Ada to byte-code compiler works as advertised

Ada to byte-code compiler works as advertised

[Mike Stark]

I have been playing with the Intermetrics AppletMagic tool, which
compiles Ada source into Java byte code, and I have successfully met all
the objectives of my experimentation except for ones that I haven't 
gotten around to but that I expect no problems with.

The main purpose of this post is to let people know that this tool works
as advertised and to thank Tucker Taft of Intermetrics for his help
in figuring out the undocumented capabilities of the beta versions
I've been using.  Intermetrics has also developed browsable
documentation for the Ada interface to the Java APIs that is similar
in concept to the Java API documentation, as well as a software
development environment that is also Java byte code.  I haven't
had a chance to play with (evaluate, I mean ;) these features, but
they look nice when you scan the Intermetrics Web page 
(http://www.inmet.com:80/javadir/download/).

For those who are more interested in details, I wrote up a summary
of what I've done below the line.  For those who just want to know
if AppletMagic is a good tool -- I'd say it's a great one, and am
looking forward to seeing non-beta versions.

Mike

-----------------------------------------------------------------
My objectives were

1.  Have a Java applet call Ada code.  The application we have been
working with is to draw a satellite ground track over a world map.  The
graphics is coded in Java, and the computation of latitude and longitude
are computed in Ada.  This applet has been tested with the flight
dynamics state as package body variables.  The interface to Ada and the
.class files compiled from Ada work fine, but we are still debugging the
graphics that are written in Java.  When this is done, the applet will
be available from the Goddard Space Flight Center Flight Dynamics 
Division WWW page (http://fdd.gsfc.nasa.gov/).  I will post an 
announcement when this is done.

2.  Write a Java class that is called from Ada.  This was easy to do
based on the AppletMagic documentation. This is done using the 
"java2ada" tool on the compiled .class file.  The only undocumented 
step is that you need to run the "adareg" tool to register the output
from java2ada with the compilation library, so that the calling routines
can import the resulting package.

3.  Write a class in Ada that is callable from Java.  This was done
by replacing the package implementation discussed above which generated
 a Java class with static functions) with an implementation that
extended java.lang.Object with a tagged private record.  Unlike the
package implementation, to get a correct record type implementation
requires the use of pragma Export for all your functions.  Once that
little undocumented issue was resolved it was easy to compile a
package containing the tagged record type, and to run javap on the 
resulting class to verify that the interface was correct.  I then
wrote a Java application to create an object of this class and use
it to generate the longitudes and latitudes for the ground track.

The only minor problem here (which I attribute entirely to the fact 
that the product is a beta version) is that my private record
components in Ada become public instance variables in the corres-
ponding Java class. This does not have any impact on how the system
runs, but to have consistent semantics between the languages the
private record components should IMO become protected instance
variables.

4.  (the part I haven't done)  Write a Java class that extends the Ada
class.  I have no real doubts that this will work, but I won't get to it
until next week. Similarly, I could write an Ada extension to the Java
class being called from Ada.  Examining the generated code leaves me
no doubt that both can be done -- I just have run out of time before the
expiration of the AM1.5 beta license.

The source code for all this work will be forwarded to Intermetrics to
include in their applet writer's guide.  I will probably finish up
my work by mid-October and send it then.

Re: Ada to byte-code compiler works as advertised

[Ray Blaak]

Mike Stark <michael.e.stark@gsfc.nasa.gov> writes:

>1.  Have a Java applet call Ada code.  The application we have been
>working with is to draw a satellite ground track over a world map.  The
>graphics is coded in Java, and the computation of latitude and longitude
>are computed in Ada.

Would it have been difficult to write the graphics in Ada as well? I have no
idea what the Ada interface to the Java graphics API is like.

Cheers,
Ray Blaak
blaak@mda.ca

Re: Ada to byte-code compiler works as advertised

[Mike Stark]

Ray Blaak wrote:
> 
> Mike Stark <michael.e.stark@gsfc.nasa.gov> writes:
> 
> >1.  Have a Java applet call Ada code.  The application we have been
> >working with is to draw a satellite ground track over a world map.  The
> >graphics is coded in Java, and the computation of latitude and longitude
> >are computed in Ada.
> 
> Would it have been difficult to write the graphics in Ada as well? I have no
> idea what the Ada interface to the Java graphics API is like.
> 
> Cheers,
> Ray Blaak
> blaak@mda.ca
Ray --

I don't believe it would be difficult, but I haven't done this.  
Intermetrics has produced an Ada package interfaces to all the 
classes in the AWT libraries (as well as all the other libraries).
In fact, I believe it would be easier (if you are already Ada literate,
that is) to stick with writing Ada code and to reference the Ada 
interfaces to the Java classes.  One reason we looked at Java calling
Ada is we have a lot of C/C++ types who want to look at Java and to 
produce a sample applet as a learning experience.  Being able to link
in the Ada code to compute longitudes and latitudes was a convenience
to people who wanted to concentrate on drawing the ground track on top
of a map.

Well, that's a long-winded answer.  The short version is that it 
should be easy to write the graphics in Ada, but since I haven't done
it personally I don't want to make a lot of wild claims.

Mike

Re: Ada to byte-code compiler works as advertised

[Vince Del Vecchio]

Many thanks for the kind words.

>>>>> On Fri, 27 Sep 1996 15:43:17 +0000,
   Mike Stark <michael.e.stark@gsfc.nasa.gov> said:

> The only minor problem here (which I attribute entirely to the fact 
> that the product is a beta version) is that my private record
> components in Ada become public instance variables in the corres-
> ponding Java class. This does not have any impact on how the system
> runs, but to have consistent semantics between the languages the
> private record components should IMO become protected instance
> variables.

This is a tricky area, for a couple of reasons.  One is that if you
inline a subprogram, then your (generated) code can make references
to all sorts of components which are Ada-private.  Another is that
because we map a package spec to `<name>.class' and body to
`<name>_.class', we have to make everything in the spec Java-public
or the body won't be able to see into it.

I agree that it is frustrating that these two seemingly similar
concepts don't correspond a little more closely.  We might be able
to come up with a scheme to use private and protected components in
at least some instances, but this is not likely to change anytime
soon.  Sorry.

> 4.  (the part I haven't done)  Write a Java class that extends the Ada
> class.  I have no real doubts that this will work, but I won't get to it
> until next week. Similarly, I could write an Ada extension to the Java
> class being called from Ada.  Examining the generated code leaves me
> no doubt that both can be done -- I just have run out of time before the
> expiration of the AM1.5 beta license.

There should be a 1.6 out soon (hopefully today) to replace the
expiring 1.5.

-Vince Del Vecchio
vdelvecc@inmet.com

New version of AppletMagic (was Re: Ada to byte-code compiler ...)

[Vince Del Vecchio]

Intermetrics' AppletMagic release 1.6 is now available at
http://www.inmet.com/javadir/download/ to replace the recently
expired 1.5.  This release expires October 31.  I apologize for
the delay between the expiration of the last version and this
release.

-Vince Del Vecchio
vdelvecc@inmet.com

Re: New version of AppletMagic (was Re: Ada to byte-code compiler ...)

[Michael Feldman]

In article <epg23vaomq.fsf_-_@mesmer.camb.inmet.com>,
Vince Del Vecchio <vdelvecc@mesmer.camb.inmet.com> wrote:

>Intermetrics' AppletMagic release 1.6 is now available at
>http://www.inmet.com/javadir/download/ to replace the recently
>expired 1.5.  This release expires October 31.  I apologize for
>the delay between the expiration of the last version and this
>release.

Please let Vince know if you think Intermetrics ought to be keeping
the _Mac_ version of this compiler up to date with the rest. 

Inmet won the Academic Ada contract which required a Mac implementation
(among others) but subsequently changed the deliverables to eliminate
the Mac version ("small market", I suppose). 

"Well, we'll cover the Mac with AppletMagic", they said at the time.

Now they seem to be doing very little to the Mac version beyond the
initial release, which was a good start but is getting really behind
the other platforms.

Does anyone think there's a market for a Mac AppletMagic?

Mike Feldman

Re: New version of AppletMagic (was Re: Ada to byte-code compiler ...)

[Tucker Taft]

Michael Feldman (mfeldman@seas.gwu.edu) wrote:

: ...
: Now they seem to be doing very little to the Mac version beyond the
: initial release, which was a good start but is getting really behind
: the other platforms.
: ...

Progress is finally being made on the Mac port.  We now have
a PowerMac for building and testing AppletMagic, and we are
in the process of building a "fat" binary (68k + PowerPC).
We hope to release this sometime in the coming week.

: Mike Feldman

-Tucker Taft   stt@inmet.com   http://www.inmet.com/~stt/
Intermetrics, Inc.  Cambridge, MA  USA

Re: New version of AppletMagic (was Re: Ada to byte-code compiler ...)

[Matthew Heaney]

In article <533bki$d5n@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael
Feldman) wrote:

>>Intermetrics' AppletMagic release 1.6 is now available at
>>http://www.inmet.com/javadir/download/ to replace the recently
>>expired 1.5.  This release expires October 31.  I apologize for
>>the delay between the expiration of the last version and this
>>release.
>
>Please let Vince know if you think Intermetrics ought to be keeping
>the _Mac_ version of this compiler up to date with the rest. 

Yes, there is a market for any and all Ada-related products for the Mac!

Remember that historically Mac programming was done in Pascal; learning Ada
would be just a baby step for those programmers.  Compiler vendors: why not
tap that market?

If you're not conviced: I just took a seminar at Carnegie Mellon.  In the
classrooms where they teach CS there were row and rows of computers...all
Macs!

I am told Carnegie Mellon is the best school in the country for learning
CS, and they chose Macs.  You the Ada compiler vendors have chosen the best
language, so why the best computer, too?

Symantec, Metrowerks: how about an Ada compiler for the Mac?

Apple: how about helping port GCC/GNAT to MPW?  (If there's anyone out
there with info about an MPW port, or about how to do such a port, or is
doing it and wants some help, please contact me.)

Perhaps a better name for the Java compiler would be AppleMagic?

-matt

--------------------------------------------------------------------
Matthew Heaney
Software Development Consultant
mheaney@ni.net
(818) 985-1271

Ada on the Mac (was: AppletMagic stuff)

[Michael Feldman]

In article <mheaney-ya023180000610961022420001@news.ni.net>,
Matthew Heaney <mheaney@ni.net> wrote:

>Yes, there is a market for any and all Ada-related products for the Mac!

I'm glad you think so; so do I. A problem has been that the Ada community
has pointed at the Mac and said "small market". And the Mac software
development market has looked at Ada and said "small market". In this
"deadly embrace", nothing happened.

>Remember that historically Mac programming was done in Pascal; learning Ada
>would be just a baby step for those programmers.  Compiler vendors: why not
>tap that market?

Many Mac programmers preferred Pascal and were furious when Apple and
the compiler vendors jumped on the C++ bandwagon with everyone else.
At least Apple has produced "universal headers" to interface with the
toolbox libraries. That is what we've built on in developing the
bindings that come with GNAT-Mac.

>If you're not conviced: I just took a seminar at Carnegie Mellon.  In the
>classrooms where they teach CS there were row and rows of computers...all
>Macs!

This is true of many universities. Rumor has it that when the "academic
Ada" contract folks walked away from the Mac, the decision was based
on a market survey that asked which universities are teaching Ada on
Macs. Of course, the answer was essentially zero, because no usable
compiler existed.

>I am told Carnegie Mellon is the best school in the country for learning
>CS, and they chose Macs.  You the Ada compiler vendors have chosen the best
>language, so why the best computer, too?

I've always thought that the Ada and Mac communities should be kindred
spirits, natural allies, because both Ada and Mac have been damaged by
the "market share is the only thing that counts" business planning
that, IMHO, is ruining our industry.

>Symantec, Metrowerks: how about an Ada compiler for the Mac?

Not very likely; they would see this as just undercutting their
existing compilers, diminishing the market for each.

>Apple: how about helping port GCC/GNAT to MPW?  (If there's anyone out
>there with info about an MPW port, or about how to do such a port, or is
>doing it and wants some help, please contact me.)

Apple won't have to do much; I think it'll happen anyway.

Keep in touch with Cygnus about their port of the _GNU_ stuff to MPW.
That is coming along, according to rumors I hear. Once there is an
up to date and stable _GNU_ infrastructure for MPW, you can start
with the sources for GNAT-MachTen and go from there, presumably by
cross-compilation and tweaking the runtime, just like any other
GNAT port. The GNU back end will exist already for MPW.
   
  * * * * * *

As has been discussed many times in this group, the Mac comes with no  
command-line interface on which a traditional development environment
can be built. One is always forced to purchase an IDE. MPW is one
popular IDE (it's produced by APPLE, but Apple has always been
wishy-washy in its support thereof); Symantec and Metrowerks'
CodeWarrior are the other two biggies. The point is that realistically,
an Ada must fit into one of these, or...

...then there's MachTen, which Tenon has marketed mainly to UN*X folks
for X and Internet applications. It's a nice, stable, BSD-ish system,
especially in its recent PPC incarnation. The nicest part of it is
that - like SoftWindows - it runs alongside MacOS, in a Mac window,
so - unlike Linux - you don;t have to decide whether to boot MacOS
or MachTen. I regularly run MachTen alongside my other MacOS
apps - Word, Excel, Netscape, Claris Home Page, etc.

It turns out that MachTen is a perfectly respectable UN*X development
environment. If you prefer an editor with Mac look-and-feel, use
BBEdit. I regularly do so to develop Ada code for the Mac; MacOS
and MachTen ASCII files are interoperable, so I can develop code
using BBEdit, save the file, then use gnatmake from the MachTen
command line to build it. On MachTen, GNAT looks like any other GNAT.

   * * * * *

Recall that GNAT is intimately tied in to the GNU software, and
it would be nearly impossible to extricate it, even if you thought
it was an advantage to do so. Metrowerks actually looked at this,
and concluded that there was no legal way to connect their back end
to the GNAT front end. GPL-style free software just doesn;t mix
with proprietary stuff in the way Metrowerks desired.

We on the GNAT-Mac team investigated many different paths to a GNAT
for the Mac, and chose MachTen mainly because the GNU stuff is an
inherent, and well-supported, part of MachTen. The MPW GCC, OTOH,
was just getting started. So was MkLinux, and we further decided
that it was uninteresting because you lose your MacOS when Linux
is running. And it doesn't even support the full range of PowerMacs yet.

  * * * * *

Sure, MachTen costs a few hundred bucks, but it is a much more
useful system than Linux, at least right now, and Tenon is
committed to keeping their GNU infrastructure stable and up to date.

If you're willing to buy a copy of MachTen, you'll find the GNAT
for Mac a pretty decent system already. Bindings to a pretty large
number of toolbox libraries are provided, and genuine, double-clickable
Mac apps can be generated. The standalone apps do not themselves
require MachTen. This is similar to the MPW style in which a program
can be built either as an MPW tool or as a free-standing app.

Currently, the only big thing missing from the GNAT-Mac distribution
is the ability to link _tasking_ programs as standalone apps. This is
changing fast; we expect to have a standalone tasking runtime ready
for release in the very near future, quite possibly with 3.07.

Keep your eyes open for announcements from Tenon about some
interesting forthcoming releases of MachTen that will include
GNAT and be priced competitively. 

>Perhaps a better name for the Java compiler would be AppleMagic?

Sigh... I wish it were so.
>
>-matt

Mike Feldman

Re: Ada on the Mac (was: AppletMagic stuff)

[Larry Kilgallen]

In article <5395s7$bu8@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:
> In article <mheaney-ya023180000610961022420001@news.ni.net>,
> Matthew Heaney <mheaney@ni.net> wrote:

>>Symantec, Metrowerks: how about an Ada compiler for the Mac?
> 
> Not very likely; they would see this as just undercutting their
> existing compilers, diminishing the market for each.

> As has been discussed many times in this group, the Mac comes with no  
> command-line interface on which a traditional development environment
> can be built. One is always forced to purchase an IDE. MPW is one
> popular IDE (it's produced by APPLE, but Apple has always been
> wishy-washy in its support thereof); Symantec and Metrowerks'
> CodeWarrior are the other two biggies. The point is that realistically,
> an Ada must fit into one of these, or...

Symantec has expressed a policy of welcoming other compilers to their
IDE, freely publishing the interface specifications.  They are being
beaten up about their Pascal non-support so they seem an unlikely
candidate to do a compiler, but adapting to their IDE might be
possible for someone else.  Symantec may have a reputation
below that of Metrowerks in the Macintosh developer community,
but for many their IDE ranks far above MPW (which is alleged
to be still quite command-line oriented).

> Sure, MachTen costs a few hundred bucks, but it is a much more
> useful system than Linux, at least right now, and Tenon is
> committed to keeping their GNU infrastructure stable and up to date.
> 
> If you're willing to buy a copy of MachTen, you'll find the GNAT
> for Mac a pretty decent system already. Bindings to a pretty large
> number of toolbox libraries are provided, and genuine, double-clickable
> Mac apps can be generated. The standalone apps do not themselves
> require MachTen. This is similar to the MPW style in which a program
> can be built either as an MPW tool or as a free-standing app.

But my understanding is the current version still requires MachTen
for executing on a 68K machine.  That may work for education (the
basis on which this thread has been proceeding), but is not viable
for most commercial applications.

> Currently, the only big thing missing from the GNAT-Mac distribution
> is the ability to link _tasking_ programs as standalone apps. This is
> changing fast; we expect to have a standalone tasking runtime ready
> for release in the very near future, quite possibly with 3.07.

Keep us posted.

Larry Kilgallen

Re: Ada on the Mac (was: AppletMagic stuff)

[Michael Feldman]

In article <1996Oct7.100737.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>Symantec has expressed a policy of welcoming other compilers to their
>IDE, freely publishing the interface specifications.  They are being
>beaten up about their Pascal non-support so they seem an unlikely
>candidate to do a compiler, but adapting to their IDE might be
>possible for someone else.  

Which compiler would you adapt to their UI? 

GNAT? That would require yet another GCC port.

AdaMagic? Evidently Intermetrics, which had a contract to develop a
  Mac port, found this to be infeasible or not cost-effective, or
  whatever, so they dropped the Mac port (with AJPO's permission).

A whole 'nother Ada 95 compiler? Who's going to come up with the
  20-25 person years usually estimated for such a thing?

>Symantec may have a reputation
>below that of Metrowerks in the Macintosh developer community,
>but for many their IDE ranks far above MPW (which is alleged
>to be still quite command-line oriented).

The problem is not selecting the IDE but doing the compiler.
>
>
(referring to MachTen producing standalone Mac apps)

>But my understanding is the current version still requires MachTen
>for executing on a 68K machine.  That may work for education (the
>basis on which this thread has been proceeding), but is not viable
>for most commercial applications.

I agree with you. If Tenon comes to see this as a business advantage,
maybe they will do it. IMHO, the way to go is not to worry too much
about the 68k as a _host_, but rather to focus on a PPC-hosted
cross-compiler capable of producing fat (68k+PPC) binaries.

As my note yesterday pointed out, MachTen has been marketed historically
as a UN*X server, not so much as a development system. This is changing,
and GNAT-Mac is a big part of the change. If it "works" - if MachTen
becomes a decently popular development system - then its creators will
have the incentives to make the enhancements we'd all like to see.
>
>> Currently, the only big thing missing from the GNAT-Mac distribution
>> is the ability to link _tasking_ programs as standalone apps. This is
>> changing fast; we expect to have a standalone tasking runtime ready
>> for release in the very near future, quite possibly with 3.07.
>
>Keep us posted.

Don't worry - we'll shout it from the rooftops.:-)
>
>Larry Kilgallen

Mike Feldman

Re: Ada on the Mac (was: AppletMagic stuff)

[Larry Kilgallen]

In article <53b87u$bm9@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:
> In article <1996Oct7.100737.1@eisner>,
> Larry Kilgallen <kilgallen@eisner.decus.org> wrote:
> 
>>Symantec has expressed a policy of welcoming other compilers to their
>>IDE, freely publishing the interface specifications.  They are being
>>beaten up about their Pascal non-support so they seem an unlikely
>>candidate to do a compiler, but adapting to their IDE might be
>>possible for someone else.  
> 
> Which compiler would you adapt to their UI? 

I don't know enough about compilers to say, but I just wanted to
ensure that any readers out there who were being daunted by the
IDE aspects knew there was an alternative to writing their own.

> The problem is not selecting the IDE but doing the compiler.

Having just heard about the selection of MachTen due to it having
a command-line style of operation, it just occured to me that
some folk who might be in a position to make a difference were
not aware of that IDE option.  Perhaps Metrowerks makes the same
offer, for all I know.

> (referring to MachTen producing standalone Mac apps)
> 
>>But my understanding is the current version still requires MachTen
>>for executing on a 68K machine.  That may work for education (the
>>basis on which this thread has been proceeding), but is not viable
>>for most commercial applications.
> 
> I agree with you. If Tenon comes to see this as a business advantage,
> maybe they will do it. IMHO, the way to go is not to worry too much
> about the 68k as a _host_, but rather to focus on a PPC-hosted
> cross-compiler capable of producing fat (68k+PPC) binaries.

Absolutely !  I can affect the mix of 68K vs. PPC machines
in the development environment, but there is no way to change
that at customer sites.  To paraphrase some politician,
"Dollars for Development, not one cent for runtime royalties."

> As my note yesterday pointed out, MachTen has been marketed historically
> as a UN*X server, not so much as a development system. This is changing,
> and GNAT-Mac is a big part of the change. If it "works" - if MachTen
> becomes a decently popular development system - then its creators will
> have the incentives to make the enhancements we'd all like to see.

And they will have a much easier time getting Apple support.
Apple may not be the richest company but they have the ear of
a lot of software developers.

Larry Kilgallen

Re: Ada on the Mac (was: AppletMagic stuff)

[Michael Feldman]

In article <1996Oct7.143041.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>Having just heard about the selection of MachTen due to it having
>a command-line style of operation, it just occured to me that
>some folk who might be in a position to make a difference were
>not aware of that IDE option.  Perhaps Metrowerks makes the same
>offer, for all I know.

I didn;t say we chose MT because it had a command line. I said we chose it
because it has a stable and supported GCC infrastructure. If one had
existed for MPW, we might have gone that way. As it happens, I'm glad
we selected MachTen, because it does provide a nice way to use Unix
capabilities without throwing away your MacOS apps.

What I said about command lines is that the Mac does not provide a
standard one, so just getting started on a GNAT port requires _some_
add-on, purchased environment that supports a GCC.

This is as though Windows 95 did not provide a command line, or if 
Solaris _required_ you to use X, with no direct way to get to a shell.

>Absolutely !  I can affect the mix of 68K vs. PPC machines
>in the development environment, but there is no way to change
>that at customer sites.  To paraphrase some politician,
>"Dollars for Development, not one cent for runtime royalties."

We are in violent agreement here.

>And they will have a much easier time getting Apple support.

I don;t know if they get any tangible support from Apple, but they and
Apple are certainly "plugged in" to each other. 

>Apple may not be the richest company but they have the ear of
>a lot of software developers.

True. Apple is far from dirt-poor, in any case. Many folks have
written Apple's obituary many times (mostly folks with an axe to
grind) but Apple soldiers on and sells millions of computers a year.
That ain;t hay. As we in the Ada community know, market share is
not the only valid measure of success.:-)

>Larry Kilgallen

Mike Feldman

Re: Ada on the Mac (was: AppletMagic stuff)

[Richard A. O'Keefe]

mfeldman@seas.gwu.edu (Michael Feldman) writes:

>I agree with you. If Tenon comes to see this as a business advantage,
>maybe they will do it. IMHO, the way to go is not to worry too much
>about the 68k as a _host_, but rather to focus on a PPC-hosted
>cross-compiler capable of producing fat (68k+PPC) binaries.

Ahem.  Apple will sell you something called MAE (Macintosh Application
Environment).  What is it?  It's a complete System 7.1, with some 7.5 bits,
running on SPARC/UNIX and HPPA/HPUX systems.  (The opposite of MachTen.)
According to the documentation I have, it does this by translating
>>68k<< code to SPARC or HPPA native code on the fly.  The document
makes no mention of translartion PPC instructions to anything else.

This suggests that a rather nice development environment might be
to use a MAE-emulated 68k as the host for UNIX work...

-- 
Australian citizen since 14 August 1996.  *Now* I can vote the xxxs out!
Richard A. O'Keefe; http://www.cs.rmit.edu.au/%7Eok; RMIT Comp.Sci.

Re: Ada on the Mac (was: AppletMagic stuff)

[Michael Feldman]

In article <53ci2v$fef$1@goanna.cs.rmit.edu.au>,
Richard A. O'Keefe <ok@goanna.cs.rmit.edu.au> wrote:

>Ahem.  Apple will sell you something called MAE (Macintosh Application
>Environment).  What is it?  It's a complete System 7.1, with some 7.5 bits,
>running on SPARC/UNIX and HPPA/HPUX systems.  (The opposite of MachTen.)
>According to the documentation I have, it does this by translating
>>>68k<< code to SPARC or HPPA native code on the fly.  The document
>makes no mention of translartion PPC instructions to anything else.

Either your documentation is really out of date, or MAE is a dead
product. I'm not very familiar with it, but if it's being kept current,
it's an alternative for those who happen to have a Sun or HP box lying
around.:-)

>This suggests that a rather nice development environment might be
>to use a MAE-emulated 68k as the host for UNIX work...

Well, recent Macs make pretty decent development machines, at prices
much lower than Sun or HP boxes. I've been working on the GNAT-Mac stuff
using a 7500 (100 MHZ PPC, PCI bus) that cost me about $2k (plus about $800.
for a nice monitor). 

I started out with 16mb RAM, but - because Apple's pre-7.5.5 virtual 
memory manager was - for GNAT development - too slow, I added $500. worth of 
RAM which brought me up to 64mb. At current RAM prices, buying lots of it
is irresistible.

Apple has really improved VM in 7.5.5. Further, Tenon has been developing
a native UN*X file system that really runs like blazes, at the cost of
setting up a separate file system partition a la OS/2 or Linux.

Mike Feldman

Hosting Ada on Mac (Was: Ada on the Mac)

[Laurent Gasser]

(Larry Kilgallen) writes:
> In article <5395s7$bu8@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes: 
> Symantec has expressed a policy of welcoming other compilers to their
> IDE, freely publishing the interface specifications. 
>
> Keep us posted.
> 
> Larry Kilgallen

Both Symantech and Metrowerks allow "Plug-in" to interact with their IDE.
So both could hosts GNAT.  But Metrowerks also comes with MPW.  And now
GCC 2.7.2 is about to be available on MPW.  Sooo...

If you want to port GNAT to Metrowerks' IDE, you have to first make GCC available
there.  By buying their environment, you already own MPW which more or less
has GCC now (recent news from gnu-mac@cygnus.com).  

I don't know how it goes with the Symantech environment, but I vaguely recall
that they produce a MPW version of their compilers.  They may ship MPW with
their IDE.

I am decided to have GNAT on my Mac.  Alas, I don't have now the MachTen 
environment.  This is unfortunate because I am a Metrowerks customer as well
(and consequently own both Metrowerks'IDE and MPW...).

Could the Mac-GNAT team let us know when the stripped down MachTen will arrive?
(info@tenon.com may do the trick)

-- 
Laurent Gasser (lga@sma.ch)
Computers do not solve problems, they execute solutions.

Re: Hosting Ada on Mac (Was: Ada on the Mac)

[Michael Feldman]

In article <53dah1INNnp@maz4.sma.ch>,
Laurent Gasser <1996Oct7.100737.1@eisner> wrote:

>Could the Mac-GNAT team let us know when the stripped down MachTen will arrive?
>(info@tenon.com may do the trick)

Hey, we are GNAT fans and Mac fans. Do you think we'd keep it a secret?:-)

Seriously, we'll shout it from the rooftops. As soon as Tenon pins down
a release date and gives us a green light to publicize it, we'll be all
over the net with it. It's getting quite close; I don;t know specifics
yet. We're tidying up the GNAT part of it now.

Mike Feldman

Re: Ada on the Mac (was: AppletMagic stuff)

[Robert Dewar]

i"Symantec has expressed a policy of welcoming other compilers to their
IDE, freely publishing the interface specifications.  They are being
beaten up about their Pascal non-support so they seem an unlikely
candidate to do a compiler, but adapting to their IDE might be
possible for someone else.  Symantec may have a reputation
below that of Metrowerks in the Macintosh developer community,
but for many their IDE ranks far above MPW (which is alleged
to be still quite command-line oriented)."


MPSW is not "quite comamnd line-oriented", it is a command interpretor,
analogous to a unix shell, but with its own highly peculiar syntax.


By the way, I find it strange that Larry should be so concerned with 68K
based Macs, for educational use it is nice to be able to support such
machines at least a little bit, but I can't see any serious interest
at a commercial level. Certainly the commercial interest we have seen
in the Mac port so far is entirely PPC based.

Re: Ada on the Mac (was: AppletMagic stuff)

[Larry Kilgallen]

In article <dewar.844783703@schonberg>, dewar@schonberg.cs.nyu.edu (Robert Dewar) writes:

> By the way, I find it strange that Larry should be so concerned with 68K
> based Macs, for educational use it is nice to be able to support such
> machines at least a little bit, but I can't see any serious interest
> at a commercial level. Certainly the commercial interest we have seen
> in the Mac port so far is entirely PPC based.

Perhaps that interest is largely for in-house use.  I cannot name
a single commercial Macintosh product which is PPC-only.  I am sure
there are some, but the fact that I cannot name one indicates to me
that they are not plentiful.

Whether one can offer a PPC-only product to customers at some point
in time depends strongly on the nature of the product.  A special-purpose
item of interest only to "power users" might make it, but for any
case where one looks to bulk purchases for multiple users, one must
take care of the 68K machines which were handed down the food chain
to the least favored users.

As Jim Hopper has pointed out, this has nothing to do with the choice
of development environment.  Most 68K commercial products are probably
built on PPC.

(For those who might not be aware, the way most of the Macintosh
products work is to ship both 68K and PPC code bundled into the
same box and sometimes into the same file.)

Larry Kilgallen

Re: Ada on the Mac (was: AppletMagic stuff)

[Michael Feldman]

In article <1996Oct8.191754.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>Perhaps that interest is largely for in-house use.  I cannot name
>a single commercial Macintosh product which is PPC-only.  I am sure
>there are some, but the fact that I cannot name one indicates to me
>that they are not plentiful.

Try Netscape Navigator Gold 3.0. PPC only.

>Whether one can offer a PPC-only product to customers at some point
>in time depends strongly on the nature of the product.  A special-purpose
>item of interest only to "power users" might make it, but for any
>case where one looks to bulk purchases for multiple users, one must
>take care of the 68K machines which were handed down the food chain
>to the least favored users.

I agree.

>As Jim Hopper has pointed out, this has nothing to do with the choice
>of development environment.  Most 68K commercial products are probably
>built on PPC.

I can;t imagine serious developers not using up-to-the-minute machines
as development platforms. 

>(For those who might not be aware, the way most of the Macintosh
>products work is to ship both 68K and PPC code bundled into the
>same box and sometimes into the same file.)

Yes, this is what is called a "fat binary".

We're starting to beat a dead horse here, Larry. We're in violent
agreement that the 68k ought to be supported. My guess is that we
will eventually see a GNAT-Mac that's hosted under MachTen/PPC,
with code generators for both platforms.

What is holding it back is that the _current_ 68k support from Tenon
does not permit linking standalone apps. We've said several times here 
that Tenon is moving out in the direction of seeing MachTen as a _Mac_
development system, not just as a UN*X that happens to run on a Mac.
Indeed, the GNAT-Mac project has helped them understand the potential.

But they cannot do everything at once, so they understandably focused
on the PPC side first.

>Larry Kilgallen

Mike Feldman

Re: Ada on the Mac (was: AppletMagic stuff)

[jim hopper]

In article <dewar.844783703@schonberg>
dewar@schonberg.cs.nyu.edu (Robert Dewar) writes:

> By the way, I find it strange that Larry should be so concerned with 68K
> based Macs, for educational use it is nice to be able to support such
> machines at least a little bit, but I can't see any serious interest
> at a commercial level. Certainly the commercial interest we have seen
> in the Mac port so far is entirely PPC based.

Robert I think you misunderstand larrys point.  He is not worried about
running gnat on a 68k mac, most developers are using ppc.  BUT there
are millions of 68k macs still out there in users hands.  no mac
software developer would develop a powermac only applications unless
that app needed more power than the 68k could provide, as you are
cutting out a huge hunk of your market.  what we need to attract
commercial mac programmers is a ppc compiler which can spit out either
ppc or 68k apps.  

its really not that unrealistic at this point.  i have built the ppc to
68k cross to assembly before.  i know how to build a cross assembler (
ihave done sparc to 68k before), the bindings to mac os we have
developed are close to what we would need for 68k apps.  the only real
missing piece at this time is the linker, and from reading up on the
ppc linker its appears that the linker really should not care if its
linking a ppc fragment or a 68k fragment if it follows apples
guidelines.  [i do admit i might be real wrong here, but my reading so
far is encouraging] i am going to be talking to tenon about this when
we get native tasking complete.

jim

Re: Ada and Mac (Was: New version of AppletMagic)

[Stan Shebs]

jhopper@erinet.com (jim hopper) writes:

> 1. make files between mpw and gcc are MUCH different.  stan made a
> translator which does some of the work, but i don't yet know how much
> work its going to be to move ada make to mpw

Point me at the GNAT makefile and I'll set up a basic mpw-make.sed.
I've done enough now that it's pretty easy to write a new one.

> 3. it appears that because of the way mpw has forced stan to translate
> the make files and patches that it will be a lot more work to maintain
> the updates than it is with gnat.

Depends on how the GNAT makefile is set up.  The base GCC makefile
editor is the most complicated of all the makefile sed scripts I've had
to write, and it's been changed 10 times in the past year, of which
4-5 were to track GCC makefile changes (internal Cygnus versions of
FSF code are more turbulent than released versions).

> bottom line, i would really like to have a version that was a
> metrowerks plugin, but mpw version is just not worth the trouble.  mpw
> offers NO advantages (other than some people already have it) over
> machten once the new codebuilder version comes out from tenon.  This
> will happen in the next month or so depending on how quick we get
> native mac tasking wrapped up.

MPW *is* Apple's officially supported programming environment.  Even
with its limitations, it can do many tricks that the IDEs can only
dream of, such as building MacOS itself...

							Stan Shebs
							Cygnus Support
							shebs@cygnus.com

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <xd4tk5e4oa.fsf@andros.cygnus.com>, Stan Shebs <shebs@andros.cygnus.com> writes:

> MPW *is* Apple's officially supported programming environment.  Even
> with its limitations, it can do many tricks that the IDEs can only
> dream of, such as building MacOS itself...

But one of the tricks it cannot do is start programmers off easily.
I am not saying that MPW is never better, but that there are many
settings in which a quality IDE (free of command lines) is better.

Larry Kilgallen

Re: Ada and Mac (Was: New version of AppletMagic)

[jim hopper]

In article <xd4tk5e4oa.fsf@andros.cygnus.com>
Stan Shebs <shebs@andros.cygnus.com> writes:

> > 1. make files between mpw and gcc are MUCH different.  stan made a
> > translator which does some of the work, but i don't yet know how much
> > work its going to be to move ada make to mpw
> 
> Point me at the GNAT makefile and I'll set up a basic mpw-make.sed.
> I've done enough now that it's pretty easy to write a new one.

Stan, thanks for your generous offer.  i will get you the stuff over
this weekend. 

> 
> > 3. it appears that because of the way mpw has forced stan to translate
> > the make files and patches that it will be a lot more work to maintain
> > the updates than it is with gnat.
> 
> Depends on how the GNAT makefile is set up.  The base GCC makefile
> editor is the most complicated of all the makefile sed scripts I've had
> to write, and it's been changed 10 times in the past year, of which
> 4-5 were to track GCC makefile changes (internal Cygnus versions of
> FSF code are more turbulent than released versions).

you know much better than i do.  it would take about 10 passes a year i
ssupect of the gnat make file as well.

> 
> > bottom line, i would really like to have a version that was a
> > metrowerks plugin, but mpw version is just not worth the trouble.  mpw
> > offers NO advantages (other than some people already have it) over
> > machten once the new codebuilder version comes out from tenon.  This
> > will happen in the next month or so depending on how quick we get
> > native mac tasking wrapped up.
> 
> MPW *is* Apple's officially supported programming environment.  Even
> with its limitations, it can do many tricks that the IDEs can only
> dream of, such as building MacOS itself...

dont get my wrong, i spent a lot of years working with mpw, i love it
for a lot of things, and continued using it long after most had gone to
nice IDE's on the mac, its just that to me Machten is mpw+.

I think your gcc c compiler on mpw is a very great contribution as it
will push the envelope for the mac, but a gnat on machten and a gnat on
mpw will be pretty much the same capabilities.

thanks again for the offer of help i will be takin you up on it sooner
than i had anticipated being able to get to it thanks to this offer.

best jim

Ada and Mac (Was: New version of AppletMagic)

[Laurent Gasser]

Well, in my unsignificant opinion, the Mac is certainly a good opportunity
for Ada.  Most of the original system was written in Pascal, but new
devlopments are now in C or even C++ (QuickDraw GX,...)  

The devlopers faithful to Pascal could be an easy win to Ada 95 (Symantech has 
clearly said that Pascal was no longer one of their goal, Metrowerks diversifies
like hell seemingly without investing a lot in Pascal).

Good news: 

* GNAT has been ported to Mac (Thank you) but requires MachTen.

* GCC 2.7.2 for MPW is progressing.  I recently received a mail from 
Stan Shebs <shebs@cygnus.com> on gnu-mac@cygnus.com telling us that
a pre-release version could be found on ftp Cygnus site. 

>I haven't gotten a lot of feedback on the previous release, but have
>pressed on and put out a new one, again in private/mac on ftp.cygnus.com:
>
>-rw-rw-r--   1 shebs    cygnus   25394845 Oct  2 17:30 pmac-960930-src.tar.gz
>-rw-rw-r--   1 shebs    cygnus   4411902 Oct  2 17:28 pmac-960930.cpt.hqx

It looks like Mac is attracting some interest from third parties.

-- 
Laurent Gasser (lga@sma.ch)
Computers do not solve problems, they execute solutions.

Re: Ada and Mac (Was: New version of AppletMagic)

[jim hopper]

> Good news: 
> 
> * GNAT has been ported to Mac (Thank you) but requires MachTen.
> 
> * GCC 2.7.2 for MPW is progressing.  I recently received a mail from 
> Stan Shebs <shebs@cygnus.com> on gnu-mac@cygnus.com telling us that
> a pre-release version could be found on ftp Cygnus site. 
> 

I am working slowly on moving gnat to this, but its not as trivial as
you may think.

1. make files between mpw and gcc are MUCH different.  stan made a
translator which does some of the work, but i don't yet know how much
work its going to be to move ada make to mpw

2. tasking runtime we did depends very much on machten libraries event
the native one is using a stripped down soon to be statically linked
version of machten libs.  so tasking runtime will have to be redone
completely.  mpw environemnt does not provide made facilities required
by fsu and provenzano threads so this is a VERY non-trivial process
certainly more than i am interested in doing so while i will probably
port the compiler in the next couple months, i will not be providing
tasking with it.

3. it appears that because of the way mpw has forced stan to translate
the make files and patches that it will be a lot more work to maintain
the updates than it is with gnat.

bottom line, i would really like to have a version that was a
metrowerks plugin, but mpw version is just not worth the trouble.  mpw
offers NO advantages (other than some people already have it) over
machten once the new codebuilder version comes out from tenon.  This
will happen in the next month or so depending on how quick we get
native mac tasking wrapped up.

jim

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <53dnlp$bq7@news.syspac.com>, jhopper@erinet.com (jim hopper) writes:

> bottom line, i would really like to have a version that was a
> metrowerks plugin, but mpw version is just not worth the trouble.  mpw

There is another group who would prefer a Metrowerks plugin --

	Macintosh programmers who don't use Ada !!!

That would be such a great route to increase interest in Ada that
even Ada fans who don't care about Macintosh should _really_ want
it to happen.

It would probably also be good for the quality of Ada IDEs,
as expectations of Metrowerks users are high, and feedback
should be strong.

Larry Kilgallen

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <1996Oct8.190225.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

regarding an Ada plug-in for CodeWarrior:

>That would be such a great route to increase interest in Ada that
>even Ada fans who don't care about Macintosh should _really_ want
>it to happen.

By "plug-in" I assume you mean an independent Ada compiler that 
happens to be invoked from the CodeWarrior IDE.

The _only_ Ada 95 compiler for Mac, any Mac, is the MachTen-hosted
one right now. Given the emerging GCC port to MPW, earlier posts
have made it apparent that a GNAT hosted under MPW is feasible, if
non-trivial.

Please, before theorizing further, think about just how you
would get an Ada 95 compiler to plug in to CodeWarrior. We've
been around this loop many times.

>It would probably also be good for the quality of Ada IDEs,
>as expectations of Metrowerks users are high, and feedback
>should be strong.

The general issue has been discussed with Metrowerks principals.
They were not _in the least_ interested in developing an Ada 95 
front end. They are willing to provide their plug-in specs,
as they would to any developer, but made it quite clear that if
GNAT was involved, they would do _nothing_ to support or promote
it, or even put it on their site for contributed software.

Pure and simple, they wanted nothing whatsoever to do with anything
under the GPL.

In my opinion, AdaMagic (or indeed AppletMagic, which is what started
this thread) would be a good candidate for a Mac port (whether free-
standing or as a CW plug-in). At least the front end exists. 

This brings us full circle - Intermetrics apparently did not find 
that an interesting idea, even though it was originally part of the 
Academic Ada contract.

Mike Feldman

Re: Ada and Mac (Was: New version of AppletMagic)

[Matthew Heaney]

In article <53gikd$1nr@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael
Feldman) wrote:

>The general issue has been discussed with Metrowerks principals.
>They were not _in the least_ interested in developing an Ada 95 
>front end.  They are willing to provide their plug-in specs,
>as they would to any developer, but made it quite clear that if
>GNAT was involved, they would do _nothing_ to support or promote
>it, or even put it on their site for contributed software.

I can't blame Metrowerks for not wanting to build an Ada 95 front end from
scratch, because 

1) It would be very difficult, and 
2) Intermetrics already did it.

>Pure and simple, they wanted nothing whatsoever to do with anything
>under the GPL.

They are in the commercial software business, so again, I can't blame them
for distancing themselves from the GPL.  Commercial shops and the Free
Software Foundation can happily coexist, not unlike commercial radio and,
say, NPR.  Paying for software isn't a moral issue.

>In my opinion, AdaMagic (or indeed AppletMagic, which is what started
>this thread) would be a good candidate for a Mac port (whether free-
>standing or as a CW plug-in). At least the front end exists. 

I agree whole-heartedly.  As a matter of fact, I contacted Intermetrics
about that very idea.  The deal is that one can buy the front end, or not
buy it but give them part of your profit.  I am still seriously considering
doing this.

Now, if Metrowerks called me today and said, "Matt, we want to market an
Ada 95 compiler for the Mac and would like you to help us plug the
Intermetrics front end into our back end," then I would happily quit my
contract and move to Austin tomorrow.  Writing an Ada compiler for the Mac
would be, well, like a dream come true.

There are already today multi-processor Macs.  What better way to exploit
that technology than by using Ada?

Multi-threading will be built into System 8.  What better way to take
advantage of that feature than by writing your apps in Ada?

And why not do the port now, while System 8 is still under development? 
The details of System 8 are still being worked out (that's probably an
understatement...), so I bet Apple would listen to suggestions for
including OS features that would facilitate an Ada run-time environment. 
And with a new port, we don't have to worry about anything breaking under a
new OS.

Metrowerks already targets the BeOS, which has a multi-tasking kernal. 
What better way to write software for the BeOS than by using Ada?

>Mike Feldman

Still hoping,
Matt

--------------------------------------------------------------------
Matthew Heaney
Software Development Consultant
mheaney@ni.net
(818) 985-1271

Re: Ada and Mac (Was: New version of AppletMagic)

[Robert Dewar]

Matt said

"They are in the commercial software business, so again, I can't blame them
for distancing themselves from the GPL.  Commercial shops and the Free
Software Foundation can happily coexist, not unlike commercial radio and,
say, NPR.  Paying for software isn't a moral issue.
"


Nor is paying for software or not paying for software what the FSF and
the idea of free software is about. As Richard Stallman has gone to pains
to point out on many occasions, the free in free software is about free
availability of the sources, and freedom to modify and distribute, it is
not about paying no money. Freeware where you can get the binary for $0
but not the sources is just as objectionable as expensive proprietary
software from this point of view, price is not the issue!

Note incidentally that free software and proprietary software can coexist
in many ways. Lots of people use GCC to build proprietary software, and
also many useful proprietary tools can be used in conjunction with gcc or
GNAT. Similarly, it is quite practical to embed GNAT in proprietary
visual IDE's. In particular, it would in fact be quite practical to
put GNAT into the Metroworks environment without any problems with
respect to the GPL (my information here comes from someone who works
closely with the company), but their lawyers are nervous (probably with
no justification whatsoever) about getting anywhere near GNAT. Fine that's
their problem, but please don't think there is any fundamental problem here.

We don't have much interest in persuing such a combination, since we don't
see a significant market place, but someone else who does see a marketplace
could certainly persue this. 

I think that this kind of combination is a more practical one than trying
to build a new Mac port from scratch, even with the Intermetrics front end.
After all, if it was so easy to build a Mac compiler with this front end,
wouldn't Intermetrics have done it, especially since they signed a contract
to do so. In practice there is still a LOT of work to be done, and clearly
Intermetrics decided at the time, working with AJPO, that this effort was
not worth while. Seeing as there were relatively few complaints about this
development, one could not be blamed for assuming that this decision was
not so unreasonable.

Building a new Ada 95 compiler, even if you use an existing front end
and an existing backend is not zero work. Putting out Object Ada was
not zero work, ask the TSP folks!

If you want a Mac version of Ada 95 with a nice command line free visual
environment, then by far the most practical route is to take either the
existing GNAT port, or the Intermetrics Java version, make as few changes
as possible to the core compiler technology, and work on embedding one of
these existing compilers in a nice environment. That approach could succeed
at reasonable cost (perhaps 1-2 person years of effort).

Robert Dewar
Ada Core Technologies

Re: Ada and Mac (Was: New version of AppletMagic)

[Matthew Heaney]

In article <dewar.845036787@merv>, dewar@merv.cs.nyu.edu (Robert Dewar) wrote:

>"They are in the commercial software business, so again, I can't blame them
>for distancing themselves from the GPL.  Commercial shops and the Free
>Software Foundation can happily coexist, not unlike commercial radio and,
>say, NPR.  Paying for software isn't a moral issue.
>"
>Nor is paying for software or not paying for software what the FSF and
>the idea of free software is about. As Richard Stallman has gone to pains
>to point out on many occasions, the free in free software is about free
>availability of the sources, and freedom to modify and distribute, it is
>not about paying no money. Freeware where you can get the binary for $0
>but not the sources is just as objectionable as expensive proprietary
>software from this point of view, price is not the issue!

Yes, I agree.  Thank you for correcting me.

>If you want a Mac version of Ada 95 with a nice command line free visual
>environment, then by far the most practical route is to take either the
>existing GNAT port, or the Intermetrics Java version, make as few changes
>as possible to the core compiler technology, and work on embedding one of
>these existing compilers in a nice environment. That approach could succeed
>at reasonable cost (perhaps 1-2 person years of effort).

A ray of hope!  I guess I've got a lot of homework to do, eh?


matt

--------------------------------------------------------------------
Matthew Heaney
Software Development Consultant
mheaney@ni.net
(818) 985-1271

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <53gikd$1nr@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:

> Please, before theorizing further, think about just how you
> would get an Ada 95 compiler to plug in to CodeWarrior. We've
> been around this loop many times.

In no way did I mean to imply that such an effort would be easy, and a
recent post from someone even went into difficulties which would be
involved connecting GNAT to Symantec's IDE (Rainbow?).  I have
no basis to believe that Metrowerks does not use a similar idea
of memory-driven compilation for their IDE.

But I also did not mean to restrict consideration to GNAT.  Yes,
that may be the only possibility for Macintosh now, but things
might change at some point in the future.

> The general issue has been discussed with Metrowerks principals.
> They were not _in the least_ interested in developing an Ada 95 
> front end. They are willing to provide their plug-in specs,
> as they would to any developer, but made it quite clear that if
> GNAT was involved, they would do _nothing_ to support or promote
> it, or even put it on their site for contributed software.
> 
> Pure and simple, they wanted nothing whatsoever to do with anything
> under the GPL.

Thanks for the background on their GPL attitude.  As for front ends,
Metroworks has been known to violently change their opinion with the
passage of time, although such possibilities cannot be counted on.

> In my opinion, AdaMagic (or indeed AppletMagic, which is what started
> this thread) would be a good candidate for a Mac port (whether free-
> standing or as a CW plug-in). At least the front end exists. 
> 
> This brings us full circle - Intermetrics apparently did not find 
> that an interesting idea, even though it was originally part of the 
> Academic Ada contract.

Intermetrics web pages express a willingess to sell their front end
for use with someone else's back end.  Some investor who wanted to
fund code generators could someday build a bridge to Metrowerks or
Symantec.  That investor would expect substantial money back, nicely
differentiating the offering from GNAT.

Larry Kilgallen
interested for the long haul :-)

Re: Ada and Mac (Was: New version of AppletMagic)

[Laurent Gasser]

In article <53dnlp$bq7@news.syspac.com>, jhopper@erinet.com (jim hopper) writes:
> > Good news: 
> > 
> > * GNAT has been ported to Mac (Thank you) but requires MachTen.
> > 
> > * GCC 2.7.2 for MPW is progressing.  I recently received a mail from 
> > Stan Shebs <shebs@cygnus.com> on gnu-mac@cygnus.com telling us that
> > a pre-release version could be found on ftp Cygnus site. 
> > 
> 
> I am working slowly on moving gnat to this, but its not as trivial as
> you may think.
> 
> [sound technical reasons deleted] 
> jim

I would like to clarify my previous postings. 

Having a GNAT delivering PowerMac stand-alone applications is what counts.  To me 
and certainly to a couple of others.  

That tasking requires a lot of extension is no surprise.  This lack of MacOS has 
been often pointed to.

All in all, GNAT hosted in MPW may well never be done.

I am willing to contribute and pay for further devlopments of GNAT though.
But today, the offer looks strange:

  You can download GNAT (essentially at null cost)
  to make it run in a shell (which costs $700).
  
Compare this with commercial offers of compilers: Metrowerks offers for $400
a subscription for three releases to C/C++, Pascal, Java for Mac 68K and PPC, 
and Windows, with CodeWarrior IDE and MPW.

That's why I am waiting for a "stripped down" MachTen environment.  The product
MachTen as it is now deserve its price.  The usage I plan to make of it 
(programming shell for GNAT) does not.
-- 
Laurent Gasser (lga@sma.ch)
Computers do not solve problems, they execute solutions.

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <53g342INNks2@maz4.sma.ch>, lga@unconfigured.xvnews.domain (Laurent Gasser) writes:

> Having a GNAT delivering PowerMac stand-alone applications is what counts.  To me 
> and certainly to a couple of others.  

I understand that your needs may differ from mine...

> I am willing to contribute and pay for further devlopments of GNAT though.
> But today, the offer looks strange:
> 
>   You can download GNAT (essentially at null cost)
>   to make it run in a shell (which costs $700).
>   
> Compare this with commercial offers of compilers: Metrowerks offers for $400
> a subscription for three releases to C/C++, Pascal, Java for Mac 68K and PPC, 
> and Windows, with CodeWarrior IDE and MPW.
> 
> That's why I am waiting for a "stripped down" MachTen environment.  The product
> MachTen as it is now deserve its price.  The usage I plan to make of it 
> (programming shell for GNAT) does not.

...while I wait for something not limited to PowerMac and don't care
so much about the price. This fragmentation of the market is
what makes pervasive availability of Ada compilers an unattractive
business proposition.  Unfortunately, I fear that if there were
many new vendors entering the space they might all want to do a
Microsoft Windows version.  Thus I fear there is no hidden rule
of economics which naturally gravitates toward my wish of good
supported Ada compilers everywhere.

While Ada may alter our environment when it comes to compiling
and debugging, it has not changed the economics of acquiring
software tools.

Larry Kilgallen

Re: Ada and Mac (Was: New version of AppletMagic)

[Tucker Taft]

A new version of AppletMagic(tm) for the Mac is (finally) now available.
It includes a "Fat" binary, and has been tested on a PowerMac.
It doesn't require MPW, Machten, CodeWarrior, etc.  However, it
does require something like the JDK (Java Development Kit).

You may download it from:

    http://www.inmet.com/javadir/download/

If you have problems, you may report them to:

    appletmagic-bugs@inmet.com

Both BinHex and MacBinary formats are available for the self-extracting
archive containing the AppletMagic compiler.

Enjoy!

-Tucker Taft   stt@inmet.com   http://www.inmet.com/~stt/
Intermetrics, Inc.  Cambridge, MA  USA

Re: Ada and Mac (Was: New version of AppletMagic)

[Jon S Anthony]

In article <53gikd$1nr@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:

> Pure and simple, they wanted nothing whatsoever to do with anything
> under the GPL.
> 
> In my opinion, AdaMagic (or indeed AppletMagic, which is what started
> this thread) would be a good candidate for a Mac port (whether free-
> standing or as a CW plug-in). At least the front end exists. 

I had the same thought.  It not only exists, it appears to be a very
clever piece of work (especially vis-a-vis retargetting various
backends), and rather more impressive than the GNAT frontend (no small
feat IMO).


> This brings us full circle - Intermetrics apparently did not find 
> that an interesting idea, even though it was originally part of the 
> Academic Ada contract.

And that's just fine.  What would seem like the more sensible road
here would be to put together a Thomson style deal (where ObjectAda
uses the AdaMagic frontend, but Thomson BE, IDE, etc).  This has
produced one of the (maybe _the_) most impressive Ada offering to
date.  Question is, who is the Thomson equivalent for the Mac in such
a scenario???

/Jon
-- 
Jon Anthony
Organon Motives, Inc.
1 Williston Road, Suite 4
Belmont, MA 02178

617.484.3383
jsa@organon.com

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <JSA.96Oct10134957@alexandria>, jsa@alexandria (Jon S Anthony) writes:

> And that's just fine.  What would seem like the more sensible road
> here would be to put together a Thomson style deal (where ObjectAda
> uses the AdaMagic frontend, but Thomson BE, IDE, etc).  This has
> produced one of the (maybe _the_) most impressive Ada offering to
> date.  Question is, who is the Thomson equivalent for the Mac in such
> a scenario???

I think that is an ideal scenario, but I believe it may be too early
for anyone to fill the implementor role.  Ada vendors are too busy with
their current projects to take a chance with an unfamiliar Macintosh
market.  Macintosh tool vendors are too busy with their current
projects to take a chance with an unfamiliar language (unfamiliar
to a lot of their customers as well).

This may be a case where GNAT will lead the way and eventually
another offering devoid of command lines, make files, etc. will
come onto the Macintosh market.

Or in another sense, Thomson may lead the way, creating a success
model in the Microsoft world which many (even Thomson) would seek
to emulate on Macintosh.

With increased interest, another component source might be
Motorola.  They have their own C/C++ compiler for PPC
and offer a version (sometimes in conjunction with Apple) for
Macintosh.  A Motorola employee was just commenting in comp.arch
today about improvements they might make in their peephole
optimizer in the back end.  They are close to the chip, and
50% of the time someone buys a PowerMac Motorola makes a sale.
They are also involved with their own Macintosh clones now,
so somehow I get the feeling they don't see compiler sales
as the product line to fund another string of PCS satellites.
At some point Motorola might be quite receptive to having
their back end used with Intermetrics front end in an PowerPC
compiler.  Of course no compiler integrator would go for this
if Intermetrics and Motorola each wanted more than 50% of the
proceeds :-).

Although Motorola is the exclusive manufacturer of 68K chips, I
don't know that they have any compiler back ends kicking around.

Larry Kilgallen

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <JSA.96Oct10134957@alexandria>,
Jon S Anthony <jsa@alexandria> wrote:

>And that's just fine.  What would seem like the more sensible road
>here would be to put together a Thomson style deal (where ObjectAda
>uses the AdaMagic frontend, but Thomson BE, IDE, etc).  This has
>produced one of the (maybe _the_) most impressive Ada offering to
>date.  Question is, who is the Thomson equivalent for the Mac in such
>a scenario???

There is none; that is the problem. Thomson is an "Ada company."
The Ada companies (Thomson, Rational, Intermetrics) look at the Mac 
and say "nah, small market." 

Both Thomson (well, its predecessor Alsys) and Rational had acceptable 
(if not stellar) Ada 83 compilers for 68k Macs under MPW, and both 
companies walked away from their products. 

At least Rational (well, its predecessor Meridian) _tried_ to sell 
the Mac compiler. I think it was ahead of its time - on the Macs
of 5 or so years ago, the Ada compiler was just real clunky, though
I know several enthusiasts (including current members of the GNAT-Mac
team:-)) who got good use out of it and were annoyed at the lack
of incremental investment to keep it up to date.

Alsys, as far as I can tell, never really tried to sell its Mac compiler.
I'm reasonably plugged in to things happening in the Ada community, and
I found out about its existence only by accident. I think Alsys gave
up even before they gave it their best marketing shot. Maybe it, too,
was ahead of its time.

The Mac companies (Symantec, MetroWerks) look at Ada and say
"nah, small market." And therein lies the deadly embrace.

Part of the intent of the Academic Ada contract was to break the
deadlock by using some government dollars to seed the development of 
a proprietary, commercialized, Ada 95 compiler for the Mac (and for 
Win 95, but that's not at issue here). Intermetrics won the contract,
but then (apparently) looked at the Mac and said "nah, small market,"
got the government to agree to a contract mod, and walked away from
the Mac. 

Something will have to happen to dispel the perception on both sides
that this is not a worthwhile market to enter. I happen to think the
GNAT-MachTen development has a decent chance of doing this. Indeed,
some government dollars (an ATIP-P contract) are being used to kick-
start the process. The government money, coupled with at least an
equal amount of volunteer time (the government calls it cost-sharing)
has actually made it happen this time. 

The compiler and toolbox binding exist now (3.05 is on the usual servers
with 3.07 on the way); the MachTen folks are enthusiastic about the 
prospects. Maybe we are finally breaking the logjam.
>
>/Jon
>-- 
Mike Feldman

Re: Ada and Mac (Was: New version of AppletMagic)

[Jon S Anthony]

In article <53mdal$6t8@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:

> Jon S Anthony <jsa@alexandria> wrote:
> 
> >And that's just fine.  What would seem like the more sensible road
> >here would be to put together a Thomson style deal (where ObjectAda
> >uses the AdaMagic frontend, but Thomson BE, IDE, etc).  This has
> >produced one of the (maybe _the_) most impressive Ada offering to
> >date.  Question is, who is the Thomson equivalent for the Mac in such
> >a scenario???
> 
> There is none; that is the problem. Thomson is an "Ada company."
> The Ada companies (Thomson, Rational, Intermetrics) look at the Mac 
> and say "nah, small market." 
> [....]
> The Mac companies (Symantec, MetroWerks) look at Ada and say
> "nah, small market." And therein lies the deadly embrace.

Sounds like an opportunity for a new company which sees the
possibilities of both sides.


> Part of the intent of the Academic Ada contract was to break the
> deadlock by using some government dollars to seed the development of 
> a proprietary, commercialized, Ada 95 compiler for the Mac (and for 
> Win 95, but that's not at issue here). Intermetrics won the contract,
> but then (apparently) looked at the Mac and said "nah, small market,"
> got the government to agree to a contract mod, and walked away from
> the Mac. 

And that is just fine.  I don't see where the contract was: "Ada on
the Mac".


> Something will have to happen to dispel the perception on both sides
> that this is not a worthwhile market to enter. I happen to think the

It's either a correct perception or it is not.  If it is, then I
suppose things will never change.  If not, then it would seem that
this is indeed an opportunity for a new company which would take the
AdaMagic front end and someone elses backend and add the value added
to glue them together in a nice Mac environment.

/Jon
-- 
Jon Anthony
Organon Motives, Inc.
Belmont, MA 02178
617.484.3383
jsa@organon.com

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <JSA.96Oct14175838@alexandria>,
Jon S Anthony <jsa@alexandria> wrote:

>Sounds like an opportunity for a new company which sees the
>possibilities of both sides.

That would be nice.:-)

[referring to the Academic Ada contract]

>And that is just fine.  I don't see where the contract was: "Ada on
>the Mac".

Actually, it was quite clear in the RFP that the MacOS (on PPC) was to 
be supported by the Academic Ada compiler. As I recall, the contract
award mentioned it as well. The article in the Fall 1995 AdaIC News
mentioned Intel and PowerPC CPU's but did not list operating
systems. 

Subsequently, the contract was modified to eliminate the Mac.
So it's not just "fine" - after the fact, the contractor asked for,
and received, a contract mod from the government. This was
discussed on this group before.

>> Something will have to happen to dispel the perception on both sides
>> that this is not a worthwhile market to enter. I happen to think the

>It's either a correct perception or it is not.  If it is, then I
>suppose things will never change.  If not, then it would seem that
>this is indeed an opportunity for a new company which would take the
>AdaMagic front end and someone elses backend and add the value added
>to glue them together in a nice Mac environment.

I could not agree more. Perhaps, if the MachTen CodeBuilder release
is a big seller and it's apparent that many people are buying it
mostly for GNAT, that will be the necessary impetus. Dunno...

Mike Feldman

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <544aj2$c6p@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:
> In article <JSA.96Oct14175838@alexandria>,
> Jon S Anthony <jsa@alexandria> wrote:

>>> Something will have to happen to dispel the perception on both sides
>>> that this is not a worthwhile market to enter. I happen to think the
> 
>>It's either a correct perception or it is not.  If it is, then I
>>suppose things will never change.  If not, then it would seem that
>>this is indeed an opportunity for a new company which would take the
>>AdaMagic front end and someone elses backend and add the value added
>>to glue them together in a nice Mac environment.
> 
> I could not agree more. Perhaps, if the MachTen CodeBuilder release
> is a big seller and it's apparent that many people are buying it
> mostly for GNAT, that will be the necessary impetus. Dunno...

I had not realized there would be any other reason to buy it.
Are there capabilities other than Ada that would drive sales,
or that Tenon might be disposed to believe are driving sales
of that future product ?

Larry Kilgallen

Re: Ada and Mac (Was: New version of AppletMagic)

[jim hopper]

In article <1996Oct17.074956.1@eisner>
kilgallen@eisner.decus.org (Larry Kilgallen) writes:

> > I could not agree more. Perhaps, if the MachTen CodeBuilder release
> > is a big seller and it's apparent that many people are buying it
> > mostly for GNAT, that will be the necessary impetus. Dunno...
> 
> I had not realized there would be any other reason to buy it.
> Are there capabilities other than Ada that would drive sales,
> or that Tenon might be disposed to believe are driving sales
> of that future product ?

Larry,

there are a lot of people who want unix on the mac, who don't need unix
networking faciliites (ie they are not going to run a web server from
it, or use it to browse the internet)  the mac stuff is all i use for
example.  these people would be more interested in a unix pointed at
using and developing unix apps.  for example students are a good
example they have to work with unix and gcc env at school and want
something for their home machine.

best jim

Re: Ada and Mac (Was: New version of AppletMagic)

[Robert Dewar]

Larry says

"> I had not realized there would be any other reason to buy it.
> Are there capabilities other than Ada that would drive sales,
> or that Tenon might be disposed to believe are driving sales
> of that future product ?"


What an extremely odd and Ada-centric view of things?

Yes, it is nice that Tenon has been willing to put some resources into 
the Ada development effort using GNAT, but they see (and I would have 
thought most people would see the same way) MachTen as a generally
useful product for anyone wanting a Unix interface that is compatible
with System 7.

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <dewar.845591538@merv>, dewar@merv.cs.nyu.edu (Robert Dewar) writes:
> Larry says
> 
> "> I had not realized there would be any other reason to buy it.
>> Are there capabilities other than Ada that would drive sales,
>> or that Tenon might be disposed to believe are driving sales
>> of that future product ?"
> 
> 
> What an extremely odd and Ada-centric view of things?
> 
> Yes, it is nice that Tenon has been willing to put some resources into 
> the Ada development effort using GNAT, but they see (and I would have 
> thought most people would see the same way) MachTen as a generally
> useful product for anyone wanting a Unix interface that is compatible
> with System 7.

For MachTen, yes, but what was under discussion was a (solidly)
rumored future Tenon product which would subset those MachTen
capabilities which are particularly used for GNAT.  Those dealing
with Tenon in that regard have properly not published exact details
of the exact nature of the subsetting, since that is Tenon's right
when they announce their offering.

Those of us who are not "wanting a Unix interface" and only accept
one as a last resort certainly see things in a different light :-)

Larry Kilgallen

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <1996Oct18.075102.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>For MachTen, yes, but what was under discussion was a (solidly)
>rumored future Tenon product which would subset those MachTen
>capabilities which are particularly used for GNAT.  Those dealing
>with Tenon in that regard have properly not published exact details
>of the exact nature of the subsetting, since that is Tenon's right
>when they announce their offering.

I do _not_ speak for Tenon; I will try to clarify what I can.

I think you misinterpreted what I and others have said. The new
release (at a significantly lower price than "full" MachTen)
will - as I understand it - focus on MachTen as a _development
system_, rather than, say, a vehicle for building an Internet 
server. 

This has nothing and everything to do with GNAT - it's
not Ada-specific at all, but will surely help to break the Ada/Mac
deadlock.
>
>Those of us who are not "wanting a Unix interface" and only accept
>one as a last resort certainly see things in a different light :-)

Well, in the short run, a UN*X interface is what you'll get.:-)

REPEAT: But you can create honest-to-goodness, double-clickable,
Toolbox-using, Mac apps, just as with MPW or CodeWarrior. 

No, there won;t be a new Mac-style editor with "compile" on some menu, 
but you can certainly use your favorite Mac text editor (BBEdit, for 
example). Keeping your source(s) in one or more BBEdit windows,
and MachTen up in another window, and typing 

"gnatmake foo"

is not _so_ terrible. It's not CodeWarrior, but it's what the     
available resources could provide.
>
>Larry Kilgallen

Mike Feldman

Re: Ada and Mac (Was: New version of AppletMagic)

[Larry Kilgallen]

In article <54ghhi$fut@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:
> In article <1996Oct18.075102.1@eisner>,
> Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>>Those of us who are not "wanting a Unix interface" and only accept
>>one as a last resort certainly see things in a different light :-)
> 
> Well, in the short run, a UN*X interface is what you'll get.:-)

I certainly understand that, and _you_ understand my reaction,
but since a Unix command interface has been touted as a big
advantage by others in this group, I just took my opportunity
to reiterate the opposing view. :-)

> No, there won;t be a new Mac-style editor with "compile" on some menu, 
> but you can certainly use your favorite Mac text editor (BBEdit, for 
> example). Keeping your source(s) in one or more BBEdit windows,

Neat! Has anybody spoken to Rich Siegel about Ada syntax coloring ?
Lacking a GUI, it would seem a good opportunity for BBEdit.

Larry Kilgallen

who still has cellophane wrap around his latest BBEdit
who only bought _one_ Mike Feldman book (but did not leave it in cellophane)

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <1996Oct21.184953.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>I certainly understand that, and _you_ understand my reaction,
>but since a Unix command interface has been touted as a big
>advantage by others in this group, I just took my opportunity
>to reiterate the opposing view. :-)

Well, I don't tout any special interface as a big advantage, because
developer UI preferences are as varied and religious as editor preferences.
Certainly there are _some_ who would find a Un*x command line an
advantage, even on a Mac. 
>
>> No, there won;t be a new Mac-style editor with "compile" on some menu, 
>> but you can certainly use your favorite Mac text editor (BBEdit, for 
>> example). Keeping your source(s) in one or more BBEdit windows,
>
>Neat! Has anybody spoken to Rich Siegel about Ada syntax coloring ?

Yes.

>Lacking a GUI, it would seem a good opportunity for BBEdit.

Yep. Rich (the Bare Bones Software principal) has given us a definite maybe. 
I supplied him, as he requested, with a lot of Ada 95 code samples and
a list of reserved words. 

If (independent of MachTen, even) you think an "Ada mode" for BBEdit
would be nice, contact Bare Bones and let 'em know.

(For the Mac-uninitiated, BBEdit is the most popular ASCII editor for
the Mac. There are shareware and $$ versions.)
>
>Larry Kilgallen
>
>who still has cellophane wrap around his latest BBEdit

Maybe you should open it up. The BBEdit folks at MacWorld were wearing 
T-shirts saying "BBEdit - it doesn't suck."

I use the one that came with MachTen.:-)

>who only bought _one_ Mike Feldman book (but did not leave it in cellophane)

Hmmm....   :-)

Mike Feldman

Re: Ada and Mac (Was: New version of AppletMagic)

[Steve Wall]

In article <1996Oct21.184953.1@eisner>, kilgallen@eisner.decus.org (Larry
Kilgallen) wrote:

[about the MachTen GNAT environment]
> 
> Neat! Has anybody spoken to Rich Siegel about Ada syntax coloring ?
> Lacking a GUI, it would seem a good opportunity for BBEdit.
> 
I don't know about BBEdit but Alpha supports Ada syntax coloring right
now, as well as C, C++, Java, Scheme, FORTRAN, HTML and others.

-- 
Steve Wall

Re: Ada and Mac (Was: New version of AppletMagic)

[Tom Halliley]

Michael Feldman wrote:
>  
> I do _not_ speak for Tenon; I will try to clarify what I can.
> 
> I think you misinterpreted what I and others have said. The new
> release (at a significantly lower price than "full" MachTen)
> will - as I understand it - focus on MachTen as a _development
> system_, rather than, say, a vehicle for building an Internet
> server.
>
> Mike Feldman

Unfortunately, "significantly lower" still means $695 ...
a price that few individuals can afford.  I was really hoping for
a $195 price ... this acknowledges what Tenon is offering as
substantial without breaking my piggy bank.


TomH


-------------------------------------------------------------------
J. Thomas Halliley | Principal Software Engineer |  619.824.0348
tomh@thomsoft.com  |            Aonix            | San Diego,  CA

Re: Ada and Mac (Was: New version of AppletMagic)

[Anita Holmgren]

Tom Halliley wrote:
> 
> Michael Feldman wrote:
> > I think you misinterpreted what I and others have said. The new
> > release (at a significantly lower price than "full" MachTen)
> > will - as I understand it - focus on MachTen as a _development
> > system_, rather than, say, a vehicle for building an Internet
> > server.
> 
> Unfortunately, "significantly lower" still means $695 ...
> a price that few individuals can afford.  I was really hoping for
> a $195 price ... this acknowledges what Tenon is offering as
> substantial without breaking my piggy bank.

Tom, I don't think that price will be a barrier with the new offering, 
while Tenon isn't ready to announce this new offering yet, I think you 
and others will be quite happy with the price.

As Mike says, the product will include a full suite of software 
development tools, including of course Ada95, and will not include 
some of the other features of MachTen.  $695 is the price of the 
current products (Power MachTen and Professional MachTen).  Those 
products are available now and the gnat tools are available now as 
well (online or a CD). So, I can say that the new offering will NOT be 
$695.

-Anita

Re: Ada and Mac (Was: New version of AppletMagic)

[Tom Halliley]

Anita Holmgren wrote:

> So, I can say that the new offering will NOT be
> $695.
> 


Great!


TomH

Re: Ada and Mac (Was: New version of AppletMagic)

[Jim Easterbrook]

In article <326CF0CC.7886@thomsoft.com>, Tom Halliley (tomh@thomsoft.com) wrote:
: Michael Feldman wrote:
: >  
: > I do _not_ speak for Tenon; I will try to clarify what I can.
: > 
: > I think you misinterpreted what I and others have said. The new
: > release (at a significantly lower price than "full" MachTen)
: > will - as I understand it - focus on MachTen as a _development
: > system_, rather than, say, a vehicle for building an Internet
: > server.
: 
: Unfortunately, "significantly lower" still means $695 ...
: a price that few individuals can afford.  I was really hoping for
: a $195 price ... this acknowledges what Tenon is offering as
: substantial without breaking my piggy bank.

$695 is the price of *full* MachTen. As far as I know, prices of the new
development system have not yet been announced.
-- 
Jim Easterbrook   <jim.easterbrook@rd.bbc.co.uk> <http://www.bbc.co.uk/rd/>
   Senior R & D Engineer, Studio Section    |          PGP key available
   BBC Research & Development               |     Phone: +44 1737 836530
   Kingswood Warren, Tadworth               |     Fax:   +44 1737 836665
   Surrey  KT20 6NP, UK                     |
***       All opinions are mine and might not be shared by the BBC      ***

Re: Ada and Mac (Was: New version of AppletMagic)

[jim hopper]

In article <1996Oct17.074956.1@eisner>
kilgallen@eisner.decus.org (Larry Kilgallen) writes:

> > I could not agree more. Perhaps, if the MachTen CodeBuilder release
> > is a big seller and it's apparent that many people are buying it
> > mostly for GNAT, that will be the necessary impetus. Dunno...
> 
> I had not realized there would be any other reason to buy it.
> Are there capabilities other than Ada that would drive sales,
> or that Tenon might be disposed to believe are driving sales
> of that future product ?

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <1996Oct17.074956.1@eisner>,
Larry Kilgallen <kilgallen@eisner.decus.org> wrote:

>> I could not agree more. Perhaps, if the MachTen CodeBuilder release
>> is a big seller and it's apparent that many people are buying it
>> mostly for GNAT, that will be the necessary impetus. Dunno...

>I had not realized there would be any other reason to buy it.
>Are there capabilities other than Ada that would drive sales,
>or that Tenon might be disposed to believe are driving sales
>of that future product ?

I guess that depends on whether you think it's a good idea to have
a pretty complete BSD-ish system on your Mac that lets you run UN*X
and MacOS side by side. I can think of a number of reasons, especially
for folks with experience in both Mac and UN*X (in school, for instance):

- lots of nice tools and utilities, especially on the Ported Apps
  disk, but also within the standard MachTen; true, these are
  command line utilities, but what the heck - if they're useful,
  they're useful.

- app development for apps that will eventually run on a "big"
  UN*X box, but could be developed on a Mac first. This is a big plus
  for GNAT, of course, but it's not limited to Ada. The relative ease
  of setting up GNU cross-compilers makes the Mac a nice cheap 
  multiplatform development box.

- studying UN*X in school, would like to have it at home, already
  have a Mac at home. Yes, MkLinux is coming along, but as has been
  said here ad nauseam, it only supports a few Mac models at this
  point and does not run alongside MacOS.

- etc. etc.

If the price is right (I dunno what the final prices will be), I think
this is a big win for Ada but also for the more general developer market.
The lack of a Mac GUI on the developer tools is the only disadvantage
I can see, and not everybody thinks that's a disadvantage.

Mike Feldman

Re: Ada and Mac (Was: New version of AppletMagic)

[Jon S Anthony]

In article <544aj2$c6p@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:

> In article <JSA.96Oct14175838@alexandria>,
> Jon S Anthony <jsa@alexandria> wrote:
>[...]
> >And that is just fine.  I don't see where the contract was: "Ada on
> >the Mac".
> 
> Actually, it was quite clear in the RFP that the MacOS (on PPC) was to 
> be supported by the Academic Ada compiler. As I recall, the contract
> award mentioned it as well. The article in the Fall 1995 AdaIC News
> mentioned Intel and PowerPC CPU's but did not list operating
> systems. 
> 
> Subsequently, the contract was modified to eliminate the Mac.
> So it's not just "fine" - after the fact, the contractor asked for,
> and received, a contract mod from the government. This was
> discussed on this group before.

Well, OK, you have me.  I appears I am just plain wrong.  I really
don't know the contract specifics and thought it was simply requiring
a readily accessible and cheap "Academic Ada Compiler".

OTOH, if the mod was OK'd by the customer, how is this still not
"fine" (well, other than it goofs up the Mac situation).


> >It's either a correct perception or it is not.  If it is, then I
> >suppose things will never change.  If not, then it would seem that
> >this is indeed an opportunity for a new company which would take the
> >AdaMagic front end and someone elses backend and add the value added
> >to glue them together in a nice Mac environment.
> 
> I could not agree more. Perhaps, if the MachTen CodeBuilder release
> is a big seller and it's apparent that many people are buying it
> mostly for GNAT, that will be the necessary impetus. Dunno...

Check


/Jon
-- 
Jon Anthony
Organon Motives, Inc.
Belmont, MA 02178
617.484.3383
jsa@organon.com

Re: Ada and Mac (Was: New version of AppletMagic)

[Michael Feldman]

In article <JSA.96Oct17172950@alexandria>,
Jon S Anthony <jsa@alexandria> wrote:

>Well, OK, you have me.  I appears I am just plain wrong.  I really
>don't know the contract specifics and thought it was simply requiring
>a readily accessible and cheap "Academic Ada Compiler".

There was a good bit more than that in the RFP. Full disclosure:
we (GW, with yours truly as team leader) submitted a proposal
against this RFP, so I know its contents fairly well.:-)

>OTOH, if the mod was OK'd by the customer, how is this still not
>"fine" (well, other than it goofs up the Mac situation).

Because it goofs up the Mac situation. 

At the time of the contract award, we got very good reviews on the
technical merits, but because we proposed GNAT as the core compiler,
we were ruled out essentially because at that time (mid-1994) there
was no way to _promise_ either validation or ongoing _commercial_
backing for the products. This may seem bizarre, because one of GNAT's 
key purposes was to support education; we would have provided the
free or nearly Intel and Mac compilers-with-IDE's that were
contemplated in the RFP. But it wasn't to be.

I can justify the government's position somewhat - they wanted to see 
their (ummm, the taxpayers') $1.5 million go to kick-start a _commercial_ 
cheap-compiler industry (at the time, GNAT was not seen as commercial). 

To the extent that funding helped to complete the AdaMagic front end and 
provide the basis for what is now Thomson ObjectAda (which uses that front 
end) one of the goals was met. IMHO, the situation would be far better if
the sponsor hadn't let the contractor walk away from the Mac. 

It is, of course, not too unusual for a bidder to promise everything in 
an RFP, then - after the contract is won and it's too late to re-bid it - 
negotiate for a mod. You are right - if the customer accepts the mod,
well, that's life. But _this_ mod undermined one of the goals of the
original project, namely to fill the Mac gap.

Luckily, GNAT's commercial maturity, coupled with a change of 
administration at AJPO, made it possible for us to apply successfully
for funding (under ATIP-P) to do the current GNAT-Mac project. But 
the funding was _significantly_ less, required significantly higher cost
-sharing, and required completion on a much tighter schedule (less than a
year end-to-end). 

So we did the best our resources allowed, namely, building on
an existing Mac-based GCC and using the existing GNAT command-line
tools instead of building yet another GUI IDE. We've met the contractual
requirements (without any mods!) and are filing off the rough edges for 
the commercial product.

And, if we're all lucky, the resulting CodeBuilder CD will be
popular and break the Mac/Ada deadlock.

Mike Feldman

Re: New version of AppletMagic (was Re: Ada to byte-code compiler

[Dr Paul Gorman]

>"Well, we'll cover the Mac with AppletMagic", they said at the time.
>
>Now they seem to be doing very little to the Mac version beyond the
>initial release, which was a good start but is getting really behind
>the other platforms.
>
>Does anyone think there's a market for a Mac AppletMagic?
>
>

Well, one of the reasons we don't teach Ada as a first year language here
is that there is no Ada compiler with a book for the Mac like there is for
Pascal.

Paul.

--
-- Dr. Paul Gorman                  e-mail   : PGorman@cs.otago.ac.nz
-- Department of Computer Science   Phone    : +64 - 3 479 8366
-- University of Otago              Fax      : +64 - 3 479 8529
-- PO Box 56, Dunedin, New Zealand 
--                WWW: http://atlas.otago.ac.nz:800/staff/PaulGorman.html

Re: Ada on the Mac (was: AppletMagic stuff)

[Stan Shebs]

dewar@schonberg.cs.nyu.edu (Robert Dewar) writes:

> i"Symantec has expressed a policy of welcoming other compilers to their
> IDE, freely publishing the interface specifications.  They are being
> beaten up about their Pascal non-support so they seem an unlikely
> candidate to do a compiler, but adapting to their IDE might be
> possible for someone else.  Symantec may have a reputation
> below that of Metrowerks in the Macintosh developer community,
> but for many their IDE ranks far above MPW (which is alleged
> to be still quite command-line oriented)."

I've actually studied the prospects for doing a GCC plugin for
Symantec, and it would be fiendishly difficult, eclipsing the MPW port
in trickiness.  The Symantec IDE basically wants to do everything with
blocks of memory, while GCC consists of multiple programs
communicating via files, so you'd have to fake out GCC's I/O routines
and come up with a way to run all the programs without the IDE being
aware of what's going.  You also have to do various things that the
GCC driver normally does for you, such as setting all the correct
preprocessor symbols - and if you try to cheat on that sort of thing,
GCC has ways to punish you... :-)

Incidentally, Metrowerks now publishes their IDE interface, and we're
looking at GCC's prospects there.

							Stan