
the SAX 1.0 API specs and DOM specs for XML data object mapping, this means,
I can unpack my JDK tar file, and find in it all the packages I need
to work with XML documents, in addition to everything else allready there. 

Check the Java Collection framework in JDK 2.0, an STL type collection
of usefull classes that many people are starting to use productively.

The Ada library has no such thing. So, when I program in Java, I am 
much more efficient and productive than in Ada, not becuase the Java 
language is better (although the Java language is very good also), but 
becuase the Java library is much more advanced.

regards,
Bob

Re: Java momentum slowing ?

[bill]

In article <373612A6.3752CC19@Botton.com>, David says...
 
>
>Java does not have a standard. It has a vendor.
>

This is funny. You are the same one who uses COM and DCOM and 
ActiveX, and all that windows specific stuff, dont't you? show us the
ISO standard for any of those. Show us the ISO standard for VB or Delphi
or Perl pr Pyhton or any of the zillion other technologies being used
everyday by millions of programmers. 

Just becuase there is no Java ISO or ANSI standard means very little.
It is a matter of time before this happens. Meanwhile, every one out
there uses the current Java packages just fine with no problem, as
long as one stays from any Java stuff from M$ one will be safe.

>You compile Python too. Python compiles to a byte code and then
>executes. You can even distribute the byte code just like Java and just
>run it :)
>

Yes, there are many things like these around. Isn't there also some sort
of Perl to java compiler in the works? The bottom line is that although
these are fun and interesting projects, the industry will by far stay with
using the Java language to generate byte code from.

>
>Java is not JDK. BTW you can compile Ada to bytecode and use all the JDK
>facilities just like Java. (See some examples on
>http://www.adapower.com/os)
>

Ada to Java technology has not taken of at all. It will interesting to
see how jGNAT (when it comes out to public) will change that. I think
it will, but to what extent, remains to be seen.  Again, I think the
industry as a whole will stick to Java as the source language to the JVM.

>
>Ada is a tool that when used properly leads to better implementations.
>If your design smells so does your code. The key is that Ada helps
>prevent human error unlike the C syntax of C/C++/Java, is easier reading
>and easier to maintain.
>

You can say that about C/C++, but you are really hand waving here when you
try to apply the same old tactics to Java.

Java has fixed many of those problems in C/C++.

Java is much better readable than C++ for example, Java has very strong
time checking, run-time checking, it is very well defined language.

It lacks some of the more advanced features that Ada had such as generics
and child packages, and it also lack some basic ones such as unsigned data
types and enumeration data types and sub ranges (very good things to
have) But other than that, it is pretty much a good language, and on
the other hand, it has things that Ada does not have, such as 
exception specification as part of the method, GC, dynamic loading
of classes, interfaces, javadoc, and I think a better and simpler
object model than Ada's, and is much more OO language than Ada is.
 
At the end of the day, one uses whatever language is best for the task
at hand.
 
>David Botton

Bill

Re: Java momentum slowing ?

[Steve]

In article <37365163.1AB89036@Botton.com>, David says...
 
>
>So there is no "Java Standard" as you wrote, not even a defacto standard
>among vendors (take a look at J++).
 
the java packages that come in the JDK are decfacto standard, the Java 
language as defined by the Sun 'Java language' reference book is defacto
standard, and so is the JVM.

No one I know uses J++, not even on windows. J++ was another attempt
by M$ to stop Java as it knew Java is making windows irrelevent. This
attempt had failed by all acounts. And J++ is effectively dead.
 
Steve.

Re: Java momentum slowing ?

[Simon Wright]

bill@nospam writes:

> In article <7h399g$4nb@drn.newsguy.com>, steve says...
> >example:
> >
> >http://www.alphaworks.ibm.com/tech/cfparse
> >
> >You'll see the platform is windoz. but it is written in Java.
>   
> I think this is just a packaging issue. IBM 'alpha works' programmers
> are too lazy to use generic packaging, they seem to use windows specific
> ways to install the software which will not work on Unix systems.

I'm just doing some Java work, _endless_ trouble with methods being
available/deprecated/not available in various browsers -- and then the
browser locks itself up because I suspended a thread in an applet --
yecch. Seems it's going to take a lot of effort to make sure this
thing works in the browser I'll be actually using. At least it doesn't
have to work in every browser in the world (yet :-)

Re: Java momentum slowing ?

[David Botton]

I think you need to see:

TASH - TCL/TK bindings for Ada

http://tash.calspan.com/


RAPID - TASH GUI Builder

ftp://ftp.usafa.af.mil/pub/dfcs/carlisle/usafa/rapid/index.html

You also should see some work in progress:

GTKAda
http://ada.eu.org/gtkada/

GTKAda on Win32
http://www.mv.com/users/jcreem/programming/win32gtk.htm

and I am also working on a cross platform GUI project that will also
support ActiveX on Win32.

BTW The reason many Java apps are marked for Win32 only is not because
they are lazy about install scripts, but rather because most Java apps
that do anything particularly useful use some sort of native code. I
have written in the last few years many Java apps, many of them made use
of COM on Win32 and thus non-portable.

Java is one of the best scripting languages I have ever used! I highly
recommend it whenever perl is too slow and the app is trivial :)

David Botton


"Pascal F. Martin" wrote:
> 
> In article <7h399g$4nb@drn.newsguy.com>,
>         steve  <steve@newsguy.com> writes:
> >
> > I am finding that Ada is actually more portable than Java.
> > [...]
> > example:
> >
> > http://www.alphaworks.ibm.com/tech/cfparse
> >
> > You'll see the platform is windoz. but it is written in Java.
> >  [...]
> > Java portability is a big spam. Ada is much much more portable
> > than Java will ever be.
> 
> I would not start a flame war here, but do you know _any_ Ada
> standard for Graphical User Interface ? 8-{
> 
> Most of the time, people use bindings to X Windows (usually custom),
> which does not work on 95 or NT, or binding to win32, which does not
> work on UNIX.
> 
> I have used a few Java (and written several Tcl/Tk) programs with
> a GUI which worked on both systems. I have, so far, never seen such
> a program from the Ada side.
> 
> Portability is always a touchy issue: beyond the language, there
> is the environment, and it seems people confuse both. If a library
> exists in one environment only, using this library makes your program
> non portable. It does not matter how portable your language is.
> 
> BTW, Ada is very poor when it comes to user interfaces tools:
> either you redo it yourself (hardware dependencies ?) or you use
> bindings for a specific OS (or OS family). At least Java is trying
> to solve this issue.
> 
> --
> 
> Pascal F. Martin.

Re: Java momentum slowing ?

[David Botton]

"Pascal F. Martin" wrote:
 while Java does
> have a standard (albeit quite unstable: AWT, then SWING, then what else ?).
> 

Java does not have a standard. It has a vendor.

> BTW, Java is not a scripting language: you need to use a compiler. Maybe
> you confused with javascript ?

You compile Python too. Python compiles to a byte code and then
executes. You can even distribute the byte code just like Java and just
run it :)

No, I am not confused, realistic.

> 
> The main points I wanted to make is that the argument "the language X is
> more portable than Y" is invalid without a distinction between language
> and environment, and that people are comparing the Ada language with another
> language + environment: apple with orange. If a valid comparison is to be
> made, it should compare similar features.

Java is not JDK. BTW you can compile Ada to bytecode and use all the JDK
facilities just like Java. (See some examples on
http://www.adapower.com/os)

> 
> As a conclusion, having used Ada for 7 years, I am still surprised how
> immature are some language zealots. The fact that Ada is a good language
> does not mean it is problem-free, or a silver bullet. I still remember the
> silly thread "The Ariane 5 rocket would have not crashed if it had been
> programmed in PL/1". A language is just a tool. No program is going to be
> more robust or portable than its design allows it to be (intentionally
> or not: I have seen easily portable designs that was not intended to be).

Ada is a tool that when used properly leads to better implementations.
If your design smells so does your code. The key is that Ada helps
prevent human error unlike the C syntax of C/C++/Java, is easier reading
and easier to maintain.

David Botton

Re: Java momentum slowing ?

[David Botton]

So how does this make Java better? If Ada is compiled to byte code and
has use of all the Java APIs, JDKs, etc.

Perhaps the debate should be is the JVM environment / script engine
better then say the native machine / OS combination and direct bindings.

BTW, The difference is you could write Java, its VM, all its libraries
and GUIs in Ada and get a better product then if you wrote Ada  in Java!
(Of course you could write Ada or Java in Perl too, another great
scripting language just like Java :)

David Botton


bob@nospam wrote:
> 
> Pepole are writing now very advanced GUI's in Java
> 
> 
> But you are missing something. The whole point of having a standard library
> that comes with the language, is to be the interface between the application
> written in the language and the environemnt.
> 

You can say that on the JVM Ada comes with them too!
 
> The Ada library has no such thing. So, when I program in Java, I am
> much more efficient and productive than in Ada, not becuase the Java
> language is better (although the Java language is very good also), but
> becuase the Java library is much more advanced.

If you are more productive in Java (including debug and testing time) it
is because you have studied and collected the available resources and
practiced. Doing the same for Ada will yield similar results.

Re: Java momentum slowing ?

[David Botton]

bill@nospam wrote:
> 
> In article <373612A6.3752CC19@Botton.com>, David says...
> 
> >
> >Java does not have a standard. It has a vendor.
> >
> 
> This is funny. You are the same one who uses COM and DCOM and
> ActiveX, and all that windows specific stuff, dont't you? show us the
> ISO standard for any of those. Show us the ISO standard for VB or Delphi
> or Perl pr Pyhton or any of the zillion other technologies being used
> everyday by millions of programmers.

That is just the point since they all have a vendor and no standard
things change under the cover all the time and there is nothing for them
to answer to.

So there is no "Java Standard" as you wrote, not even a defacto standard
among vendors (take a look at J++).

David Botton


---------

Just as an example (one of many possible) when Microsoft specified the
method ITypeInfo::GetRefTypeOfImplType they specified an Unsigned
Integer as an index. Yet since things have changed in the COM world (of
course this is not supposed to happen) you need to pass in a negative
index to get the TKIND_INTERFACE out of a TKIND_DISPATCH that is a dual
interface (something MS added after the original COM/OLE specs were
drawn up to support faster access to OLE objects from early binding
languages and tools.) Which means I have to write code to go against the
MS specs and bindings:

   -----------------------------
   -- Get_Interface_Reference --
   -----------------------------

   procedure Get_Interface_Reference(Source : in Object;
                                     Index  : Integer;
                                     In_To  : in out Object)
   is
      function To_UINT is
         new Ada.Unchecked_Conversion (Integer, Win32.UINT);
      
      Reference : aliased Win32.OleAuto.HREFTYPE;
   begin
      Check_HR(
        
Source.ITypeInfo_Pointer.lpVtbl.GetRefTypeOfImplType(Source.ITypeInfo_Pointer,
         To_UINT(Index),
         Reference'Unchecked_Access ) );

      Get_Reference(Source, Reference, In_To);

   end Get_Interface_Reference;

Re: Renaming Ada (Re: Java momentum slowing ?)

[Marin David Condic]

Nick Roberts wrote:
> 
> databases.  Two days later, they sent a written confirmation that they could
> not supply me with a compiler for the "Aider" language.  Shucks.
> 
Having spent a two year exile in New Jersey, I recognize that as the
"Joysey" pronunciation of the word "Ada". Maybe it just didn't translate
well across borders?

MDC
-- 
Marin David Condic
Real Time & Embedded Systems, Propulsion Systems Analysis
United Technologies, Pratt & Whitney, Large Military Engines
M/S 731-95, P.O.B. 109600, West Palm Beach, FL, 33410-9600
***To reply, remove "bogon" from the domain name.***

Visit my web page at: http://www.flipag.net/mcondic

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On Thu, 06 May 1999 17:42:37 GMT, Richard D Riehle
<laoxhai@ix.netcom.com> wrote:

>The Java momemtum is certainly not slowing.  I just returned
>from the STC conference in Salt Lake City.  This conference once
>had an Ada track.  No more.  The Ada track has been replaced by
>a Java track.  
>
>Richard Riehle
>richard@adaworks.com
>http://www.adaworks.com

I'm just doing a comparison with what this momentum was 2 years ago.  

Today, i don't hear no more of NC's, thin clients, don't see much Java
applets on the Web. 

 Java is just another proprietary technology. Don't see where the
revolution is except for SUN or IBM.

From a customer point of view, i don't see the advantage to replace a
monopoly by one another. And, as Windows still represents more than 80
% of the desktop OS, you would have to run Java apps on this platform
for a while. So, you rely on M$ for the OS and on SUN$ for the JVM !
If you have much money to spend, why not.

Enter Linux. It's Free, it's Fast, it's Simple, it's Customizable,
it's not Proprietary. Not bad. 

 
  

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On 8 May 1999 23:17:20 -0700, steve  <steve@newsguy.com> wrote:

>
>see all the Java stuff to download. most run on windoz only.
>and this is by IBM itself, who is the biggest supporter for
>Java these days.
>
>example:
>
>http://www.alphaworks.ibm.com/tech/cfparse
>
>You'll see the platform is windoz. but it is written in Java.
>
>I've worked on a place where coding was done in Java, but
>we all were required to only use windows for development,
>since the application was supposed to run on windows only, even
>though we used only Java code. 
>
>Java portability is a big spam. Ada is much much more portable
>than Java will ever be. 

Don't know if Ada is much more portable, but running Java on Windows
is the biggest stupidity i've ever seen ! 

A slow system and a slow language. What a deal !

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On Mon, 10 May 1999 01:24:22 GMT, pmartin@mail.earthlink.net (Pascal
F. Martin) wrote:

>.. and Ada has no GUI interface defined for it

But, i don't see why it would have one ? 

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On 8 May 1999 23:17:20 -0700, steve  <steve@newsguy.com> wrote:

>I've worked on a place where coding was done in Java, but
>we all were required to only use windows for development,
>since the application was supposed to run on windows only, even
>though we used only Java code. 

And don't forget to code and test for the right JVM on Windows. The
one from SUN, the one from MSoft and the one from IBM !!!! The
probability that they are compatible is largely < 1 ... 

Good luck ! 
 

Re: Renaming Ada (Re: Java momentum slowing ?)

[Jean-Marten Marchi]

On Thu, 06 May 1999 21:12:51 -0400, David Botton <David@Botton.com>
wrote:

>If there was a Delphi style system put together with Ada in its belly, I
>think that a name change like Pascal -> Delphi, might work.
>
>Java++ would just get you an entrance to the court room where you would
>have Sun on one side for the Java name and MS on the other for the ++.
>
>David Botton
>
>
>Matthew Whiting wrote:
>> 
>> Maybe Ada should be renamed ... Java++  :-)

I don't think naming it with Java something is a very good idea since
Java will fail. Avoid Java as much as you could. 

Maybe GNAT ++ ? so you could surf on the Linux Wave. And this wave is
the good one. 

 

Re: Java momentum slowing ?

[Pascal Obry]

Just to add that I left out in my previous message, the possible choices 
the differents JVM (Microsoft, SUN, IBM).

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--| http://ourworld.compuserve.com/homepages/pascal_obry
--|
--| "The best way to travel is by means of imagination"

Re: Java momentum slowing ?

[Pascal F. Martin]

In article <7h55i0$2ro3@drn.newsguy.com>,
	bob@nospam writes:
> I have no idea what you mean but 'quite unstable' in the above.

I have a colleague who have had trouble with different versions of Java:
he worked first with one that did had SWING (Java 1.2, I think), then tried
one another which did not (Java 1.1), to find himself forced to upgrade
with a beta product (now released version) which was still Java 1.1, but
has a slightly modified SWING, because of the 1.1 AWT. All that in a short
amount of time.

This is almost as complex as identifying which subset of Ada95 has been
implemented by each Ada vendor :-).

> tie me down to one platform forever, MFC and VB type of applications.

I hate window: I have no problem bashing these. However we tried 3 Java
environments, with each enough performance problems we decided Java is not
ready yet for our core application (redraw of the SWING table widget is
enlightening).

To be honest, we are still excited about Java; it gives us the one feature
we really want: we can load dialogs dynamically. Because our application
contains many dialogs, and we have to be able to adapt or add dialogs for
each customer, this is a real advantage. We also can upgrade live, which
is also important because our application is kind of 24/7.

> And you are confusing things. SWING use AWT for the native components,

I know that. It is just that when you buy a product, you have to be careful
(see above). This is probably the least portable aspect of Java.

> But you are missing something. The whole point of having a standard library
> that comes with the language, is to be the interface between the application
> written in the language and the environemnt.

.. and Ada has no GUI interface defined for it. Everyone comes with his
own proposal: Tk, GTK, motif or win32 bindings, etc.. all differents, with
little hope of mixing them in the same program (= integration nightmare).
I like the freedom that comes with all these different interfaces, but at
some point, one need to settle for a common one.

Let be serious: there will never be any standard to cover all environments.
However, I don't know that many commercial project on Windows or UNIX
that has no GUI. It is now the standard user interface, period. I hope GTK
will fill the need (I use Gnome at home, you see.. :).

> while the Ada ones are small, limited, and frozen.

The Ada communauty seems to have difficulties to define new standards:
the world around changes fast (GUIs, CGI, XML,..), and the Ada speachs
seem to have been written 5 years ago. Even the printers we buy today
have a WEB interface ! I am sure this is possible with Ada; I am also
sure this is more expensive, only because you have to redo it.


As a personnal conclusion, I have used C and Ada intensively; I start
looking at Java. So far, the most portable programs I have seen was well
designed C programs. Ada programs may be difficult to port when bindings
are used: implicit assumptions on C definitions, availability of all
bindings on each target, etc.. Portability is hard work, not a given.

This will close my contribution to this thread. Thank you for listening.

-- 

Pascal F. Martin.

Re: Java momentum slowing ?

[Pascal Obry]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 914 bytes --]


David Botton <David@Botton.com> a �crit dans le message :
373635DA.EA07F905@Botton.com...
> So how does this make Java better? If Ada is compiled to byte code and
> has use of all the Java APIs, JDKs, etc.
>
> Perhaps the debate should be is the JVM environment / script engine
> better then say the native machine / OS combination and direct bindings.
>

This is indeed something I completly agree. What is good in this new
technology
is the JVM (this is just a new OS, well software OS), Java is just a new
language
that is far better than C/C++ but I still like Ada better. Most of the
features added/removed
from C/C++ in Java were in Ada! This is why I see all the Ada to JVM
compilers has
very important for the Ada world. We *must* have an Ada compiler for this
new
OS as I think it is very important to have an Ada compiler under Windows as
more and more developpers are going this way...

Pascal.

Re: Java momentum slowing ?

[Pascal Obry]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1175 bytes --]


Pascal F. Martin <pmartin@mail.earthlink.net> a �crit dans le message :
MwkZ2.2415$2j3.3581@clnws01.we.mediaone.net...
> In article <7h399g$4nb@drn.newsguy.com>,
> steve  <steve@newsguy.com> writes:
> >
> > I am finding that Ada is actually more portable than Java.
> > [...]
> > example:
> >
> > http://www.alphaworks.ibm.com/tech/cfparse
> >
> > You'll see the platform is windoz. but it is written in Java.
> >  [...]
> > Java portability is a big spam. Ada is much much more portable
> > than Java will ever be.
>
> I would not start a flame war here, but do you know _any_ Ada
> standard for Graphical User Interface ? 8-{
>

No. You are right. But it depends what you are thinking when you say
"standard". Java is not a standard either. But here is not my point. For
graphical stuff we have some very good and fast growing technologies :
Tk/Tcl and GTK both are cross-plateforms are very cost effective
in terms of developpement, are very well designed.

Also, keep in mind that the most important GUI today is the browser !

In this respect Ada is (as all others languages) just fine. In CGI mode
you just have to output some HTML, XML, CSS... infos.

Pascal.

Re: Java momentum slowing ?

[Rob]

In article <7h68lv$b8l$1@cf01.edf.fr>, "Pascal says...
>
 
>We *must* have an Ada compiler for this new
>OS as I think it is very important to have an Ada compiler under Windows as
>more and more developpers are going this way...
>
 
Alot of developers and companies around we I work are starting to switch
to Linux away from windows. These are companies that write e-commerse,
ISP's, database applications, and things like that. I know of one that
replaced all their NT servers with Linux, they will save $100,000 a year
in software license fees alone.

Rob

Re: Java momentum slowing ?

[Mitch]

In article <7h694a$cnv$1@cf01.edf.fr>, "Pascal says...
 
>No. You are right. But it depends what you are thinking when you say
>"standard". Java is not a standard either. But here is not my point. For
>graphical stuff we have some very good and fast growing technologies :
>Tk/Tcl and GTK both are cross-plateforms are very cost effective
>in terms of developpement, are very well designed.
>

compared to Java SWING, they are a toy. You really can't be serious
to compare what you can do with Swing compared to tcl/tk, are you?

>Also, keep in mind that the most important GUI today is the browser !
>

really? tell us then how well you design a full GUI with pull down
menus, file dialogs, tree structures, wizards, drag/drop, 2D, 3D,
animation, etc.. with just HTML tags in a browser? 
 
>In this respect Ada is (as all others languages) just fine. In CGI mode
>you just have to output some HTML, XML, CSS... infos.
>
 
you are confusing presentation of static data, with having an interactive 
user interface. 

The ability to have powerfull GUI that is portable is very important. 
right now, there are win32, Swing, and the rest lesser able technologies 
such as gtk, qt, tcl/tk and the rest. 

Java offers the best solution as it is portable, very rich in functionality
and well defined and dcoumented, and full support in the industry from
every one, expect MS of course, since it is not in the interest of MS
for Java to successed. But Java is by all account a success so far and
continues to improve.


best,
Mitch.

Re: Renaming Ada (Re: Java momentum slowing ?)

[Nick Roberts]

My own little anecdote on this is when I once made a telephone enquiry to
some useless local software dealer, whose name shall remain a secret to
protect the guilty, as to whether they could supply me with an Ada compiler.
The reply came back that they could not, and furthermore they could not find
any reference to the language in any of their wonderful multifarious
databases.  Two days later, they sent a written confirmation that they could
not supply me with a compiler for the "Aider" language.  Shucks.

-------------------------------------
Nick Roberts
-------------------------------------

Re: Java momentum slowing ?

[David Botton]

Now you know some one who does, and I know many more :) Only J++ allows
access to COM and many other facilities, so it is the only choice for
some applications (unless you want to start writing JNI from the start.)
J++ didn't fail, it is being dumped in response to Sun.

David Botton


Steve wrote:
> 
> 
> the java packages that come in the JDK are decfacto standard, the Java
> language as defined by the Sun 'Java language' reference book is defacto
> standard, and so is the JVM.

Nope.

> 
> No one I know uses J++, not even on windows. J++ was another attempt
> by M$ to stop Java as it knew Java is making windows irrelevent. This
> attempt had failed by all acounts. And J++ is effectively dead.
> 
> Steve.

Re: Java momentum slowing ?

[Florian Weimer]

pmartin@mail.earthlink.net (Pascal F. Martin) writes:

[GUI standards]

> Let be serious: there will never be any standard to cover all environments.

There once was a project called Fresco which aimed for a cross-plattform
GUI toolkit (X11, Win32 and Mac were targeted) standardized by the X
Consortium.  Even an Ada binding was planned.

Unfortunately, the project has died.  I don't why they failed and if
every such project is dommed to failure. :-/

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On Sun, 09 May 1999 18:32:12 GMT, pmartin@mail.earthlink.net (Pascal
F. Martin) wrote:

>BTW, Ada is very poor when it comes to user interfaces tools:


I disagree ; Ada95 has solved problems that Ada83 had with GUIs,
particularly with the fact that your application is called by the
system and not the contrary as it is the case with system programming.
As far as you have the binding you can do what you want. I have even
built a very thin binding for OS/2 PM ! 

>either you redo it yourself (hardware dependencies ?) or you use
>bindings for a specific OS (or OS family).

Most often, GUIs are tied to an OS, but nothing prevents a GUI to run
on multiple OS's. In this case what you need is a binding to a GUI.

> At least Java is trying to solve this issue.
Happily, Ada doesn't try to solve any issue around GUI, nor file
system or communication protocols.
1) it gives you choice,
2) the best GUI is not invented yet,

Ada95 designers concentrated on computer language issues and they did
well. Other  issues like concurrency or distributed computing have
been put in annexes. For what i know, it was not the designers goal to
build a Corba or anything like that. 


 

Re: Java momentum slowing ?

[bob]

In article <3737AB6D.FACB718C@Botton.com>, David says...

>
>I was just thinking, is it possible to write a Java app that makes
>system calls, etc with out writing code in C/C++, Ada or another
>NON-scripting language?
>
 
I don't know why you keep calling Java a scripting language. but
if it makes you feel somehow better that is fine, but you are
wrong.

Just becuase the output of a java compiler happened to be something
called bytecode instead of something called assembler or even machine
instructions does not make Java a scripting langauage.

But to answer your question, yes of course. you use JNI (Java native
interface) to call something not written in Java. It is trivial to do
really. most people do that to call C or C++ code from Java. of course
your application become non-portable as much as it were if all the code
was written in only Java.

>
>(OK you could use TowerJ to compile to native code and finagle
>something, but that would not be "standard" java :P

Again, you are really confused. just becuase you can compile Java
to native machine instructions instead of to byte code instructions,
does not somehow changes it to become 'non-standard'.

just-in-time compilers do that all the time on the fly, the hot spot 
optimizer does that also, inside the JVM. Does that somehow make 
the source code non-standard?

They only time you java code is non-standard if you use M$ Java tools. As
long as you stay away from M$ things, you are safe. 

Bob
 

Re: Java momentum slowing ?

[David Botton]

I was just thinking, is it possible to write a Java app that makes
system calls, etc with out writing code in C/C++, Ada or another
NON-scripting language?

David Botton


(OK you could use TowerJ to compile to native code and finagle
something, but that would not be "standard" java :P

Re: Java momentum slowing ?

[Lance Kibblewhite]

bob@nospam wrote:

>But to answer your question, yes of course. you use JNI (Java native
>interface) to call something not written in Java. It is trivial to do
>really. most people do that to call C or C++ code from Java. of course
>your application become non-portable as much as it were if all the code
>was written in only Java.

Perhaps you could clarify this for me with an example.

Lets start with something simple. Say I have a routine, in a shared
library, written in some other language, that takes a single
parameter, say a 32-bit integer passed by reference, as opposed to by
value.

How exactly does JNI enable me to call that routine.

Re: Java momentum slowing ?

[David Botton]

You have to first wrap the call in a C (or Ada for that matter) function
that can be called through the JNI interface. I decided just to drop the
issue when bob@nospam came back with JNI since JNI is in native code and it
doesn't answer my question! Meaning there is now way to directly all
anything outside the VM with out first wrapping it in some native code JNI
module.

David Botton

Lance Kibblewhite wrote:

> bob@nospam wrote:
>
> >But to answer your question, yes of course. you use JNI (Java native
> >interface) to call something not written in Java. It is trivial to do
> >really. most people do that to call C or C++ code from Java. of course
> >your application become non-portable as much as it were if all the code
> >was written in only Java.
>
> Perhaps you could clarify this for me with an example.
>
> Lets start with something simple. Say I have a routine, in a shared
> library, written in some other language, that takes a single
> parameter, say a 32-bit integer passed by reference, as opposed to by
> value.
>
> How exactly does JNI enable me to call that routine.

Re: Java momentum slowing ?

[Lance Kibblewhite]

David Botton wrote:

>You have to first wrap the call in a C (or Ada for that matter) function
>that can be called through the JNI interface. I decided just to drop the
>issue when bob@nospam came back with JNI since JNI is in native code and it
>doesn't answer my question! Meaning there is now way to directly all
>anything outside the VM with out first wrapping it in some native code JNI
>module.

Exactly correct. Thick bindings are needed, and these cannot be
written in Java.

>David Botton
>
>Lance Kibblewhite wrote:
>
>> bob@nospam wrote:
>>
>> >But to answer your question, yes of course. you use JNI (Java native
>> >interface) to call something not written in Java. It is trivial to do
>> >really. most people do that to call C or C++ code from Java. of course
>> >your application become non-portable as much as it were if all the code
>> >was written in only Java.
>>
>> Perhaps you could clarify this for me with an example.
>>
>> Lets start with something simple. Say I have a routine, in a shared
>> library, written in some other language, that takes a single
>> parameter, say a 32-bit integer passed by reference, as opposed to by
>> value.
>>
>> How exactly does JNI enable me to call that routine.

Re: Java momentum slowing ?

[Dave]

In article <373A1E6D.8CDB54C9@Botton.com>, David says...
>
 
>
>I don't understand why people have a hard time dealing with this. It
>doesn't make Java "BAD", it just means that Java is not the language /
>OS killer that Sun (and not so long ago M$) would force down your throat
>to believe.
>
 
You do not have to believe it, you can see it.

Java now is the most used language in any new applications or technology.
By the end of the year, it is estimated that there will be more than 1,000,000
java programmers, more than C++ and C combined. And Java is only less than
5 years old. Imagine what things will be like in 5 or 10 years from now.

Everywhere you look, Java is now used. People are converting exisitng
application from C++ to Java, and starting new ones in Java.

I work as a contractor, and I can't even find a new C/C++ project, (not
that I will accept one, given a choice to use Java), all the new ones 
are in Java: Java/Databases, Java/Corba, Java/XML, java/Networking, 
java everywhere.

I do not know where you live, but if you live anywhere where there are
high tech companies, I do not see how you can not see this. This is not
only in the US, but worldwide. We had a large project done for a 
company in India, and they asked for it to be in Java only.  face it,
C/C++ are dead for application development these days. Only a crazy 
person will develop something new in C++ nowadays. This is bad for M$, 
but good for the rest of the world, and good for us programmers also. 

Free us from M$ and its crapy technology it wanted to shove down our throats.
 
Notice that freedom from M$ can only be good for Ada also. If M$ remained
the only player in town, all what you will ever work with is VB and VC++
and COM.

Breaking this mold, will give a chance for other technologies and languages
to come in. Look at now what people are starting to look at along with Java:
Perl, Python, XML, CROBA,  etc.. these are all open technologies, and none
is under M$ control. Java has opened the door, now M$ is scrambling to
try to figure what to do about it.

Programmers who use Java love it, and Sun is not forcing any one to use
Java. IBM, Oracle, HP, Cisco, you name it, they are all doing major
development with Java. But with M$ case, it was the only who had control and
owned its techology secrets. Java is an open technology (but I agree it
could be made more open by Sun giving away all the source code), and Sun 
is giving away alot of it free to the world to use and enjoy. and many 
people are also building a clean room implementation of java, since 
the specs are open.

I can't see how you can compare what good Sun has done for the software
world, to M$ attempt to control it. But If I have a choice between who will
control the software world, I'd much rather it be Sun than M$. At Least
Sun products do not crash every 5 minutes.

Dave

Re: Java momentum slowing ?

[Hyman Rosen]

lrk@pobox.com (Lance Kibblewhite) writes:
> David Botton wrote:
> >You have to first wrap the call in a C (or Ada for that matter) function
> >that can be called through the JNI interface.
> 
> Exactly correct. Thick bindings are needed, and these cannot be
> written in Java.

Not quite correct. It's true that an interface routine must be written
in a native language, but there is a technique available called "shared
stubs" which allows a single interface routine to call essentially any
native function. You can get source code and description from
<http://java.sun.com/products/jdk/faq/jni-examples/jnistb10.tar.gz>.

Re: Java momentum slowing ?

[David Botton]

I can do that in Perl too (create a "function" factory) using a linked
in native code module. The point is that there is no direct method with
in the language and as you see here, it requires another language in the
middle, in this case C.

If you notice the license included has the famous Java don't do that
clause :)

This software is not designed or intended for use in on-line control of
aircraft, air traffic, aircraft navigation or aircraft communications;
or in
the design, construction, operation or maintenance of any nuclear
facility. Licensee represents and warrants that it will not use or
redistribute the Software for such purposes.


Java is what java is, a C syntaxified TCL/TK.

I don't understand why people have a hard time dealing with this. It
doesn't make Java "BAD", it just means that Java is not the language /
OS killer that Sun (and not so long ago M$) would force down your throat
to believe.

The JVM has some great uses and I have used it for many cool
applications, like providing programmability to the user who can
dynamically add to the application, etc. Do to all the marketing fluff,
the JVM has become one of the best and most portable scripting engines
around.

David Botton


Hyman Rosen wrote:
> 
> lrk@pobox.com (Lance Kibblewhite) writes:
> > David Botton wrote:
> > >You have to first wrap the call in a C (or Ada for that matter) function
> > >that can be called through the JNI interface.
> >
> > Exactly correct. Thick bindings are needed, and these cannot be
> > written in Java.
> 
> Not quite correct. It's true that an interface routine must be written
> in a native language, but there is a technique available called "shared
> stubs" which allows a single interface routine to call essentially any
> native function. You can get source code and description from
> <http://java.sun.com/products/jdk/faq/jni-examples/jnistb10.tar.gz>.

Re: Java momentum slowing ?

[Keith Thompson]

David Botton <David@Botton.com> writes:
[...]
> Java is what java is, a C syntaxified TCL/TK.
> 
> I don't understand why people have a hard time dealing with this. It
> doesn't make Java "BAD", it just means that Java is not the language /
> OS killer that Sun (and not so long ago M$) would force down your throat
> to believe.
> 
> The JVM has some great uses and I have used it for many cool
> applications, like providing programmability to the user who can
> dynamically add to the application, etc. Do to all the marketing fluff,
> the JVM has become one of the best and most portable scripting engines
> around.

As far as I know, there is no necessary connection between Java the
language and the JVM (Java Virtual Machine).  They were designed by
the same group of people as part of the same project, but Java can be
compiled to machine code like any other language, and any other
language can be compiled to Java byte codes.  (Both uses of the phrase
"any other language" are admittedly exaggerations.)

-- 
Keith Thompson (The_Other_Keith) kst@cts.com  <http://www.ghoti.net/~kst>
San Diego Supercomputer Center  <http://www.sdsc.edu>                 <*>
Techno-geek.  Mouse bigger than phone.  Bites heads off virtual chickens.

Re: Java momentum slowing ?

[Pat Rogers]

Dave <Dave@newsguy.com> wrote in message
news:7hdi8f$1plb@drn.newsguy.com...
<snip>
> Java now is the most used language in any new applications or
technology.

Creditable references would be useful here.

> By the end of the year, it is estimated that there will be more than
1,000,000
> java programmers, more than C++ and C combined.

And again.

Re: Java momentum slowing ?

[Simon Wright]

David Botton <David@Botton.com> writes:

> The fact that the world chooses to use Java in domains that Java
> would be a square peg does not change what Java is, the best
> scripting language out there.

David,

I don't know why you keep on saying this; it's a troll, surely?
haven't you caught enough fish yet?

Look, the thing about Tcl/Tk etc that makes them scripting languages,
surely, is that they make it easy to call other tools. Yet you
yourself have several times pointed out how hard it is to do that from
within Java!


-Simon

Re: Java momentum slowing ?

[David Botton]

One monopoly to another. When will people learn. If you get nice and
cozy with Sun for a little while, you will find that their practices are
not that much better the MS. Although any Unix beats out Windows
platforms, agreed.

The fact that the world chooses to use Java in domains that Java would
be a square peg does not change what Java is, the best scripting
language out there.

David Botton


> I can't see how you can compare what good Sun has done for the software
> world, to M$ attempt to control it. But If I have a choice between who will
> control the software world, I'd much rather it be Sun than M$. At Least
> Sun products do not crash every 5 minutes.
> 
> Dave

[OT] Re: Java momentum slowing ?

[Clayton Weaver]

MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

The java uses that I am seeing frequently as I browse new software
tools (not specifically related to WWW browser plugins) indicate
that java is being used as a perl/tcl/VB replacement, an interpreted
scripting language with strong type checking, object orientation,
and garbage collection. You can get OO with any of those other three with
some work, but not at the java "designed in from the beginning" level,
and of course without the type checking. The strong type-checking 
is likely important for code that is generated from formal specifications
(with case tools or manual coding) and doesn't run until the code is
verified to satisfy the spec. It's the same advantage cited by Ada
advocates when discussing development time and maintenance issues,
the ability to mechanically prove that the code meets the specification
of design requirements.

Follups to comp.lang.misc.

Regards,

Clayton Weaver
<mailto:cgweav@eskimo.com>
(Seattle)

"Everybody's ignorant, just in different subjects."  Will Rogers



-- 

Clayton Weaver
<mailto:cgweav@eskimo.com>
(Seattle)

Re: Java momentum slowing ?

[Pascal Obry]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 803 bytes --]


Simon Wright <simon@pogner.demon.co.uk> a �crit dans le message :
x7vlneyktju.fsf@pogner.moho...
> I'm just doing some Java work, _endless_ trouble with methods being
> available/deprecated/not available in various browsers -- and then the
> browser locks itself up because I suspended a thread in an applet --
> yecch. Seems it's going to take a lot of effort to make sure this
> thing works in the browser I'll be actually using. At least it doesn't
> have to work in every browser in the world (yet :-)

No, but I'am sure you'll like to have your program working in the next
browser version and you just can't assume that... be prepared for
the maintenance :-)

That's the problem, the WWW is just MMM (My computer running
My program just for Me), Java did reverse the W :-) :-) :-)

Pascal.

Re: Java momentum slowing ?

[Jean-Marten Marchi]

On 12 May 1999 22:05:40 -0700, Keith Thompson <kst@cts.com> wrote:

>As far as I know, there is no necessary connection between Java the
>language and the JVM (Java Virtual Machine).  They were designed by
>the same group of people as part of the same project, but Java can be
>compiled to machine code like any other language, and any other
>language can be compiled to Java byte codes.  (Both uses of the phrase
>"any other language" are admittedly exaggerations.)

But Java compiled to machine code is an afterthought. Other languages
can be compiled after all, that's not making them efficient. 

Re: Java momentum slowing ?

[Bob]

In article <7i54v3$qh9$1@nnrp1.deja.com>, olefevre@my-dejanews.com says...
>
>In article <7ghad5$1mma@drn.newsguy.com>, Bob@nospam wrote:
>
>> HotSPot was just released by SUN, and it is for windows only. Not
>> even for Solaris.
>

>Please don't spread FUD, particularly as an anonymous poster. SUN
>does have a Solaris HotSpot JVM: it just isn't a free download
>item. Surely you don't expect everything to be free just because
>it's in Java?
 
That was no FUD.

THis is what it says on Sun web page:

http://java.sun.com/

"Java HotSpotTM Performance Engine 
 Revisit the launch of Sun's new performance technology -- feature articles,
 documentation, industry comment -- and download the Java HotSpot
 performance engine binaries free of charge for the Windows NT/98/95
 platforms."

Now tell us where is even mentions Solaris in the above? This is the
only thing about Hotspot on their web site.   

How could any one reading the above guess that hotspot is available on
other platforms? 

may be you being a Sun insider and knows it all, can tell us the URL 
that shows where to get hotspot for Solaris?

Bob

Re: Java momentum slowing ?

[olefevre]

In article <7ghad5$1mma@drn.newsguy.com>, Bob@nospam wrote:

> HotSPot was just released by SUN, and it is for windows only. Not
> even for Solaris.

Please don't spread FUD, particularly as an anonymous poster. SUN
does have a Solaris HotSpot JVM: it just isn't a free download
item. Surely you don't expect everything to be free just because
it's in Java?

-- O.L.


--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---