Peter da Silva (peter@nmti.com) wrote:
: In article <3oqk5u$lu1@louie.udel.edu>,
: Mark C. Chu-Carroll <carroll@auriga.cis.udel.edu> wrote:
: > [delete fairly long story about people who took public comments on a
: > public project, and used it to develop their own private commercial
: > product.]
:
: [... and which sounds similar to the Mosaic/Netscape situation or Aladdin
:      Ghostscript or Eudora or BSDI or any number of other products that were
:      developed and initially distributed for free, with enhanced versions
:      coming later commercially. Which is ALL PERFECTLY FINE. The original
:      stuff is still out there for anyone who wants to reinvent the wheel
:      (GNU Ghostscript, Free/NetBSD, ...), and there's no bloody reason at
:      all someone shouldn't be able to produce a *product* out of a *project*
:      and make money at it! ...]
:
: Geez. What's next, picketing Caldera?
Or RTI, umm Ingres, umm ASK, umm CA.

OK, I could see picketing CA :-)

In article <3oqk5u$lu1@louie.udel.edu>,
Mark C. Chu-Carroll <carroll@auriga.cis.udel.edu> wrote:
> [delete fairly long story about people who took public comments on a
> public project, and used it to develop their own private commercial
> product.]

[... and which sounds similar to the Mosaic/Netscape situation or Aladdin
     Ghostscript or Eudora or BSDI or any number of other products that were
     developed and initially distributed for free, with enhanced versions
     coming later commercially. Which is ALL PERFECTLY FINE. The original
     stuff is still out there for anyone who wants to reinvent the wheel
     (GNU Ghostscript, Free/NetBSD, ...), and there's no bloody reason at
     all someone shouldn't be able to produce a *product* out of a *project*
     and make money at it! ...]

Geez. What's next, picketing Caldera?
--
Peter da Silva                                            `-_-'
Network Management Technology Incorporated                 'U`
1601 Industrial Blvd.     Sugar Land, TX  77478  USA
+1 713 274 5180                                "Har du kramat din varg idag?"

[Mark S. Hathaway]

|   When they "stole" the GNAT program they took some ideas and stuff that
|   people provided them.  Without the value embodied by those inputs they
|   wouldn't have the same product.  How will those people receive
|   reimbursement for their efforts?  Apparently the NYU people think they
|   should have to buy the latest GNAT.  Somehow it doesn't seem right.

"apparently" is the keyword here, and it is false.  go check the facts
before you face a libel suit for slandering honest, hard-working people who
have ensured that the source will remain free, and _cannot_ be hijacked by
anyone, not even themselves.

there is something else at work in the bean-counter attitude to free
software, something that should explain why people are more than willing to
post hostile guesswork, libelous lies, and to work so hard to destroy the
good name and credibility of people who give things away.  it's almost as
if... as if... _Bill Gates_ was holding their puppet strings.

I'm not associated with the GNAT project, and I don't use Ada 95 (yet).
however, I do read what some of the folks involved have written, and it is
clear that they shouldn't have to face Mark S. Hathaway and his ilk in
addition to getting their jobs done on time.

#<Erik>
--
sufficiently advanced political correctness is indistinguishable from sarcasm

In article <1995May9.000314@hobbit> hathawa2@marshall.edu (Mark S. Hathaway) writes:
>> In article <3o5mr8$aj2@nsgate.envisionet.net>,
>> cjames@cec-services.com (Colin James III (The Rt Rev'd)) writes:
>
>>> In <3o5j3d$j8h@News1.mcs.com>, jim.fleming@bytes.com wrote this with
>>> possible deletions.
>
>>> Also, since it "appears" as if you are associatted with a University,
>>> you might also want to keep in mind your social obligations to people
>>> that pay you to teach their children.
>
>> I concur.  One of the major problems with academics is that they teach
>> what interests them rather than what is useful and commercially helpful.
>>
>> The best example is the several academic prostitutes at NYU who make up
>> the GNAT team.  DoD has dumped millions into making a free Ada compiler.
>> The result is an Ada compiler written in GNU C, not Ada, which has not
>> been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
>> port unless there's a gnu c compiler around for the platform, and which
>> has absolutely no support.  Now get this:  when our tax dollars ran out,
>> the prostitutes formed a *profit* corporation to support GNAT.  When
>> asked if this was conflict of interest, that is making money off a pork
>> barrel project, they said no and that since GNAT source code was
>> available to anyone on internet then anyone could compete with them.  The
>> moral issue is are they providing _their_ source code, and the answer is
>> simply no.  Hence the taxpayer is the victim of another academic scam,
>> based on senselessly squandered dollars by government bureaucrats trying
>> to keep their jobs a mouth-breathers.
>
>
>This is an interesting situation.  And, I know of another.

[delete fairly long story about people who took public comments on a
public project, and used it to develop their own private commercial
product.]

>The question is, "Who's software is it?"  Perhaps these projects that are
>supposed to be for the public should be copyrighted GNU or Public Domain
>from the start so nobody can "hijack" it later and sell (lease) it for
>money.  People's input shouldn't be stolen so easily.

Which is *exactly* what the GNAT team did. All development of GNAT was
done under the GPL. *Anything* that the new GNAT corporation develops
will continue to fall under the GPL.

Don't believe what CJ3 claimed. He's lying.

>When they "stole" the GNAT program they took some ideas and stuff that
>people provided them.  Without the value embodied by those inputs they
>wouldn't have the same product.  How will those people receive reimbursement
>for their efforts?  Apparently the NYU people think they should have to
>buy the latest GNAT.  Somehow it doesn't seem right.

Wrong.

Want GNAT? Go get it from cs.nyu.edu. It is available there, for
free. It will *continue* to be there, for free. Any new developments
for GNAT will continue to be available, either at that site, or some
other, for free (under the GPL).

The new corporation is selling *support* for GNAT. Get GNAT from the
distribution site, and you're essentially on your own, just like with
most other GNU software. IF you're a commcercial shop that requires
reliable, prompt support if you have trouble, you have to pay for it.
It's exactly what Cygnus software does with GCC. GCC is free, source
code available to anyone who wants it. If you have a problem with GCC,
you're on your own, unless you've purchased a support contract from
Cygnus.

What's wrong with that?

 <MC>
--
|| Mark Craig Chu-Carroll: <MC> ||"If love remains
|| CIS Grad, Univ of Delaware   || Though everything is lost
|| PGP public key available     || We will pay the price
|| finger carroll@cis.udel.edu  || But we will not count the cost" -Neil Peart

>I concur.  One of the major problems with academics is that they teach
>what interests them rather than what is useful and commercially helpful.

Which is what seperates a University from a second-rate trade school.  The
pursuit of knowledge in any given field should not be dependent on what is
currently "commercially helpful".  Or are you suggesting that all CS
departments should be teaching Visual Basic under DOS/Windoze?

>The best example is the several academic prostitutes at NYU who make up
>the GNAT team.  DoD has dumped millions into making a free Ada compiler.

As well it should if it wants the language to spread...  Do you think C
would be where it is today if all early UN*X systems had not come with a
bundled C Compiler?

>The result is an Ada compiler written in GNU C, not Ada, which has not
>been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
>port unless there's a gnu c compiler around for the platform, and which
>has absolutely no support.  Now get this:  when our tax dollars ran out,

The GNU C Compiler is available for more platforms than any other C Compiler.
Furthermore, it has very good support for front-ends for other languages.
By taking advantage of this, the GNAT team ended up with a compiler which is
_much_ more portable than it would be otherwise.
As for support, the GCC development team has traditionally had a much better
record as far as turnaround time on bug-complaints than most software
providers.

>the prostitutes formed a *profit* corporation to support GNAT.  When
>asked if this was conflict of interest, that is making money off a pork
>barrel project, they said no and that since GNAT source code was
>available to anyone on internet then anyone could compete with them.  The

Since GNAT is covered by the Gnu Public License, any fixes or additions they
make will be freely available, and it will greatly help the usefulness of
the product if those who wish to can get paid full-time support for it,
much as GCC/G++, and the other GNU tools became even more popular after
Cygnus support and other companies started offering paid support contracts on
them.

>moral issue is are they providing _their_ source code, and the answer is
>simply no.  Hence the taxpayer is the victim of another academic scam,

As mentioned, the answer is actually yes.  If they make any improvements to
GNAT which are asavailable only to their customers, they are in violation
of the GPL.

>based on senselessly squandered dollars by government bureaucrats trying
>to keep their jobs a mouth-breathers.

I consider a free Ada Compiler with the efficiency of GCC and professional
support anything but a waste of my money or yours.

    Jim Wise
    wisej@acf4.nyu.edu

In <3oi69i$leg@cmcl2.NYU.EDU>, wisej@acf4.nyu.edu wrote this with
possible deletions.
>
>>I concur.  One of the major problems with academics is that they teach
>>what interests them rather than what is useful and commercially
helpful.
>
>Which is what seperates a University from a second-rate trade school.
The
>pursuit of knowledge in any given field should not be dependent on what
is
>currently "commercially helpful".  Or are you suggesting that all CS
>departments should be teaching Visual Basic under DOS/Windoze?
>
>>The best example is the several academic prostitutes at NYU who make up
>>the GNAT team.  DoD has dumped millions into making a free Ada
compiler.
>
>As well it should if it wants the language to spread...  Do you think C
>would be where it is today if all early UN*X systems had not come with a
>bundled C Compiler?
>
>>The result is an Ada compiler written in GNU C, not Ada, which has not
>>been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
>>port unless there's a gnu c compiler around for the platform, and which
>>has absolutely no support.  Now get this:  when our tax dollars ran
out,
>
>The GNU C Compiler is available for more platforms than any other C
Compiler.
>Furthermore, it has very good support for front-ends for other
languages.
>By taking advantage of this, the GNAT team ended up with a compiler
which is
>_much_ more portable than it would be otherwise.
>As for support, the GCC development team has traditionally had a much
better
>record as far as turnaround time on bug-complaints than most software
>providers.
>
>>the prostitutes formed a *profit* corporation to support GNAT.  When
>>asked if this was conflict of interest, that is making money off a pork
>>barrel project, they said no and that since GNAT source code was
>>available to anyone on internet then anyone could compete with them.
The
>
>Since GNAT is covered by the Gnu Public License, any fixes or additions
they
>make will be freely available, and it will greatly help the usefulness
of
>the product if those who wish to can get paid full-time support for it,
>much as GCC/G++, and the other GNU tools became even more popular after
>Cygnus support and other companies started offering paid support
contracts on
>them.
>
>>moral issue is are they providing _their_ source code, and the answer
is
>>simply no.  Hence the taxpayer is the victim of another academic scam,
>
>As mentioned, the answer is actually yes.  If they make any improvements
to
>GNAT which are asavailable only to their customers, they are in
violation
>of the GPL.
>
>>based on senselessly squandered dollars by government bureaucrats
trying
>>to keep their jobs a mouth-breathers.
>
>I consider a free Ada Compiler with the efficiency of GCC and
professional
>support anything but a waste of my money or yours.
>
>                                Jim Wise
>                                wisej@acf4.nyu.edu

Today I read this message by mistake, but I'll respond.

Your argument is invalid for a number of reasons with which I won't
bother you.

However, I find it consistent that you would support gnat, an NYU effort,
since your email address bears an "nyu.edu" site.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Colin James III, Principal Scientist    cjames@cec-services.com
CEC Services, 2080 Kipling St, Lakewood, CO  80215-1502   U S A
303.231.9437  Voice;  303.231.9438  Facsimile
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Excerpts from netnews.comp.std.c++: 7-May-95 Re: Academic prostitutes
at.. by Colin James III @cec-ser
> Today I read this message by mistake, but I'll respond.
>
> Your argument is invalid for a number of reasons with which I won't
> bother you.

Thank you very much for this holy contribution to the discussion.

> However, I find it consistent that you would support gnat, an NYU effort,
> since your email address bears an "nyu.edu" site.

I find it consistent that you don't know much about what you are talking about.
--
Ian S. Nelson  <bonovox@cmu.edu>                  finger for PGP key
Carnegie Mellon Computer Science/Math
Home Page:http://www.contrib.andrew.cmu.edu/usr/in22/ian.html
My opinions are not the school's, although they should be!

In article <3o79ad$q9n@agate.berkeley.edu> porco@pathos.Berkeley.EDU (Travis C. Porco) writes:

   The existence of a successful free Ada95 compiler for many
   platforms brings a full-bore modern object-oriented language with
   generics, tasking, and the works to lots of people who formerly
   were pretty much limited to one of the mutually incompatible C++'s.
   In fact GNAT Ada is a highly useful tool and its architects are to
   be commended for their work.



People have always had other choices. Take Modula-3 for example.

Back in 1991 I found a very nice and free portable compiler for
Modula-3, a "full-bore modern object-oriented language" with generics,
tasking (threads), garbage collection, exceptions, strong modules &
interfaces, provisions for systems programming, and a very short and
concise definition (~50 pages!) with a portable multi-threaded
library.

Sine 1991 the language has stayed the same, which means there is a lot
of very stable and well-documented code written in it, including
network objects with distributed garbage collection, a multi-threaded
window system, widget libraries, portable operating system interface,
animation libraries, distributed scripting langauges among other
things.

For an intro, see:

      ftp://ftp.gte.com/pub/m3/linux-journal.html.

-- Farshad
--
Farshad Nayeri
nayeri@gte.com

In article <3o8r03$eq9@gateway.skipstone.com> rajat@bugs.skipstone.com (Rajat Datta) writes:

   In time, probably a few more months, GNAT will be another fine
   addition to the collection of free software resources available
   over the net.

That's certainly true.

   Certainly GNAT can be better packaged, but that is more likely a
   problem with pthreads not being standardly available.  GNAT also
   needs more supporting cast members, like an X11 binding, but some
   of that stuff is being developed as we speak.

Or you can try SRC Modula-3 now which has had standard thread support,
garbage collection, and support X11 bindings for at least 4 years.

Some more stuff is being developed as we speak.

Modula-3 Introduction: ftp://ftp.gte.com/pub/m3/linux-journal.html.

-- Farshad
--
Farshad Nayeri
nayeri@gte.com

In article <NAYERI.95May4221058@tahoe.gte.com>,
Farshad Nayeri <nayeri@gte.com> wrote:
>
>Or you can try SRC Modula-3 now which has had standard thread support,
>garbage collection, and support X11 bindings for at least 4 years.
>
>Some more stuff is being developed as we speak.
>
>Modula-3 Introduction: ftp://ftp.gte.com/pub/m3/linux-journal.html.
>

Already have.  In other posts I have pointed out that the GNAT package
is significantly lacking in comparison with Modula-3 for Linux.

But the existence of the Modula-3 package has little to do with GNAT.
Diversity is good.  I'm glad Modula-3 is available.  I'm glad that GNAT
is developing.  Anything that adds to the free software resources is
great for small business people like myself since it significantly reduces
my costs in competing with the big guys.

rajat

>>>>> "CJ" == Colin James III (The Rt Rev'd) <cjames@cec-services.com> writes:

    CJ> Your argument is invalid for a number of reasons with which I won't
    CJ> bother you.

Proof by secret argument - one of the more pathetic rhetorical techniques.

- Chris
--
Christopher Barber        cbarber@bbn.com        http://guava.bbn.com/~cbarber

> In article <3o5mr8$aj2@nsgate.envisionet.net>,
> cjames@cec-services.com (Colin James III (The Rt Rev'd)) writes:

>> In <3o5j3d$j8h@News1.mcs.com>, jim.fleming@bytes.com wrote this with
>> possible deletions.

>> Also, since it "appears" as if you are associatted with a University,
>> you might also want to keep in mind your social obligations to people
>> that pay you to teach their children.

> I concur.  One of the major problems with academics is that they teach
> what interests them rather than what is useful and commercially helpful.
>
> The best example is the several academic prostitutes at NYU who make up
> the GNAT team.  DoD has dumped millions into making a free Ada compiler.
> The result is an Ada compiler written in GNU C, not Ada, which has not
> been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
> port unless there's a gnu c compiler around for the platform, and which
> has absolutely no support.  Now get this:  when our tax dollars ran out,
> the prostitutes formed a *profit* corporation to support GNAT.  When
> asked if this was conflict of interest, that is making money off a pork
> barrel project, they said no and that since GNAT source code was
> available to anyone on internet then anyone could compete with them.  The
> moral issue is are they providing _their_ source code, and the answer is
> simply no.  Hence the taxpayer is the victim of another academic scam,
> based on senselessly squandered dollars by government bureaucrats trying
> to keep their jobs a mouth-breathers.


This is an interesting situation.  And, I know of another.

There are several ways chess players can communicate and use the net.
One of these is a site which runs software wich allows users to telnet
in and play chess interactively with one another.  Well, the software
was developed over a period of time by one or two primary programmers
with much input (suggested improvements & features and bug reports) from
ordinary chess players.  After they had produced a nice interactive
playing & chatting environment the few "took it private".  They recently
began charging for use of it.  There's a heated debate raging on
rec.games.chess about it now.  The immediate response of another group
dedicated to having a free site took an early version of the software and
began redoing all the work from the ground up.  Most people are switching
to the new site with the old software and are rapidly improving it to get
it back to where they wanted it.

The question is, "Who's software is it?"  Perhaps these projects that are
supposed to be for the public should be copyrighted GNU or Public Domain
from the start so nobody can "hijack" it later and sell (lease) it for
money.  People's input shouldn't be stolen so easily.

When they "stole" the GNAT program they took some ideas and stuff that
people provided them.  Without the value embodied by those inputs they
wouldn't have the same product.  How will those people receive reimbursement
for their efforts?  Apparently the NYU people think they should have to
buy the latest GNAT.  Somehow it doesn't seem right.


Mark S. Hathaway       <hathawa2@marshall.edu>

In article <1995May9.000314@hobbit>,
Mark S. Hathaway <hathawa2@marshall.edu> wrote:
>When they "stole" the GNAT program they took some ideas and stuff that
>people provided them.  Without the value embodied by those inputs they
>wouldn't have the same product.  How will those people receive reimbursement
>for their efforts?  Apparently the NYU people think they should have to
>buy the latest GNAT.  Somehow it doesn't seem right.

You're responding to some misinformation.

The NYU people (i.e. the ACT company they started) have not "stolen"
anything.  Their sources are freely available.  If you get a support
contract for GNAT from ACT, you can get ACT to fix bugs.  The bug fixes
become part of the freely available sources.

ACT is analogous to Cygnus, which sells support contracts for the GNU C
compiler.  Any bugs that Cygnus fixes end up in the *public* version.
ACT does the same thing with the GNU Ada compiler.  ACT does not sell
Ada compilers -- they sell support contracts.

Anybody else who wants to start a business selling support has access to
the same GNAT sources.

The GNAT sources are, and always have been, under the GNU Copyleft, just
like all the other GNU software.

- Bob

From: ka@socrates.hr.att.com (Kenneth Almquist)
Organization: AT&T
Date: Wed, 3 May 1995 02:13:16 GMT

cjames@cec-services.com (Colin James III (The Rt Rev'd)) writes a bunch
of nonsense which wouldn't be worth replying to if he had posted it to
comp.lang.ada, but since he posted to groups containing people who may
be unfamiliar with GNAT, I will respond.

> The result is an Ada compiler written in GNU C, not Ada,

The front end is written in Ada.  GNAT uses the GCC back end, which is
written in C.

> which has not been completed,

True, but there are only a handful of unimplemented features at this point.

> is not Ada83 _or_ Ada94 compliant,

This is just a repetition of the preceding point.  (Except that it's
Ada 95, not Ada 94.  I believe the standard was completed in the closing
hours of 1994, but it was not formally issued by ISO until 1995.)

> is very difficult to port unless there's a gnu c compiler around for the
> platform,

No, the GCC code generator is quite easy to port as code generators go.
Of course it's much easier to port GNAT to a platform that has GNU C
ported to it, which is one of the big advantages of using the GNU back
end.

> and which has absolutely no support.

False.

> Now get this:  when our tax dollars ran out,

As Colin wrote above, GNAT "has not been completed."  The DoD funding
runs for another two months.

> the prostitutes formed a *profit* corporation to support GNAT.

Above Colin claimed that GNAT "has absolutely no support."

> When asked if this was conflict of interest, that is making money off a
> pork barrel project, they said no

Quite right.  I suspect that the government grant to the GNAT project was
*indended* to spur private economic activity.

> and that since GNAT source code was available to anyone on internet then
> anyone could compete with them.

True.

> The moral issue is are they providing _their_ source code, and the answer
> is simply no.

GNAT is covered by the GNU General Public License.  Enhancements made by
ACT (the corporation you refer to above) will be freely available.

> Hence the taxpayer is the victim of another academic scam,

No scam involved.  The GNAT team contracted to build an Ada 95 compiler,
and seems to be on schedule.

> based on senselessly squandered dollars by government bureaucrats
> to keep their jobs a mouth-breathers.

This is pretty much incoherent.  I should point out, however, that
"government bureaucrats" are not the only poeple who fund work which
doesn't have an immediate economic payoff.  The development of the
Standard Template Library by HP is one example.
    Kenneth Almquist
--
| Fidonet:  ka@socrates.hr.att.com 1:133/411.412
| Internet: ka@socrates.hr.att.com
| Gateway:  Galaxy Information System (GIS) Atlanta

>>>>> Mark C Chu-Carroll <carroll@auriga.cis.udel.edu> writes:

> Fifth, support *is* available, if you're willing to pay for it, from a
> variety of sources, including the new corporation that was set up by
> the NYU people, and I believe also Cygnus support.

Cygnus does not support GNAT at this time, nor am I aware of any plans to
do so.

Jason

cjames@cec-services.com (Colin James III (The Rt Rev'd)) writes a bunch
of nonsense which wouldn't be worth replying to if he had posted it to
comp.lang.ada, but since he posted to groups containing people who may
be unfamiliar with GNAT, I will respond.

> The result is an Ada compiler written in GNU C, not Ada,

The front end is written in Ada.  GNAT uses the GCC back end, which is
written in C.

> which has not been completed,

True, but there are only a handful of unimplemented features at this point.

> is not Ada83 _or_ Ada94 compliant,

This is just a repetition of the preceding point.  (Except that it's
Ada 95, not Ada 94.  I believe the standard was completed in the closing
hours of 1994, but it was not formally issued by ISO until 1995.)

> is very difficult to port unless there's a gnu c compiler around for the
> platform,

No, the GCC code generator is quite easy to port as code generators go.
Of course it's much easier to port GNAT to a platform that has GNU C
ported to it, which is one of the big advantages of using the GNU back
end.

> and which has absolutely no support.

False.

> Now get this:  when our tax dollars ran out,

As Colin wrote above, GNAT "has not been completed."  The DoD funding
runs for another two months.

> the prostitutes formed a *profit* corporation to support GNAT.

Above Colin claimed that GNAT "has absolutely no support."

> When asked if this was conflict of interest, that is making money off a
> pork barrel project, they said no

Quite right.  I suspect that the government grant to the GNAT project was
*indended* to spur private economic activity.

> and that since GNAT source code was available to anyone on internet then
> anyone could compete with them.

True.

> The moral issue is are they providing _their_ source code, and the answer
> is simply no.

GNAT is covered by the GNU General Public License.  Enhancements made by
ACT (the corporation you refer to above) will be freely available.

> Hence the taxpayer is the victim of another academic scam,

No scam involved.  The GNAT team contracted to build an Ada 95 compiler,
and seems to be on schedule.

> based on senselessly squandered dollars by government bureaucrats
> to keep their jobs a mouth-breathers.

This is pretty much incoherent.  I should point out, however, that
"government bureaucrats" are not the only poeple who fund work which
doesn't have an immediate economic payoff.  The development of the
Standard Template Library by HP is one example.
    Kenneth Almquist

>The best example is the several academic prostitutes at NYU who make up
>the GNAT team.

Why don't you keep this sort of ad-hominem attack somewhere else?  I'm glad
the GNAT team has put together a free Ada compiler; compared to some other
languages of comparable functionality, it's a pleasure to be able to use
Ada--instead of C++.

Now C++ is a fascinating language and I hope it always has some devotees
and that they write great programs using it.

But I'm glad the GNAT team has given us a choice--making it accessible to
everyone.  Ada is not the ultimate language either, but it is
darn good, and it suffered the consequences of being ahead of its time.

The existence of a successful free Ada95 compiler for many platforms
brings a full-bore modern object-oriented language with generics, tasking,
and the works to lots of people who formerly were pretty much limited to
one of the mutually incompatible C++'s.  In fact GNAT Ada is a highly
useful tool and its architects are to be commended for their work.

I don't care if their corporation is profit or nonprofit, or whether it is
funded by the military or not.  They've got a "useful and commercially
helpful" product.

--Travis Porco

I won't comment on the inimitable CJIII style, but there are as usual
several plain errors of fact that should be corrected:

>>>>"The result is an Ada compiler written in GNU C, not Ada,"

WRONG: the entire front end is written in Ada 95. The backend is the
existing GCC backend, which is written in standard C, not GNU C. The
runtime library is close to 100% Ada, with a little bit of C glue for
system functions.

>>>>"which has not been completed"

Well that's true, the contract does not end till the end of June, we expect
to complete the full implementation of Ada 95 by then.

>>>>"is not Ada83 _or_ Ada94 compliant"

First, and it is Ada95, not Ada94. Second, GNAT is pretty close to complete
now, and expected to be validated within the next few months. Certainly for
example GNAT is closer to a complete Ada 95 implementation than most C++
compilers in terms of following the ISO standards.

>>>>"is very difficult to port unless there's a gnu c compiler around
>>>> for the platform"

WRONG! It is quite easy to port GNAT to new platforms, with the exception
 of tasking (tasking will be easier when Pthreads becomes standardized and
 widely implemented, since the default tasking is built on Pthreads). Porting
 is done by standard cross-compilation techniques (the presence of a GNU C
 compiler is irrelevant). Ports exist for the following at least:

      PC (DOS/Windows)
      PC (NT)
     *PC (Linux)
     *PC (OS/2)
      PC (Nextstep)
      PC (SCO Unix)
      PC (FreeBSD)
      PC (Solaris)
     *Sun (SunOS)
     *Sun (Solaris)
      Alpha (OSF1)
     *HP (HPUX)
      1750-A
     *SGI (IRIX 5)
      SGI (IRIX 4)
      RS-6000 (AIX)
      DG Avion
      Dec station

   * = tasking available
>>>>"and which has absolutely no support"

  WRONG! NYU is providing informal free support, and has been for some time.
  We fix bugs on a best efforts basis, and new versions with new features
  and bug fixes come out about once a month. Commercial support is also
  available from:

      SGI (IRIX 5)
      Labtek (NT)
      ACT (several versions)


>>>>"Now get this:  when our tax dollars ran out,
>>>> the prostitutes formed a *profit* corporation to support GNAT.  When
>>>> asked if this was conflict of interest, that is making money off a pork
>>>> barrel project, they said no and that since GNAT source code was
>>>> available to anyone on internet then anyone could compete with them.  The
>>>> moral issue is are they providing _their_ source code, and the answer is
>>>> simply no.  Hence the taxpayer is the victim of another academic scam,
>>>> based on senselessly squandered dollars by government bureaucrats trying
>>>> to keep their jobs a mouth-breathers.

  WRONG! We are most certainly providing ALL our source code. All code is
  written under the GPL (GNU Public License), as required (suprisingly,
  this took some doing), but the original Federal contract. This guarantees
  that no one (including us) can possibly hijack this code. That actually
  is the whole point of the GPL. There have been a number of instances in
  which code originally developed at government expense has been placed
  in the public domain, and then companies, sometimes formed by the
  original developers, have taken the PD version, enhanced it and made
  it proprietary. This is certainly legal, and I certainly would not say
  there is any moral issue, but it does seem undesirable, and one of the
  reasons we pushed for the GPL approach was to make this impossible.

  Our company (ACT) is formed so that continued GNAT development can be
  funded by those actually using it, which seems at least to us to be
  more appropriate than continued government funding. We welcome other
  companies getting involved with GNAT (as you see above), Silicon
  Graphics is using GNAT for their Ada 95 technology, and will support
  it directly.

Finally, GNAT is fully available, and will continue to be available as
it is enhanced by ACT and others. Pick up a copy and have a look. Almost
all the ports listed above are available by anonymous FTP from cs.nyu.edu
in the directories pub/gnat.

Robert Dewar

"Cygnus does not support GNAT at this time, nor am I aware of any plans to
do so."

That's correct, we cooperate closely with Cygnus, who is in the business
of supporting GCC, but so far, they have not moved to get into the GNAT
maintenance business. That indeed is one of the main reasons that we
formed a separate company to concentrate on Ada. I think this makes sense
as a reasonable specialization of concerns. As you would expect, Cygnus
is much more oriented to the C and C++ world, since that is where there
customers interests lie.

In article <dewar.799505883@gnat>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>>>>>"and which has absolutely no support"
>
>  WRONG! NYU is providing informal free support, and has been for some time.
>  We fix bugs on a best efforts basis, and new versions with new features
>  and bug fixes come out about once a month. Commercial support is also
>  available from:
>
>      SGI (IRIX 5)
>      Labtek (NT)
>      ACT (several versions)
>
>

I can certainly personally attest to the fact that NYU is providing informal
free support, and that the response to my bug reports have been prompt and
informative.  The quality of the responses have been better than responses I
have received from commercial companies that I have paid for.

Certainly GNAT can be better packaged, but that is more likely a problem with
pthreads not being standardly available.  GNAT also needs more supporting cast
members, like an X11 binding, but some of that stuff is being developed as we
speak.  In time, probably a few more months, GNAT will be another fine addition
to the collection of free software resources available over the net.

rajat

In <3o5j3d$j8h@News1.mcs.com>, jim.fleming@bytes.com wrote this with
possible deletions.

> Also, since it
>"appears" as if you are associatted with a University, you might also
>want to keep in mind your social obligations to people that pay you
>to teach their children.
>

I concur.  One of the major problems with academics is that they teach
what interests them rather than what is useful and commercially helpful.

The best example is the several academic prostitutes at NYU who make up
the GNAT team.  DoD has dumped millions into making a free Ada compiler.
The result is an Ada compiler written in GNU C, not Ada, which has not
been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
port unless there's a gnu c compiler around for the platform, and which
has absolutely no support.  Now get this:  when our tax dollars ran out,
the prostitutes formed a *profit* corporation to support GNAT.  When
asked if this was conflict of interest, that is making money off a pork
barrel project, they said no and that since GNAT source code was
available to anyone on internet then anyone could compete with them.  The
moral issue is are they providing _their_ source code, and the answer is
simply no.  Hence the taxpayer is the victim of another academic scam,
based on senselessly squandered dollars by government bureaucrats trying
to keep their jobs a mouth-breathers.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Colin James III, Principal Scientist    cjames@cec-services.com
CEC Services, 2080 Kipling St, Lakewood, CO  80215-1502   U S A
303.231.9437  Voice;  303.231.9438  Facsimile
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

In article <3o5mr8$aj2@nsgate.envisionet.net>,
Colin James III (The Rt Rev'd) <cjames@cec-services.com> wrote:
>In <3o5j3d$j8h@News1.mcs.com>, jim.fleming@bytes.com wrote this with
>possible deletions.
>
>> Also, since it
>>"appears" as if you are associatted with a University, you might also
>>want to keep in mind your social obligations to people that pay you
>>to teach their children.
>>
>
>I concur.  One of the major problems with academics is that they teach
>what interests them rather than what is useful and commercially helpful.

The following paragraph has no observable connection to the preceeding,
even though it claims to be an example.

>The best example is the several academic prostitutes at NYU who make up
>the GNAT team.  DoD has dumped millions into making a free Ada compiler.
>The result is an Ada compiler written in GNU C, not Ada, which has not
>been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
>port unless there's a gnu c compiler around for the platform, and which
>has absolutely no support.  Now get this:  when our tax dollars ran out,
>the prostitutes formed a *profit* corporation to support GNAT.  When
>asked if this was conflict of interest, that is making money off a pork
>barrel project, they said no and that since GNAT source code was
>available to anyone on internet then anyone could compete with them.

Had you actually downloaded the source code you would have seen that the
vast majority of it is in fact Ada95 source.  The rest is either being
replaced by Ada95 code or is necessary to interact with the GNU system.

> The
>moral issue is are they providing _their_ source code, and the answer is
>simply no.  Hence the taxpayer is the victim of another academic scam,
>based on senselessly squandered dollars by government bureaucrats trying
>to keep their jobs a mouth-breathers.
               ^^^^^^^^^^^^^^^^^^^^^^
What does this mean?

I believe I hear a personal axe being ground...

>
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>Colin James III, Principal Scientist    cjames@cec-services.com
>CEC Services, 2080 Kipling St, Lakewood, CO  80215-1502   U S A
>303.231.9437  Voice;  303.231.9438  Facsimile
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>

--
Pat Rogers
progers@acm.org

I have no affliliation with NYU or the GNAT team, other than as a user of GNAT.

In article <3o5mr8$aj2@nsgate.envisionet.net> cjames@cec-services.com (Colin James III (The Rt Rev'd)) writes:
>In <3o5j3d$j8h@News1.mcs.com>, jim.fleming@bytes.com wrote this with
>possible deletions.
>
>> Also, since it
>>"appears" as if you are associatted with a University, you might also
>>want to keep in mind your social obligations to people that pay you
>>to teach their children.
>>
>
>I concur.  One of the major problems with academics is that they teach
>what interests them rather than what is useful and commercially helpful.
>
>The best example is the several academic prostitutes at NYU who make up
>the GNAT team.  DoD has dumped millions into making a free Ada compiler.
>The result is an Ada compiler written in GNU C, not Ada, which has not
>been completed, is not Ada83 _or_ Ada94 compliant, is very difficult to
>port unless there's a gnu c compiler around for the platform, and which
>has absolutely no support.  Now get this:  when our tax dollars ran out,
>the prostitutes formed a *profit* corporation to support GNAT.  When
>asked if this was conflict of interest, that is making money off a pork
>barrel project, they said no and that since GNAT source code was
>available to anyone on internet then anyone could compete with them.  The
>moral issue is are they providing _their_ source code, and the answer is
>simply no.  Hence the taxpayer is the victim of another academic scam,
>based on senselessly squandered dollars by government bureaucrats trying
>to keep their jobs a mouth-breathers.

Sorry. I couldn't just watch this garbage shoot past, without commenting.

First of all: GNAT is a complete compiler, with full source code
available.  Anyone who wants GNAT has access to every line of GNAT
source written by anyone, anywhere. In other words, yes, anyone who
wants it gets _their_ source code.

Second, GNAT is based on the GNU compiler technology. What that means
is that anyone, anywhere, with access to a solid C compiler, can
generate a new backend for a given piece of hardware by writing a GCC
processor description. That's not a horribly difficult piece of work,
compared to porting a more conventional compiler. Further, if anyone,
anywhere ports *ANY* of the GNU compilers to a given piece of
hardware, they get all of the others for free.

Third, GNAT is *not* written in C. It is written in a blend of C and
Ada. In order to use the GNU technologies as a springboard to make it
easier to port Ada, it is necessary to use *some* C, because all of the
GNU software is written in C. GNAT is written as a blend.

Fourth, the GNAT team took a contract from the DOD to complete an
Ada9x compiler. This they *have* done. It is not yet certified; the
DOD contract did not call for certification.

Fifth, support *is* available, if you're willing to pay for it, from a
variety of sources, including the new corporation that was set up by
the NYU people, and I believe also Cygnus support. So if you want support,
you can have it. If you're not willing to pay for it, you're on your
own. Sounds fair to me...

Sixth, I believe that the move to the corporation was not just an
intelligent move, but a *moral*, correct one. Let's face it, the GNAT
folks could have stuck with the DOD, and bled them for money for years
to come. Instead, they delivered on their contract, and then cut
themselves off from the government funding, and will support the
future development of the project off of funds earned by the
project. Sounds quite honest and aboveboard to me! Is it a conflict of
interest? Somehow, I don't see any *conflict*. GNAT is *free*, and
they give it away. Then, they make use of the free technology to establish
a business. Just like anyone else could do. You could start up a business
to compete with the GNAT people, and you'd have access to *exactly* the
same source code that they do.

Horrors. God forbid! A defense contractor did something honest. We'd
better mobilize, and make sure that the business fails! We can't have
people doing defense contracts, it'll make everyone else look too bad!

There is absolutely *nothing* wrong with the GNAT project. Get over it.

--
|| Mark Craig Chu-Carroll: <MC> ||"I live to see my fondest dreams realized,
|| CIS Grad, Univ of Delaware   || Of living under the sights and sounds of
|| PGP key available by finger  || the gifted, Who gave so much to me"
|| carroll@cis.udel.edu         ||   - _Feed_the_Fire_, Happy Rhodes

In article <3o5mr8$aj2@nsgate.envisionet.net>,
Colin James III (The Rt Rev'd) <cjames@cec-services.com> wrote:
>
>I concur.  One of the major problems with academics is that they teach
>what interests them rather than what is useful and commercially helpful.
>

Remarkable that high-tech companies tend to spring up around universities
then.

Speaking as a (very) small business entrepreneur, I have found our local
university system to be extremely helpful to our completely commercial
venture.  While certainly not all of the university students we have
interviewed for possible employment have been a great fit, neither have
most of the programmers out in the commercial world.

rajat