Is the Ada World Embarrassed by the Defense Industry?

Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 813 bytes --]

While looking at the SIGAda 2000 web site, I notice that the role of Ada in
defense applications is minimized (even after the explicit requirements in
this area were dropped). For example, the list of "recent" successful
Ada-based systems includes only commercial projects, some five years old,
although one of the most recent Ada success stories occurred just a few days
ago (October 24). I also notice that an interview last year with Tucker Taft
included the statement "These days we�re focused mostly on commercial
success stories..." I can understand wanting to promote commercial
applications, but isn't this going a little overboard?

http://www.acm.org/sigs/sigada/conf/sigada2000/
http://206.144.247.86/vol18_no18/interview/118-1.html
http://www.lmaeronautics.com/news/press/jsf/jsf_1stflight.html

Re: Is the Ada World Embarrassed by the Defense Industry?

[Tucker Taft]

Ken Garlington wrote:
> 
> While looking at the SIGAda 2000 web site, I notice that the role of Ada in
> defense applications is minimized (even after the explicit requirements in
> this area were dropped). For example, the list of "recent" successful
> Ada-based systems includes only commercial projects, some five years old,
> although one of the most recent Ada success stories occurred just a few days
> ago (October 24). I also notice that an interview last year with Tucker Taft
> included the statement "These days we?re focused mostly on commercial
> success stories..." I can understand wanting to promote commercial
> applications, but isn't this going a little overboard?

In my experience (which is admittedly limited), DoD folks are not
interested in military success stories; they want to see technology
that is commercially successful.  Unfortunately, it seems that commercial
folks are also not particularly interested in a language that is associated
with defense.  Hence, there seems no particular upside in spending energy
researching defense success stories, as nobody seems to be willing to
use them as a reason to consider Ada.

If the story really emphasized how Ada was *not* mandated, and was chosen
because of its inherent advantages, and it delivered successfully, then
the fact that it was a military project would hopefully not get in the
way of someone using it as a reason to consider Ada.  In general, the commercial
marketplace has a bit of a love/hate relationship with the military
marketplace.  And the military marketplace seems to have a love/hate relationship
with itself ;-).

> 
> http://www.acm.org/sigs/sigada/conf/sigada2000/
> http://206.144.247.86/vol18_no18/interview/118-1.html
> http://www.lmaeronautics.com/news/press/jsf/jsf_1stflight.html

-- 
-Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
Technical Director, Commercial Division, AverStar (formerly Intermetrics)
(http://www.averstar.com/services/IT_consulting.html)  Burlington, MA  USA

Re: Is the Ada World Embarrassed by the Defense Industry?

[Robert A Duff]

Tucker Taft <stt@averstar.com> writes:

>...  In general, the commercial
> marketplace has a bit of a love/hate relationship with the military
> marketplace.

Why does Ada get a bad name for being associated with the U.S. military,
but TCP/IP gets no such bad name?  Remember Arpanet?  ;-)

- Bob

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"Tucker Taft" <stt@averstar.com> wrote in message
news:39FDB80A.728339A9@averstar.com...

: In my experience (which is admittedly limited), DoD folks are not
: interested in military success stories; they want to see technology
: that is commercially successful.  Unfortunately, it seems that commercial
: folks are also not particularly interested in a language that is
associated
: with defense.  Hence, there seems no particular upside in spending energy
: researching defense success stories, as nobody seems to be willing to
: use them as a reason to consider Ada.

Interesting - I know more than one defense project that decided not to use
Ada, in part because "obviously" no one else in the defense industry was
sticking with it for new projects. Of course, you may be right that *DoD*
folks aren't interested, but what does that have to do (in today's
environment) with whether or not it's used on the average defense project?

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ted Dennison]

Whoa, that was scary.

I saw the subject and for a moment thought Greg Ahnorian had come back.
:-)

Whew!

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[E. Robert Tisdale]

Ken Garlington wrote:

> While looking at the SIGAda 2000 web site,
> I notice that the role of Ada in defense applications is minimized
> (even after the explicit requirements in this area were dropped).
> For example, the list of "recent" successful Ada-based systems
> includes only commercial projects, some five years old,
> although one of the most recent Ada success stories occurred
> just a few days ago (October 24).  I also notice that
> an interview last year with Tucker Taft included the statement,
> "These days we're focused mostly on commercial success stories..."
> I can understand wanting to promote commercial applications,
> but isn't this going a little overboard?

Apparently, national defense, and the U.S. Navy in particular,
has finally turned toward Commercial Off-The-Shelf (COTS)
solutions after encouraging a commercial computer industry
for the past 50 years.

It is possible to implement reliable applications
in other programming languages through diligence,
discipline and exhaustive testing.  It just costs more.
One can only assume that the commercial developer
weighed these costs against all of the other costs
relevant to application program development when
they decided which programming language(s) to use.

The problem for the military
is to test and evaluate all of these applications
and select the best value.

If application program source codes are transferred
to the military, they must find and/or train programmers
to modify and maintain that source code.
It is easier to find and train C and C++ programmers
than it is to find and train Ada programmers today
so there is a strong incentive to prefer C or C++ over Ada.

A lot can be done to incorporate safety and reliability
into C and C++ compilers and class libraries
but these languages are inherently unsafe
and there is very little that can be done about it
without changing the languages themselves.

Re: Is the Ada World Embarrassed by the Defense Industry?

[James Rogers]

"E. Robert Tisdale" wrote:
> 
> 
> It is possible to implement reliable applications
> in other programming languages through diligence,
> discipline and exhaustive testing.  It just costs more.
> One can only assume that the commercial developer
> weighed these costs against all of the other costs
> relevant to application program development when
> they decided which programming language(s) to use.

Nice sentiment, but my experience tells me this is nonesense.

I am currently involved in teaching software skills to software
professionals. Unfortunately, I have not been able to find any
customers wanting to learn Ada. I have found a number of customers
wanting to learn Java. The most curious group of these customers
are the ones wanting to convert all their legacy COBOL code to Java.

Such decisions are not being made by the people who understand the
legacy code. They are being made by management who want to feel like
they are keeping up with the times. All their friends in other companies
are using Java. That is all the reason they need. After that, they
hire some Java consultants, who naturally say that Java is the only
language worth using for ANY purpose. With such a glowing
recommendation,
the management team feels fully justified in their decision to switch
to Java.

After making the decision they sometimes ask what performance gains they
will see. The answer is always that they will see significant
performance
losses using Java. Ah well, I guess that is the price of keeping up with
your peers.

> 
> The problem for the military
> is to test and evaluate all of these applications
> and select the best value.
> 
> If application program source codes are transferred
> to the military, they must find and/or train programmers
> to modify and maintain that source code.
> It is easier to find and train C and C++ programmers
> than it is to find and train Ada programmers today
> so there is a strong incentive to prefer C or C++ over Ada.

This is also nonesense. It is easier to find C and C++ programmers
than Ada programmers. It is definitely NOT easier to train C and
C++ programmers than Ada programmers.

Among other things, you must train C and C++ programmers to avoid the
numerous dangerous features of the language, particularly when
developing
real-time system. You can spend less time teaching Ada, without multiple
class hours spent on avoiding dangerous language features.

If you want to spend the same amount of time teaching Ada, you can teach
the real-time aspects of synchronous and asynchronous tasking. These
subjects are NEVER taught in C and C++ courses. They are occasionally
taught
as SEPARATE COURSES for those C and C++ developers who want to explore
the
strange domain of concurrent systems.

> 
> A lot can be done to incorporate safety and reliability
> into C and C++ compilers and class libraries
> but these languages are inherently unsafe
> and there is very little that can be done about it
> without changing the languages themselves.

Exactly.

Jim Rogers
Colorado Springs, Colorado

Re: Is the Ada World Embarrassed by the Defense Industry?

[John Kern]

James Rogers wrote:
> 
> "E. Robert Tisdale" wrote:
> >

[[snip]]

> This is also nonesense. It is easier to find C and C++ programmers
> than Ada programmers. It is definitely NOT easier to train C and
> C++ programmers than Ada programmers.
> 
> Among other things, you must train C and C++ programmers to avoid the
> numerous dangerous features of the language, particularly when
> developing
> real-time system. You can spend less time teaching Ada, without multiple
> class hours spent on avoiding dangerous language features.

I find it somewhat amusing that in the automotive arena, there is a
whole industry of products available to keep the typical C programmer
from screwing up.  Our software process recommends, among the typical
things, 'Safer C' training from Les Hatton, MISRA C coding guidelines,
static analysis tools like QA C, OSEK compliant RTOSes for standardized
kernel access, and UML tools that help us draw pictures of what we would
have already done with Ada packages anyway.  Also, there are the
individual compiler issues and training, and the misguided attempts to
achieve reuse by adding copious precompiler directives, etc.

Have you ever noticed that whenever you service your car at a
dealership, there is almost always a 'silent' software Technical
Bulletin (recall) that has to be done?

Re: Is the Ada World Embarrassed by the Defense Industry?

[Pat Rogers]

"John Kern" <jkern3@visteon.com> wrote in message
news:3A002AFA.910DC18E@NOSPAM.visteon.com...
>
>
> James Rogers wrote:
> >
> > "E. Robert Tisdale" wrote:
> > >
>
> [[snip]]
>
> > This is also nonesense. It is easier to find C and C++ programmers
> > than Ada programmers. It is definitely NOT easier to train C and
> > C++ programmers than Ada programmers.
> >
> > Among other things, you must train C and C++ programmers to avoid
the
> > numerous dangerous features of the language, particularly when
> > developing
> > real-time system. You can spend less time teaching Ada, without
multiple
> > class hours spent on avoiding dangerous language features.
>
> I find it somewhat amusing that in the automotive arena, there is a
> whole industry of products available to keep the typical C
programmer
> from screwing up.  Our software process recommends, among the
typical
> things, 'Safer C' training from Les Hatton, MISRA C coding
guidelines,
> static analysis tools like QA C, OSEK compliant RTOSes for
standardized
> kernel access, and UML tools that help us draw pictures of what we
would
> have already done with Ada packages anyway.  Also, there are the
> individual compiler issues and training, and the misguided attempts
to
> achieve reuse by adding copious precompiler directives, etc.

Too, the MISRA guidelines for C actually recommend Ada (among others)
as an alternative to C.

---
Patrick Rogers                      Consulting and Training in:
http://www.classwide.com      Deadline Schedulability Analysis
progers@classwide.com        Software Fault Tolerance
(281)648-3165                       Real-Time/OO Languages

Adam ... does not deserve all the credit; much is due to Eve, the
first woman, and Satan, the first consultant.
Mark Twain

Re: Is the Ada World Embarrassed by the Defense Industry?

[Pat Rogers]

"E. Robert Tisdale" <edwin@netwood.net> wrote in message
news:39FDE9E4.35F615A6@netwood.net...
> Ken Garlington wrote:
>
> > While looking at the SIGAda 2000 web site,
> > I notice that the role of Ada in defense applications is minimized
> > (even after the explicit requirements in this area were dropped).
> > For example, the list of "recent" successful Ada-based systems
> > includes only commercial projects, some five years old,
> > although one of the most recent Ada success stories occurred
> > just a few days ago (October 24).  I also notice that
> > an interview last year with Tucker Taft included the statement,
> > "These days we're focused mostly on commercial success stories..."
> > I can understand wanting to promote commercial applications,
> > but isn't this going a little overboard?
>
> Apparently, national defense, and the U.S. Navy in particular,
> has finally turned toward Commercial Off-The-Shelf (COTS)
> solutions after encouraging a commercial computer industry
> for the past 50 years.

The DoD is institutionally clueless about software development.  (Oh
sure there are plenty of bright, capable people involved; but as an
organization it is lost.)  They went from a simple language policy
("use Ada for the things that make sense") to a laughable one that
requires good will on the part of their contractors ("do a reasonable
language trade-off study including Ada").  What a sad joke!

> It is possible to implement reliable applications
> in other programming languages through diligence,
> discipline and exhaustive testing.  It just costs more.
> One can only assume that the commercial developer
> weighed these costs against all of the other costs
> relevant to application program development when
> they decided which programming language(s) to use.

ROTFL

Thanks -- that (and the lack of a smiley) made my day!

> The problem for the military
> is to test and evaluate all of these applications
> and select the best value.
>
> If application program source codes are transferred
> to the military, they must find and/or train programmers
> to modify and maintain that source code.
> It is easier to find and train C and C++ programmers
> than it is to find and train Ada programmers today
> so there is a strong incentive to prefer C or C++ over Ada.

References please, otherwise this propagates myth.  IMHO training Ada
people is easy.  Finding good Ada people is easy if you pay well.

> A lot can be done to incorporate safety and reliability
> into C and C++ compilers and class libraries
> but these languages are inherently unsafe
> and there is very little that can be done about it
> without changing the languages themselves.

I don't see how the first part of that sentence agrees with the last.
(And I agree with the last part.)

---
Patrick Rogers                      Consulting and Training in:
http://www.classwide.com      Deadline Schedulability Analysis
progers@classwide.com        Software Fault Tolerance
(281)648-3165                       Real-Time/OO Languages

Adam ... does not deserve all the credit; much is due to Eve, the
first woman, and Satan, the first consultant.
Mark Twain

Re: Is the Ada World Embarrassed by the Defense Industry?

[Lao Xiao Hai]

"E. Robert Tisdale" wrote:

> Apparently, national defense, and the U.S. Navy in particular,
> has finally turned toward Commercial Off-The-Shelf (COTS)
> solutions after encouraging a commercial computer industry
> for the past 50 years.

An Ada RTE is COTS.

> It is possible to implement reliable applications
> in other programming languages through diligence,
> discipline and exhaustive testing.  It just costs more.
> One can only assume that the commercial developer
> weighed these costs against all of the other costs
> relevant to application program development when
> they decided which programming language(s) to use.

Indeed!!!!!!??????   Most organizations that I see choosing
C++ over Ada have done very little in the way of careful
study.   Certainly no U.S. military organization has thought
this through very carefully.   It is, in fact, quite scary.  An
organization that could not manage a  single-language
policy is under the illusion that it can manage a multiple-
language policy.


> .The problem for the military

> is to test and evaluate all of these applications

> and select the best value.

And which ISO standard validated C++ compiler
will they select.   OOOOOPS!  There aren't any.

> If application program source codes are transferred
> to the military, they must find and/or train programmers
> to modify and maintain that source code.
> It is easier to find and train C and C++ programmers
> than it is to find and train Ada programmers today
> so there is a strong incentive to prefer C or C++ over Ada.

It is easier to find programmers who claim to know C++ than
programmers who claim to know Ada.  Their skill level varies
all over the place and most of them should not be allowed
anywhere within ten miles of the source code.    As for training.
I teach C++ and I teach Ada.  Ada is easier to teach than C++.
Properly taught, Ada is easier for the students to learn.

> A lot can be done to incorporate safety and reliability
> into C and C++ compilers and class libraries
> but these languages are inherently unsafe
> and there is very little that can be done about it
> without changing the languages themselves.

C++ is the peanut brittle of programming languages.  It gets
stuck in your teeth, breaks easily, and cause serious rotting.
Someone once said, "C is its own virus."   This same comment
extends to C++.

Richard Riehle

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ted Dennison]

In article <39FE461D.275F1363@ix.netcom.com>,
  Lao Xiao Hai <laoxhai@ix.netcom.com> wrote:
>
> Someone once said, "C is its own virus."   This same comment
> extends to C++.

That probably came from the old "Worse is Better" paper, by
rpg@lucid.com. (If I was a good CLOS person, I'd probably know who "rpg"
is.) see
http://www.ai.mit.edu/docs/articles//good-news/subsection3.2.1.html . It
contains the comment:

"Unix and C are the ultimate computer viruses."


--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[mjsilva]

In article <39FE461D.275F1363@ix.netcom.com>,
  Lao Xiao Hai <laoxhai@ix.netcom.com> wrote:
>
> Indeed!!!!!!??????   Most organizations that I see choosing
> C++ over Ada have done very little in the way of careful
> study.   Certainly no U.S. military organization has thought
> this through very carefully.   It is, in fact, quite scary.  An
> organization that could not manage a  single-language
> policy is under the illusion that it can manage a multiple-
> language policy.

I'm still interested in hearing if any of those who have switched from
Ada to The Radiant Future of language X are finding that there's
trouble in paradise X.  It seems that if there is trouble in paradise X
then we Ada advocates should be sure and document it for the benefit of
others who are considering such a switch, or who are simply evaluating
languages.  OTOH, if everybody is happy as a clam using X then I guess
we need to rethink some of our assumptions.

Mike




Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[E. Robert Tisdale]

mjsilva@my-deja.com wrote:

> I'm still interested in hearing
> if any of those who have switched from Ada
> to The Radiant Future of language X
> are finding that there's trouble in paradise X.
> It seems that if there is trouble in paradise X
> then we Ada advocates should be sure and document it
> for the benefit of others who are considering such a switch,
> or who are simply evaluating languages.
> OTOH, if everybody is happy as a clam using X
> then I guess we need to rethink some of our assumptions.

Mike,

We are talking about programming languages -- NOT religions.
Aren't we?

Re: Is the Ada World Embarrassed by the Defense Industry?

[mjsilva]

In article <39FEFBF8.D62508B8@netwood.net>,
  "E. Robert Tisdale" <edwin@netwood.net> wrote:
> mjsilva@my-deja.com wrote:
>
> > I'm still interested in hearing
> > if any of those who have switched from Ada
> > to The Radiant Future of language X
> > are finding that there's trouble in paradise X.
> > It seems that if there is trouble in paradise X
> > then we Ada advocates should be sure and document it
> > for the benefit of others who are considering such a switch,
> > or who are simply evaluating languages.
> > OTOH, if everybody is happy as a clam using X
> > then I guess we need to rethink some of our assumptions.
>
> Mike,
>
> We are talking about programming languages -- NOT religions.
> Aren't we?

Yep.


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Jeff Carter]

"E. Robert Tisdale" wrote:
> We are talking about programming languages -- NOT religions.
> Aren't we?

Aren't they the same thing? :)

-- 
Jeff Carter
"Illegitimate-faced bugger-folk!"
Monty Python & the Holy Grail

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

Actually, as the person who started this thread, I was talking about
marketing. It just seems odd to make choices that damage your historical
customer base (military) before you've adequately captured your new base
(non-military).

Re: Is the Ada World Embarrassed by the Defense Industry?

[Tucker Taft]

Ken Garlington wrote:
> 
> Actually, as the person who started this thread, I was talking about
> marketing. It just seems odd to make choices that damage your historical
> customer base (military) before you've adequately captured your new base
> (non-military).

I think we are interested in appealing to military customers.  If you have
recommendations of the best way to do so, please fire away.
I honestly had the sense that broadcasting military success stories
was actually counterproductive in some circles, including some military
circles.

-- 
-Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
Technical Director, Commercial Division, AverStar (formerly Intermetrics)
(http://www.averstar.com/services/IT_consulting.html)  Burlington, MA  USA

Re: Is the Ada World Embarrassed by the Defense Industry?

[E. Robert Tisdale]

Tucker Taft wrote:

> Ken Garlington wrote:
>
> > Actually, as the person who started this thread,
> > I was talking about marketing.
> > It just seems odd to make choices
> > that damage your historical customer base (military)
> > before you've adequately captured your new base (non-military).
>
> I think we are interested in appealing to military customers.
> If you have recommendations of the best way to do so, please fire away.
> I honestly had the sense that broadcasting military success stories
> was actually counterproductive in some circles,
> including some military circles.

I'm not sure what choices Ken Garlington is talking about.

I'm not sure that it is counterproductive
to "broadcast" military success stories.
But I do have a very strong impression
that the military is far more interested in commercial successes
that can be used in military applications.

I worked on
the Vector, Signal and Image Processing Library (VSIPL)
Application Programmer's Interface (API) standard

    http://www.vsipl.org/

for a while.
Originally, I expected that there would be a great deal of interest
in an Ada language binding for the VSIPL API standard
but almost no one -- including the participants
from DARPA and the U.S. Navy -- was interested.
They weren't (and still aren't) even particularly concerned
about a C++ language binding for the VSIPL API standard.
The ANSI C language binding is the only one
which has been specified so far.
The reason is that, until recently, only ANSI C compilers
were available for embedded Digital Signal Processing chips
which are likely targets for a standard VSIPL implementation.
There are now very good C++ compilers for high end DSP chips
so I expect that the VSIPL Forum will, eventually, specify
a C++ language binding for the VSIPL API standard.
There are still almost no really good Ada compilers
for any embedded DSP chips
so I don't expect movement on an Ada language binding
for the VSIPL API standard any time in the near future.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"E. Robert Tisdale" <edwin@netwood.net> wrote in message
news:3A007746.E656E341@netwood.net...

: Originally, I expected that there would be a great deal of interest
: in an Ada language binding for the VSIPL API standard
: but almost no one -- including the participants
: from DARPA and the U.S. Navy -- was interested.

You may want to read the full thread to get caught up. In any case,
hopefully everyone now understands that, with acquisition reform, defense
contractors have a lot more latitude on the decisions they make in the
software area. Given that defense contractors do a large percentage of the
defense work, it's not a good idea to limit your defense marketing to the
Government.

: There are still almost no really good Ada compilers
: for any embedded DSP chips
: so I don't expect movement on an Ada language binding
: for the VSIPL API standard any time in the near future.

Actually, there is an Ada compiler for the SHARC, and apparently more than
one Ada compiler that can generate C code compatible with TI DSP processors.

Re: Is the Ada World Embarrassed by the Defense Industry?

[E. Robert Tisdale]

Ken Garlington wrote:

> E. Robert Tisdale wrote:
>
> : There are still almost no really good Ada compilers
> : for any embedded DSP chips
> : so I don't expect movement on an Ada language binding
> : for the VSIPL API standard any time in the near future.
>
> Actually, there is an Ada compiler for the SHARC,
> and apparently more than one Ada compiler
> that can generate C code compatible with TI DSP processors.

Can you cite a URL?

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"E. Robert Tisdale" <edwin@netwood.net> wrote in message
news:3A00B859.B31B77CB@netwood.net...
: Ken Garlington wrote:
:
: > E. Robert Tisdale wrote:
: >
: > : There are still almost no really good Ada compilers
: > : for any embedded DSP chips
: > : so I don't expect movement on an Ada language binding
: > : for the VSIPL API standard any time in the near future.
: >
: > Actually, there is an Ada compiler for the SHARC,
: > and apparently more than one Ada compiler
: > that can generate C code compatible with TI DSP processors.
:
: Can you cite a URL?

In no particular order:

General index:
http://adaic.org/docs/flyers/comp-tool.html

The Intermetrics/AverStar/Titan story (SHARC/C support):
http://www.analog.com/publications/press/products/adapr.html
http://www.averstar.com/services/IT_consulting.html#ada
http://www.averstar.com/news/titan.html

The ICC story (SHARC/C support)
http://www.irvine.com/products.html

Re: Is the Ada World Embarrassed by the Defense Industry?

[Jeff Carter]

Ken Garlington wrote:
> Actually, there is an Ada compiler for the SHARC, and apparently more than
> one Ada compiler that can generate C code compatible with TI DSP processors.

Wasn't it a Tartan Ada compiler for a TI DSP that produced smaller and
faster code than hand-optimized assembler in the famous "Ada beats
assembler" article?

-- 
Jeff Carter
"Perfidious English mouse-dropping hoarders."
Monty Python & the Holy Grail

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"Jeff Carter" <jrcarter@acm.org> wrote in message
news:3A00E427.70D5E7C@acm.org...
: Ken Garlington wrote:
: > Actually, there is an Ada compiler for the SHARC, and apparently more
than
: > one Ada compiler that can generate C code compatible with TI DSP
processors.
:
: Wasn't it a Tartan Ada compiler for a TI DSP that produced smaller and
: faster code than hand-optimized assembler in the famous "Ada beats
: assembler" article?

Yes, and in theory you can still buy that Ada83 compiler from TI (which
bought that part of Tartan). However, I don't get a good feeling as to the
level of support.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Gautier]

> : Wasn't it a Tartan Ada compiler for a TI DSP that produced smaller and
> : faster code than hand-optimized assembler in the famous "Ada beats
> : assembler" article?

Ken Garlington:

> Yes, and in theory you can still buy that Ada83 compiler from TI (which
> bought that part of Tartan). However, I don't get a good feeling as to the
> level of support.

I'm not a DSP guru, but there is at least a recent Ada95 for a TI chip:

http://www.ti.com/sc/docs/news/1999/99032.htm

______________________________________________________
Gautier  --  http://members.nbci.com/gdemont/gsoft.htm

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"Gautier" <gautier.demontmollin@maths.unine.ch> wrote in message
news:3A016D19.1E7FCC10@maths.unine.ch...
: > : Wasn't it a Tartan Ada compiler for a TI DSP that produced smaller and
: > : faster code than hand-optimized assembler in the famous "Ada beats
: > : assembler" article?
:
: Ken Garlington:
:
: > Yes, and in theory you can still buy that Ada83 compiler from TI (which
: > bought that part of Tartan). However, I don't get a good feeling as to
the
: > level of support.
:
: I'm not a DSP guru, but there is at least a recent Ada95 for a TI chip:
:
: http://www.ti.com/sc/docs/news/1999/99032.htm

This is the Irvine (ICC) compiler that I mentioned in a previous post. The
website, again, is http://www.irvine.com

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"Tucker Taft" <stt@averstar.com> wrote in message
news:3A006EB1.BA1B841F@averstar.com...
: Ken Garlington wrote:
: >
: > Actually, as the person who started this thread, I was talking about
: > marketing. It just seems odd to make choices that damage your historical
: > customer base (military) before you've adequately captured your new base
: > (non-military).
:
: I think we are interested in appealing to military customers.  If you have
: recommendations of the best way to do so, please fire away.
: I honestly had the sense that broadcasting military success stories
: was actually counterproductive in some circles, including some military
: circles.

Advertising military success stories, based on contracts where the language
was specified by the Government, is probably counter-productive. Advertising
success stories where Ada was chosen by a military contractor *on their
own*, based on a business case specific to defense applications, is much
more productive in the defense market. Defense contractors look at what
other companies in their same industry are doing, just like any other
business segment. I have heard more than once that statement "Once the Ada
mandate was dropped, all the other military contractors stopped using Ada,
so why should we use it?" You know what? If you look at what's been
published post-mandate, it's not an easy statement to refute.

At this point, it probably doesn't matter. However, I suspect that not
drawing this distinction in the decision to concentrate on non-defense
applications was a mistake.

Ada vs. C++ in defense projects

[Michael P. Card]

[-- Attachment #1: Type: text/plain, Size: 2871 bytes --]

Mike-

I am on a project now which is mixed Ada and C++. I find it frightening
that many defense contractors are pushing C++ for critical defense systems
like the one I am working on now, though thankfully I am on a team which is
working the Ada part. C++ is, IMO, an extremely poor choice for defense
systems for many reasons, not the least of which include portability,
readability, maintainability, and memory corruption due to invalid type
casts and over-writing array bounds. As far as I am concerned, there is no
financial or technical justification for using C++ on these kinds of
projects.

IMO, the only reason it is chosen by mgmt and some engineers is that most
of us remember the big defense downturn of the late 80's and early 90's. If
you want to go work for Microsoft or a dot com, or if you want your resume
ready just in case, it is a lot better to be able to say "I managed a team
of 50 C++ programmers and we developed a 40 KSLOC distributed real-time C++
application" or "As a S/W engineer at company X, I wrote 10 KSLOC of C++ on
my last project." Because of this "resume factor," engineers and managers
in the defense industry are willing (albeit often unintentional)
collaborators on the move to C++.

When you couple this with amazing trends like the preference for Windows NT
as the information infrastructure for the CVN-77 (new Navy carrier), you
can begin to believe that it would be in America's best interest for the
government to pay M$ to re-write Windows, Office, Access, Project and SQL
Server in Ada. The way I see it, M$ would like it since they could improve
their products at taxpayer expense, consumers would get more reliable
software, and the DoD would get a better infrastructure for the CVN-77 and
future projects! (tongue-in-cheek here)

- Mike

mjsilva@my-deja.com wrote:

> In article <39FE461D.275F1363@ix.netcom.com>,
>   Lao Xiao Hai <laoxhai@ix.netcom.com> wrote:
> >
> > Indeed!!!!!!??????   Most organizations that I see choosing
> > C++ over Ada have done very little in the way of careful
> > study.   Certainly no U.S. military organization has thought
> > this through very carefully.   It is, in fact, quite scary.  An
> > organization that could not manage a  single-language
> > policy is under the illusion that it can manage a multiple-
> > language policy.
>
> I'm still interested in hearing if any of those who have switched from
> Ada to The Radiant Future of language X are finding that there's
> trouble in paradise X.  It seems that if there is trouble in paradise X
> then we Ada advocates should be sure and document it for the benefit of
> others who are considering such a switch, or who are simply evaluating
> languages.  OTOH, if everybody is happy as a clam using X then I guess
> we need to rethink some of our assumptions.
>
> Mike
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

[-- Attachment #2: Card for Michael P. Card --]
[-- Type: text/x-vcard, Size: 344 bytes --]

begin:vcard 
n:Card;Michael
tel;fax:315-456-0441
tel;work:315-456-3022
x-mozilla-html:TRUE
org:Lockheed Martin ;Ocean, Radar, and Sensor Systems
version:2.1
email;internet:michael.p.card@lmco.com
title:Principal Software Engineer
adr;quoted-printable:;;Electronics Park=0D=0ABuilding 6, Room 201;Syracuse;NY;13221;USA
fn:Michael Card
end:vcard

Re: Ada vs. C++ in defense projects

[Jeff Stimson]

Most of my career has been based on writing applications in Ada.  That has
changed recently since absolutley no one
wants anything other than C++ right now.  I've just ramped up in C++ and am
doing projects with it.  I am not new to C since
I have also used that for a number of years.

My God.
Is everyone blind to the inherent dangers in programming with C++ ?  My last
project in Ada (about 20K SLOC) was delivered to the
customer about 8 months ago with no defects.  That's zero, zip, 0, nadda.
We have heard nothing back from them in terms of
problems or bugs.  I know for a fact that the current project (approx same
size) in C++ will not turn out the same.

I really despair at the dissapearance of Ada, and it is going away.  We can
fight and scream all we want but it is not being taught
in colleges or universities, it is not being promoted by companies that have
influence, and it still has a stigma with being 'that military
language'.

Sigh.


"Michael P. Card" <michael.p.card@lmco.com> wrote in message
news:3A02DB88.8A4232D1@lmco.com...
> Mike-
>
> I am on a project now which is mixed Ada and C++. I find it frightening
> that many defense contractors are pushing C++ for critical defense systems
> like the one I am working on now, though thankfully I am on a team which
is
> working the Ada part. C++ is, IMO, an extremely poor choice for defense
> systems for many reasons, not the least of which include portability,
> readability, maintainability, and memory corruption due to invalid type
> casts and over-writing array bounds. As far as I am concerned, there is no
> financial or technical justification for using C++ on these kinds of
> projects.
>
> IMO, the only reason it is chosen by mgmt and some engineers is that most
> of us remember the big defense downturn of the late 80's and early 90's.
If
> you want to go work for Microsoft or a dot com, or if you want your resume
> ready just in case, it is a lot better to be able to say "I managed a team
> of 50 C++ programmers and we developed a 40 KSLOC distributed real-time
C++
> application" or "As a S/W engineer at company X, I wrote 10 KSLOC of C++
on
> my last project." Because of this "resume factor," engineers and managers
> in the defense industry are willing (albeit often unintentional)
> collaborators on the move to C++.
>
> When you couple this with amazing trends like the preference for Windows
NT
> as the information infrastructure for the CVN-77 (new Navy carrier), you
> can begin to believe that it would be in America's best interest for the
> government to pay M$ to re-write Windows, Office, Access, Project and SQL
> Server in Ada. The way I see it, M$ would like it since they could improve
> their products at taxpayer expense, consumers would get more reliable
> software, and the DoD would get a better infrastructure for the CVN-77 and
> future projects! (tongue-in-cheek here)
>
> - Mike
>
> mjsilva@my-deja.com wrote:
>
> > In article <39FE461D.275F1363@ix.netcom.com>,
> >   Lao Xiao Hai <laoxhai@ix.netcom.com> wrote:
> > >
> > > Indeed!!!!!!??????   Most organizations that I see choosing
> > > C++ over Ada have done very little in the way of careful
> > > study.   Certainly no U.S. military organization has thought
> > > this through very carefully.   It is, in fact, quite scary.  An
> > > organization that could not manage a  single-language
> > > policy is under the illusion that it can manage a multiple-
> > > language policy.
> >
> > I'm still interested in hearing if any of those who have switched from
> > Ada to The Radiant Future of language X are finding that there's
> > trouble in paradise X.  It seems that if there is trouble in paradise X
> > then we Ada advocates should be sure and document it for the benefit of
> > others who are considering such a switch, or who are simply evaluating
> > languages.  OTOH, if everybody is happy as a clam using X then I guess
> > we need to rethink some of our assumptions.
> >
> > Mike
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>

Re: Ada vs. C++ in defense projects

[Robert Love]

>>>>> "Jeff" == Jeff Stimson <jstimson@home.com> writes:


    Jeff> My God.  Is everyone blind to the inherent dangers in
    Jeff> programming with C++ ?  My last project in Ada (about 20K
    Jeff> SLOC) was delivered to the customer about 8 months ago with
    Jeff> no defects.  That's zero, zip, 0, nadda.  We have heard
    Jeff> nothing back from them in terms of problems or bugs.  I know
    Jeff> for a fact that the current project (approx same size) in
    Jeff> C++ will not turn out the same.


Don't tell us, tell your customer.  Yeah, I know you don't go talking
bad about mgmt decisions to the cusotmer but you can find ways to
point this out so the customer uncovers this himself.

-- 
=============================================================
| Support Signature Minimalism                              |
=============================================================

Re: Ada vs. C++ in defense projects

[E. Robert Tisdale]

Jeff Stimson wrote:

> Most of my career has been based on writing applications in Ada.
> That has changed recently
> since absolutely no one wants anything other than C++ right now.
> I've just ramped up in C++ and am doing projects with it.
> I am not new to C since I have also used that for a number of years.
>
> My God.
> Is everyone blind to the inherent dangers in programming with C++?
> My last project in Ada (about 20K SLOC)
> was delivered to the customer about 8 months ago with no defects.
> That's zero, zip, 0, nadda.
> We have heard nothing back from them in terms of problems or bugs.
> I know for a fact that the current project (approx. same size) in C++
> will not turn out the same.
>
> I really despair at the disappearance of Ada, and it is going away.
> We can fight and scream all we want
> but it is not being taught in colleges or universities,
> it is not being promoted by companies that have influence,
> and it still has a stigma with being 'that military language'.

Ada has made you lazy and careless.
You can write programs in C that are just as safe
by the simple application of super-human diligence.

Re: Ada vs. C++ in defense projects

[Jeff Carter]

"E. Robert Tisdale" wrote:
> Ada has made you lazy and careless.
> You can write programs in C that are just as safe
> by the simple application of super-human diligence.

This is a great quotation. I wish every manager who was deciding to use
C/++ in a mission/safety critical application had to memorize it.

-- 
Jeff Carter
"You tiny-brained wipers of other people's bottoms!"
Monty Python & the Holy Grail

Re: Is the Ada World Embarrassed by the Defense Industry?

[Pascal Obry]

Most of your message is nonsense to me. But others have already pointed
this out. Let me just add:

"E. Robert Tisdale" <edwin@netwood.net> writes:

> It is easier to find and train C and C++ programmers
> than it is to find and train Ada programmers today
> so there is a strong incentive to prefer C or C++ over Ada.
> 

Maybe, but it is easier to find GOOD Ada programmers than C
programmers. Ada programmers have a strong background on
design, modularity, readability... at least they have heard
about that!

Pascal.

-- 

--|------------------------------------------------------------
--| Pascal Obry                               Team-Ada Member |
--|                                                           |
--| EDF-R&D-MTI-NTIC- T A I C                                 |
--|                       Intranet: http://cln46gb            |
--| Bureau N-023            e-mail: p.obry@der.edf.fr         |
--| 1 Av G�n�ral de Gaulle  voice : +33-1-47.65.50.91         |
--| 92141 Clamart CEDEX     fax   : +33-1-47.65.50.07         |
--| FRANCE                                                    |
--|------------------------------------------------------------
--|
--|         http://perso.wanadoo.fr/pascal.obry
--|
--|   "The best way to travel is by means of imagination"

Re: Is the Ada World Embarrassed by the Defense Industry?

[Jean St-Pierre]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1956 bytes --]

This is a typical case of "being right, but being dead".

You're right to say that an Ada programmer is forced to follow better
programming practices than your typical C programmer. However, I know at
least two companies that have rewritten the code of their software tools
from Ada to C++ mostly because they were not able to maintain a stable team
of Ada programmers at a reasonable cost (to be fair, there were also some
portability and compiler issues). Those Ada programmers were good, but
certainly not easy to find.


Pascal Obry <p.obry@der.edf.fr> wrote in message
news:uy9z5tox1.fsf@der.edf.fr...
>
> Most of your message is nonsense to me. But others have already pointed
> this out. Let me just add:
>
> "E. Robert Tisdale" <edwin@netwood.net> writes:
>
> > It is easier to find and train C and C++ programmers
> > than it is to find and train Ada programmers today
> > so there is a strong incentive to prefer C or C++ over Ada.
> >
>
> Maybe, but it is easier to find GOOD Ada programmers than C
> programmers. Ada programmers have a strong background on
> design, modularity, readability... at least they have heard
> about that!
>
> Pascal.
>
> --
>
> --|------------------------------------------------------------
> --| Pascal Obry                               Team-Ada Member |
> --|                                                           |
> --| EDF-R&D-MTI-NTIC- T A I C                                 |
> --|                       Intranet: http://cln46gb            |
> --| Bureau N-023            e-mail: p.obry@der.edf.fr         |
> --| 1 Av G�n�ral de Gaulle  voice : +33-1-47.65.50.91         |
> --| 92141 Clamart CEDEX     fax   : +33-1-47.65.50.07         |
> --| FRANCE                                                    |
> --|------------------------------------------------------------
> --|
> --|         http://perso.wanadoo.fr/pascal.obry
> --|
> --|   "The best way to travel is by means of imagination"

Re: Is the Ada World Embarrassed by the Defense Industry?

[Larry Kilgallen]

In article <8tmmat$e79$1@dns3.cae.ca>, "Jean St-Pierre" <jstp.xremovethisx@cae.ca> writes:
> This is a typical case of "being right, but being dead".
> 
> You're right to say that an Ada programmer is forced to follow better
> programming practices than your typical C programmer. However, I know at
> least two companies that have rewritten the code of their software tools
> from Ada to C++ mostly because they were not able to maintain a stable team
> of Ada programmers at a reasonable cost (to be fair, there were also some
> portability and compiler issues). Those Ada programmers were good, but
> certainly not easy to find.

So they hired guaranteed long-term C++ programmers who were incapable
of learning Ada ?

Re: Is the Ada World Embarrassed by the Defense Industry?

[Jean St-Pierre]

Larry Kilgallen <Kilgallen@eisner.decus.org.nospam> wrote in message
news:cKo3$rZJqqvZ@eisner.decus.org...
> In article <8tmmat$e79$1@dns3.cae.ca>, "Jean St-Pierre"
<jstp.xremovethisx@cae.ca> writes:
> > This is a typical case of "being right, but being dead".
> >
> > You're right to say that an Ada programmer is forced to follow better
> > programming practices than your typical C programmer. However, I know at
> > least two companies that have rewritten the code of their software tools
> > from Ada to C++ mostly because they were not able to maintain a stable
team
> > of Ada programmers at a reasonable cost (to be fair, there were also
some
> > portability and compiler issues). Those Ada programmers were good, but
> > certainly not easy to find.
>
> So they hired guaranteed long-term C++ programmers who were incapable
> of learning Ada ?

No, the turnover rate was not lower with the C++ programmers but it did not
jeopardize the tool development and maintenance, which was the case with the
Ada version. It was simply a logistical and financial decision (availability
of programmers, cost of training, cost of maintaining multi-platform port,
cost of compilers and cross-compilers, etc). The cost of training was due to
the fact that it was very difficult to find Ada-proficient people at the
time, so Ada training had to be paid for.

I'm not saying it was a technically good decision (I guess we will see in
the long term), but it was almost unavoidable, considering the
circumstances.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Wes Groleau]

> Maybe, but it is easier to find GOOD Ada programmers than C
> programmers. Ada programmers have a strong background on
> design, modularity, readability... at least they have heard
> about that!

Many of us (Ada folks) have also heard of information hiding 
and abstraction.

Many of us were also doing "object-oriented design" long before
the word became a "silver bullet."

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

Re: Is the Ada World Embarrassed by the Defense Industry?

[Wes Groleau]

> A lot can be done to incorporate safety and reliability
> into C and C++ compilers and class libraries
> but these languages are inherently unsafe
> and there is very little that can be done about it
> without changing the languages themselves.

One problem with COTS for military and similar problem domains is
that you need a system where reliability requirements are high on
the priority list and you are trying to implement it with components
designed with other priorities.

Another is that (in some cases) instead of specifying the requirements
that the problem domain needs, you try to reverse-engineer the
requirements of the components you have decided to use.

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

Re: Is the Ada World Embarrassed by the Defense Industry?

[mark_lundquist]

In article <39FDE9E4.35F615A6@netwood.net>,
  "E. Robert Tisdale" <edwin@netwood.net> wrote:
> Ken Garlington wrote:
>
> > While looking at the SIGAda 2000 web site,
> > I notice that the role of Ada in defense applications is minimized
> > (even after the explicit requirements in this area were dropped).
> > For example, the list of "recent" successful Ada-based systems
> > includes only commercial projects, some five years old,
> > although one of the most recent Ada success stories occurred
> > just a few days ago (October 24).  I also notice that
> > an interview last year with Tucker Taft included the statement,
> > "These days we're focused mostly on commercial success stories..."
> > I can understand wanting to promote commercial applications,
> > but isn't this going a little overboard?

Defense-related success stories do little to counter the myth that Ada
is a "military language".  On the other hand, years-old succes stories
of any kind do little to dispel the myth that "nobody uses Ada anymore".

>
> Apparently, national defense, and the U.S. Navy in particular,
> has finally turned toward Commercial Off-The-Shelf (COTS)
> solutions after encouraging a commercial computer industry
> for the past 50 years.

I think COTS is a natural area to focus on for growing the Ada market.

-- mark

Mark Lundquist
Senior Software Engineer
Rational Software


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[E. Robert Tisdale]

mark_lundquist@my-deja.com wrote:

> I think COTS is a natural area to focus on for growing the Ada market.

Perhaps.
But Ada programmers don't appear to be interested.
No Ada programmers have expressed an interest
in an Ada language binding for the VSIPL API standard.
I don't think that a single Ada programmer has ever participated
in the discussion on the VSIPL Form web page

    http://www.vsipl.org/

Ada programmers don't appear to be much interested
in numerical computing at all.
I don't think that anyone uses Ada for numerical computing anymore.
There are, at best, only some adapter interfaces
for numerical libraries written in Fortran and/or C.
I think that many people in the military believe that
they have been abandoned by the Ada community.
They have kept a candle in the window for decades now
and I don't think that you can blame them for being distracted
by more enthusiastic C and C++ programmers.
Ada has been given up for dead.

Re: Is the Ada World Embarrassed by the Defense Industry?

[mark_lundquist]

In article <3A031100.68F2B59@netwood.net>,
  "E. Robert Tisdale" <edwin@netwood.net> wrote:
> mark_lundquist@my-deja.com wrote:
>
> > I think COTS is a natural area to focus on for growing the Ada
market.
>
> Perhaps.
> But Ada programmers don't appear to be interested.

What would it mean for Ada programmers to be interested in growing the
Ada market in the COTS segment?  How would that growth area depend on
Ada programmers being interested?  (Seems more to depend upon the
interest of those who would *pay* Ada programmers :-)

> No Ada programmers have expressed an interest
> in an Ada language binding for the VSIPL API standard.
> I don't think that a single Ada programmer has ever participated
> in the discussion on the VSIPL Form web page
>
>     http://www.vsipl.org/

I can envision a number of possible reasons why that might be, so
without knowing anything about it, it's hard for me to draw any
conclusions...

You've got to admit that vector processing is a somewhat narrow
specialty, though.

> Ada programmers don't appear to be much interested
> in numerical computing at all.

I would say that programmers in general are not interested in numerical
computing!  Most of computing is not numerics.  (Nothing against it...)

You seem to be in a double-minority: deeply interested in numerics and
vector processing, and also an Ada advocate :-)

> I don't think that anyone uses Ada for numerical computing anymore.
> There are, at best, only some adapter interfaces
> for numerical libraries written in Fortran and/or C.

What would you like to see in the way of Ada support for numeric
computing?

> I think that many people in the military believe that
> they have been abandoned by the Ada community.

Interesting... what kind of support do you think these people in the
military would have wanted to have from the "Ada community"?  (Are you
talking about Ada vendors here, third-party tool vendors, or what?)

I work for Rational Software in the Ada products group, and have for a
long time -- way back for Verdix Corp. and then w/ Rational when the
two companies merged.  I was a developer until 1-1/2 years ago, then
moved to a developer role in a different part of Rational's business
for about a year, then went back to the Ada products group, now as part
of the product management team.  So I'm very interested in what people
think the Ada vendors ought to be doing, different perspectives on the
Ada market, where it's going and how to grow it, etc.

Mark Lundquist
Rational Software



Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[E. Robert Tisdale]

mark_lundquist@my-deja.com wrote:

> E. Robert Tisdale wrote:
>
> > mark_lundquist@my-deja.com wrote:
> >
> > > I think COTS is a natural area to focus on
> > > for growing the Ada market.
> >
> > Perhaps.
> > But Ada programmers don't appear to be interested.
>
> What would it mean for Ada programmers to be interested
> in growing the Ada market in the COTS segment?
> How would that growth area
> depend upon Ada programmers being interested?
> (Seems more to depend upon the interest of those
>  who would *pay* Ada programmers :-)
>
> > No Ada programmers have expressed an interest
> > in an Ada language binding for the VSIPL API standard.
> > I don't think that a single Ada programmer has ever participated
> > in the discussion on the VSIPL Form web page
> >
> >     http://www.vsipl.org/
>
> I can envision a number of possible reasons why that might be,
> so without knowing anything about it,
> it's hard for me to draw any conclusions...
>
> You've got to admit that
> vector processing is a somewhat narrow specialty, though.
>
> > Ada programmers don't appear to be much interested
> > in numerical computing at all.
>
> I would say that programmers in general
> are not interested in numerical computing!
> Most of computing is not numerics.  (Nothing against it...)
>
> You seem to be in a double-minority:
> deeply interested in numerics and vector processing,
> and also an Ada advocate :-)
>
> > I don't think that anyone uses Ada for numerical computing anymore.
> > There are, at best, only some adapter interfaces
> > for numerical libraries written in Fortran and/or C.
>
> What would you like to see in the way of Ada support
> for numeric computing?
>
> > I think that many people in the military believe that
> > they have been abandoned by the Ada community.
>
> Interesting...
> What kind of support do you think these people in the military
> would have wanted to have from the "Ada community"?
> (Are you talking about Ada vendors here,
>  third-party tool vendors, or what?)
>
> I work for Rational Software in the Ada products group
> and have for a long time -- way back for Verdix Corp.
> and then w/ Rational when the two companies merged.
> I was a developer until 1-1/2 years ago,
> then moved to a developer role in a different part of Rational's business
> for about a year, then went back to the Ada products group,
> now as part of the product management team.
> So I'm very interested in what people think
> the Ada vendors ought to be doing,
> different perspectives on the Ada market,
> where it's going and how to grow it, etc.

I would expect someone from the Ada community to
  participate in the VSIPL Forum (come to meetings),
  propose an Ada language binding for the VSIPL API
  and implement a high performance VSIPL library
  with an Ada language binding for one or more of the
  embedded DSP chips which support an Ada compiler.

Numerical applications,
especially digital signal and image processing applications,
are very near and dear to the heart of the U.S. Military.
The U.S. Military has been reaching out to the Ada community
for decades now but the Ada community does not reach back.
I can find dozens of high performance numerical libraries
implemented in C and C++ but almost none in Ada.

Should the Military train Ada programmers to maintain
applications written in Ada and also train them to program
in C an C++ so that they can maintain numerical applications?
I think that they believe that they are better off with programmers
who are proficient in a general purpose computer programming
language like C or C++ instead of investing in what appears to be
a special purpose computer programming language like Ada.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Larry Kilgallen]

In article <3A033C49.44E2CC2D@netwood.net>, "E. Robert Tisdale" <edwin@netwood.net> writes:

> I would expect someone from the Ada community to
>   participate in the VSIPL Forum (come to meetings),
>   propose an Ada language binding for the VSIPL API
>   and implement a high performance VSIPL library
>   with an Ada language binding for one or more of the
>   embedded DSP chips which support an Ada compiler.

So have you done that, and what were the results ?

Re: Is the Ada World Embarrassed by the Defense Industry?

[John Magness]

Just as a matter of personal curiosity, does Rational provide any cross
compilers for  processors commonly used for real time embedded systems?
Which chips?
John.
ps.  Does Rational have a version of Apex/Ada95 suitably priced for the
home user?
i.e. Why can't I buy a shrink wrapped Rational/Apex/Ada 95 ( and other
languages)
like I can for Code Warrior, Visual C++, and the former Borland products at
CompUSA ?  One of the reasons that C, C++ and Java are popular is that
students (and schools) can purchase compilers with decent IDEs at a low
price.  But of course this would put Rational in direct competition with
Microsoft, threatening any existing mutual business relationships.
John.

mark_lundquist@my-deja.com wrote:

> In article <3A031100.68F2B59@netwood.net>,
>   "E. Robert Tisdale" <edwin@netwood.net> wrote:
> > mark_lundquist@my-deja.com wrote:
> >
> > > I think COTS is a natural area to focus on for growing the Ada
> market.
> >
> > Perhaps.
> > But Ada programmers don't appear to be interested.
>
> What would it mean for Ada programmers to be interested in growing the
> Ada market in the COTS segment?  How would that growth area depend on
> Ada programmers being interested?  (Seems more to depend upon the
> interest of those who would *pay* Ada programmers :-)
>
> > No Ada programmers have expressed an interest
> > in an Ada language binding for the VSIPL API standard.
> > I don't think that a single Ada programmer has ever participated
> > in the discussion on the VSIPL Form web page
> >
> >     http://www.vsipl.org/
>
> I can envision a number of possible reasons why that might be, so
> without knowing anything about it, it's hard for me to draw any
> conclusions...
>
> You've got to admit that vector processing is a somewhat narrow
> specialty, though.
>
> > Ada programmers don't appear to be much interested
> > in numerical computing at all.
>
> I would say that programmers in general are not interested in numerical
> computing!  Most of computing is not numerics.  (Nothing against it...)
>
> You seem to be in a double-minority: deeply interested in numerics and
> vector processing, and also an Ada advocate :-)
>
> > I don't think that anyone uses Ada for numerical computing anymore.
> > There are, at best, only some adapter interfaces
> > for numerical libraries written in Fortran and/or C.
>
> What would you like to see in the way of Ada support for numeric
> computing?
>
> > I think that many people in the military believe that
> > they have been abandoned by the Ada community.
>
> Interesting... what kind of support do you think these people in the
> military would have wanted to have from the "Ada community"?  (Are you
> talking about Ada vendors here, third-party tool vendors, or what?)
>
> I work for Rational Software in the Ada products group, and have for a
> long time -- way back for Verdix Corp. and then w/ Rational when the
> two companies merged.  I was a developer until 1-1/2 years ago, then
> moved to a developer role in a different part of Rational's business
> for about a year, then went back to the Ada products group, now as part
> of the product management team.  So I'm very interested in what people
> think the Ada vendors ought to be doing, different perspectives on the
> Ada market, where it's going and how to grow it, etc.
>
> Mark Lundquist
> Rational Software
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ken Garlington]

"John Magness" <jmagness@swbell.net> wrote in message
news:3A16D250.70FD97C7@swbell.net...
: Just as a matter of personal curiosity, does Rational provide any cross
: compilers for  processors commonly used for real time embedded systems?
: Which chips?

According to their web site, Rational Apex Embedded supports the following
targets, several of which are used in real-time embedded systems:

PowerPC (602, 603, 603e, 604, 604e, 505, 8xx, 403),

The Intel Architecture

Motorola 68k family

MIPS I, II

Honeywell RH32

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ted Dennison]

John Magness wrote:

> i.e. Why can't I buy a shrink wrapped Rational/Apex/Ada 95 ( and other
> languages)
> like I can for Code Warrior, Visual C++, and the former Borland products at
> CompUSA ?  One of the reasons that C, C++ and Java are popular is that

Little-known fact: CompUSA actually *charges* publishers for shelf space in
their stores. A publisher has to have pretty deep pockets and a good
expectation of moving lots of product to even think about getting shelf space
in a national chain. Its generally not worth the expense unless your company
has multiple products to sell. This is the reason small game deveopment
companies are typically foced to sign themselves into servitude with one of the
major software publishing houses.


> students (and schools) can purchase compilers with decent IDEs at a low
> price.  But of course this would put Rational in direct competition with

Students (and schools) can purchase Ada compilers with decent IDEs at a low
price too. Aonix offers educational pricing for ObjectAda, and Gnat (IDE's
don't get much better than Emacs) is essentially free (can't get much cheaper
than free) for anyone who doesn't need priority support. ACT might even offers
schools discounts on support.

--
T.E.D.

Home - mailto:dennison@telepath.com  Work - mailto:dennison@ssd.fsi.com
WWW  - http://www.telepath.com/dennison/Ted/TED.html  ICQ  - 10545591

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ted Dennison]

In article <3A031100.68F2B59@netwood.net>,
  "E. Robert Tisdale" <edwin@netwood.net> wrote:
> I don't think that anyone uses Ada for numerical computing anymore.

I don't think that anyone uses Ada for numerical computing any *less*
either. :-)

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Gautier]

E. Robert Tisdale:

> Ada programmers don't appear to be much interested
> in numerical computing at all.
> I don't think that anyone uses Ada for numerical computing anymore.

I know at least _one_ person who does - and maybe in
an area where there has been no Ada programming at all
(finite elements - CFD). :-)

> There are, at best, only some adapter interfaces
> for numerical libraries written in Fortran and/or C.

Well, where is the problem ? It allows the few Ada programmer
to use debugged Fortran / C libraries, while cutting
their debug time in writing their own code in Ada!

______________________________________________________
Gautier  --  http://members.nbci.com/gdemont/gsoft.htm

Re: Is the Ada World Embarrassed by the Defense Industry?

[Laurent Guerby]

"E. Robert Tisdale" <edwin@netwood.net> writes:
> [...] Ada programmers don't appear to be much interested in numerical
> computing at all.  I don't think that anyone uses Ada for numerical
> computing anymore. [...]

That's funny. Any rocket, missile or transportation system (a domain
where Ada is used) is full of large pieces of numerical code written
in Ada.  At work, the Ada software I work on is full of numerical code
(equity derivative pricer: PDE, Monte Carlo, FFT, stats, etc...).

-- 
Laurent Guerby <guerby@acm.org>

Re: Is the Ada World Embarrassed by the Defense Industry?

[Ted Dennison]

In article <8666m0gby8.fsf@acm.org>,
  Laurent Guerby <guerby@acm.org> wrote:
> "E. Robert Tisdale" <edwin@netwood.net> writes:
> > [...] Ada programmers don't appear to be much interested in
> > numerical computing at all.  I don't think that anyone uses Ada for
> That's funny. Any rocket, missile or transportation system (a domain
> where Ada is used) is full of large pieces of numerical code written
> in Ada.  At work, the Ada software I work on is full of numerical code
> (equity derivative pricer: PDE, Monte Carlo, FFT, stats, etc...).

Our simulation code has quite a lot of it too. Anything that goes
through FAA certification has to match the reference aircraft to within
a rather small tolerance. That requires very accurate modelling of
forces and moments and such.

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[Lao Xiao Hai]

mark_lundquist@my-deja.com wrote:

> > Apparently, national defense, and the U.S. Navy in particular,
> > has finally turned toward Commercial Off-The-Shelf (COTS)
> > solutions after encouraging a commercial computer industry
> > for the past 50 years.
>
> I think COTS is a natural area to focus on for growing the Ada market.

As is illustrated by the now famous example of the USS Yorktown going
dead in the water due to a problem in its Microsoft-based software.

COTS products, when selected through the minimum technical review that
seems to characterize its management, are often as fraught with risk as
home-grown products.   At least with home-grown products, we can apply
some oversight to the developmenet process.

COTS is not a silver bullet.

Richard Riehle

Re: Is the Ada World Embarrassed by the Defense Industry?

[mark_lundquist]

In article <39FDE9E4.35F615A6@netwood.net>,
  "E. Robert Tisdale" <edwin@netwood.net> wrote:
>
> One can only assume that the commercial developer
> weighed these costs against all of the other costs
> relevant to application program development when
> they decided which programming language(s) to use.

You'd think...

But I have a hard time giving them the benefit of the doubt, because I
really do believe that if this were true we'd see Ada being used a lot
more!

So just out of curiosity, has anyone out there ever been involved with
or have first-hand knowledge of a project where an actual business case
analysis was undertaken with respect to language choice?  Where it was
really decided to try to quantify the expected cost/benefits for the
various alternatives and make the decision on that basis?  I'd be very
interested to hear about it...

Mark Lundquist
Rational Software


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Is the Ada World Embarrassed by the Defense Industry?

[DuckE]

> So just out of curiosity, has anyone out there ever been involved with
> or have first-hand knowledge of a project where an actual business case
> analysis was undertaken with respect to language choice?  Where it was
> really decided to try to quantify the expected cost/benefits for the
> various alternatives and make the decision on that basis?  I'd be very
> interested to hear about it...

In our group we don't tend to select language and development environments
by individual projects.  We select an environment that will meet our needs
for many projects for years to come.

A few years ago the development envrionment VAX VMS and target RTOS: VAXELN
started to approach their end of life.  Our applications were written in
VAXELN Pascal.

We were faced with many choices.  We had to choose a new target
architecture, new development environment, and new programming language.  We
could have stayed with Pascal, but would have had to change the dialect.

I had little difficulty finding resources on the Internet to argue the
advantages of moving to Ada 95.  No such resources were available (that I
could find) supporting moving to other environments.

Still the decision to move to Ada was a hard sell.  We had a meeting in our
department to make the decison on how to proceed.  I was able to present
numerous case studies in favor of moving to Ada.  The only arguement against
was "Nobody uses it".

We made that decision to use Ada, and have been thankful ever since.

SteveD

>
> Mark Lundquist
> Rational Software

Re: Is the Ada World Embarrassed by the Defense Industry?

[Frode Tennebø]

DuckE wrote:
[snip]
>
> I had little difficulty finding resources on the Internet to argue the
> advantages of moving to Ada 95.  No such resources were available (that I
> could find) supporting moving to other environments.
>
> We made that decision to use Ada, and have been thankful ever since.

Congratulation with a well-considered choise. Perhaps you could
compose a paper on this and put it on the Internet as a resource? :)

 -Frode


-- 
^ Frode Tenneb� | email: frodet@nvg.org | Frode@IRC  ^
| with Standard.Disclaimer; use Standard.Disclaimer; |

Re: Is the Ada World Embarrassed by the Defense Industry?

[mark]

In article <hh02u8.ddd.ln@leia>,
  Frode =?ISO-8859-1?Q?Tenneb=F8?= <frodet@nvg.org> wrote:
> DuckE wrote:
> [snip]
> >
> > I had little difficulty finding resources on the Internet to argue
the
> > advantages of moving to Ada 95.  No such resources were available
(that I
> > could find) supporting moving to other environments.
> >
> > We made that decision to use Ada, and have been thankful ever since.
>
> Congratulation with a well-considered choise. Perhaps you could
> compose a paper on this and put it on the Internet as a resource? :)
>
>  -Frode

Steve:

I'd also really encourage you to write something up.  But you might
consider publishing/presenting it somewhere where it would have more
impact than just floating it out on the net, where it's more likely to
be seen by people like us :-) than by those who aren't already
interested in Ada.

You might get in touch with Richard Riehle for some contacts, pointers
etc. about magazines and conferences -- he's written Ada articles for
JOOP, Embedded Systems Programming, etc. and is involved with the TOOLS
conference (http://www.tools-conferences.com/).

You also might get contact the Ada Resource Association for the same
kind of help, or other ideas about the best way to publicize your
story.  They can also put you in touch with writers who could co-author
something with you if you don't feel up to writing an article yourself.

What kind of application / industry sector are you in?  Are the
naysayers still grumbling, or are they saying "I guess Ada isn't so bad
after all", or are they saying "whew, I'm really glad we went with
Ada?"  Some quotes from people that were against Ada before would be
great.  What about the "problem" that "nobody uses Ada?"  Has there
been any negative impact from that at all, financially or otherwise?
Or was that really nothing more than a Chicken Little argument?  How
hard is to find excellent engineers who are willing to learn Ada, and
how expensive is it to train them?  That kind of stuff...

Interested!
-- mark

Mark Lundquist
Rational Software


Sent via Deja.com http://www.deja.com/
Before you buy.

Ada vs. C++ in defense projects

[michael_p_card]

Hello everyone (this is in response to Mike Silva, Richard Riehle et
al)-

I am on a project now which is mixed Ada and C++. I find it frightening
that many defense contractors are pushing C++ for critical defense
systems. C++ is, IMO, an extremely poor choice for defense systems for
many reasons, not the least of which include portability, readability,
maintainability, and memory corruption due to invalid type casts and
over-writing array bounds. As far as I am concerned, there is no
financial or technical justification for using C++ on these kinds of
projects. RE: the "high cost of training people in Ada," I know of an
excellent source of Ada training we have used in the past at about
$1K/student. I would wager we have spent far more than that per C++
programmer chasing bugs caused by uninitialized data structures, bad
type casts, overwritten array bounds etc.

IMO, the only reason C++ is chosen by mgmt and some engineers is that
most of us remember the big defense downturn of the late 80's and early
90's. If you want to go work for Microsoft or a dot com, or if you want
your resume ready just in case, it is a lot better to be able to say "I
managed a team of 50 C++ programmers and we developed a 40 KSLOC
distributed real-time C++ application" or "As a S/W engineer at company
X, I wrote 10 KSLOC of C++ on my last project." Because of this "resume
factor," engineers and managers in the defense industry are willing
(albeit often unintentional) collaborators on the move to C++.

When you couple this with amazing trends like the preference for Windows
NT as the information infrastructure for the CVN-77 (new Navy carrier),
you can begin to believe that it would be in America's best interest for
the government to pay M$ to re-write Windows, Office, Access, Project
and SQL Server in Ada. The way I see it, M$ would like it since they
could improve their products at taxpayer expense, consumers would get
more reliable software, and the DoD would get a better infrastructure
for the CVN-77 and future projects! (tongue-in-cheek here)

- Mike


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Ada vs. C++ in defense projects

[Tom Hargraves]

The US company I work for are currently bidding for a multi million $ US
Navy project and specifying java as the implementation language. This proves
there is no logic in this world, and is the reason I am learning Java. Such
is life...

I am just thankful they didn't choose C++

Tom H.


<michael_p_card@my-deja.com> wrote in message
news:8tv04t$n80$1@nnrp1.deja.com...
> Hello everyone (this is in response to Mike Silva, Richard Riehle et
> al)-
>
> I am on a project now which is mixed Ada and C++. I find it frightening
> that many defense contractors are pushing C++ for critical defense
> systems. C++ is, IMO, an extremely poor choice for defense systems for
> many reasons, not the least of which include portability, readability,
> maintainability, and memory corruption due to invalid type casts and
> over-writing array bounds. As far as I am concerned, there is no
> financial or technical justification for using C++ on these kinds of
> projects. RE: the "high cost of training people in Ada," I know of an
> excellent source of Ada training we have used in the past at about
> $1K/student. I would wager we have spent far more than that per C++
> programmer chasing bugs caused by uninitialized data structures, bad
> type casts, overwritten array bounds etc.
>
> IMO, the only reason C++ is chosen by mgmt and some engineers is that
> most of us remember the big defense downturn of the late 80's and early
> 90's. If you want to go work for Microsoft or a dot com, or if you want
> your resume ready just in case, it is a lot better to be able to say "I
> managed a team of 50 C++ programmers and we developed a 40 KSLOC
> distributed real-time C++ application" or "As a S/W engineer at company
> X, I wrote 10 KSLOC of C++ on my last project." Because of this "resume
> factor," engineers and managers in the defense industry are willing
> (albeit often unintentional) collaborators on the move to C++.
>
> When you couple this with amazing trends like the preference for Windows
> NT as the information infrastructure for the CVN-77 (new Navy carrier),
> you can begin to believe that it would be in America's best interest for
> the government to pay M$ to re-write Windows, Office, Access, Project
> and SQL Server in Ada. The way I see it, M$ would like it since they
> could improve their products at taxpayer expense, consumers would get
> more reliable software, and the DoD would get a better infrastructure
> for the CVN-77 and future projects! (tongue-in-cheek here)
>
> - Mike
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Re: Ada vs. C++ in defense projects

[E. Robert Tisdale]

Tom Hargraves wrote:

> The US company I work for are currently bidding
> for a multi million $ U.S. Navy project and
> specifying java as the implementation language.
> This proves there is no logic in this world,
> and is the reason I am learning Java.

Are you sure?

> Such is life...
>
> I am just thankful they didn't choose C++

In the comp.lang.c++ newsgroup,
Subject: Re: C++ Interpreter,
David Forrai <d.forrai-@-ieee.org> wrote:

> There is a C++ virtual machine in development called Internet C++.
> I have no comment/experience with this system.
> You can get info at
>
>     http://members.xoom.com/icvmcpp/

The idea is to compile the Java source code to "byte code"
that can execute on any platform
which supports a Java Virtual Machine (JVM).

If you can find an Ada compiler which emits byte code for a JVM,
you may be able to convince them that they should implement
safety critical applications in Ada instead of Java.
Otherwise, you should probably just forget about Ada.

Re: Ada vs. C++ in defense projects

[Pat Rogers]

"E. Robert Tisdale" <edwin@netwood.net> wrote in message
news:3A04C67F.FEB1D90A@netwood.net...
<snip>
> If you can find an Ada compiler which emits byte code for a JVM,
> you may be able to convince them that they should implement
> safety critical applications in Ada instead of Java.
> Otherwise, you should probably just forget about Ada.

I'm curious about something, based upon this and other posts by you. I
know you've done some work in C++ and have submitted posts on the
subject in other newsgroups.  What is your professional experience
with Ada?

pat rogers
progers@classwide.com

Re: Ada vs. C++ in defense projects

[Robert Dewar]

In article <3A04C67F.FEB1D90A@netwood.net>,
  "E. Robert Tisdale" <edwin@netwood.net> wrote:
> If you can find an Ada compiler which emits byte code for a
> JVM,
> you may be able to convince them that they should implement
> safety critical applications in Ada instead of Java.
> Otherwise, you should probably just forget about Ada.


JGNAT from Ada Core Technologies is such a system ....

Robert Dewar
Ada Core Technologies
www.gnat.com


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Ada vs. C++ in defense projects

[E. Robert Tisdale]

Robert Dewar wrote:

> E. Robert Tisdale wrote:
> > If you can find an Ada compiler which emits byte code for a
> > JVM,
> > you may be able to convince them that they should implement
> > safety critical applications in Ada instead of Java.
> > Otherwise, you should probably just forget about Ada.
>
> JGNAT from Ada Core Technologies is such a system ....
>
> Robert Dewar
> Ada Core Technologies
> www.gnat.com

I only found a short announcement

    Ada Core Technologies and ACT Europe
    announce JGNAT Professional release 1.1a. JGNAT Professional,
    the Ada 95 development environment for the Java platform,
    is a port of the GNAT Ada 95 technology to the Java environment.
    JGNAT supports the development of both applications and applets.

    JGNAT comprises a compiler generating Java bytecode
    fully compatible with Java virtual machines
    conforming to Sun's specification (JDK 1.1 and above),
    and a set of tools to aid in developing Ada programs for the Java platform.
    Contact sales@gnat.com for more information or contact:

at

    http://www.gnat.com/

by clicking on

    Announcing JGNAT Pro release 1.1a

Re: Ada vs. C++ in defense projects

[David C. Hoos, Sr.]

JGNAT, along with some papers on it is freely available at
ftp://ftp.cs.nyu.edu/pub/gnat/jgnat/

E. Robert Tisdale <edwin@netwood.net> wrote in message
news:3A04F336.B780D887@netwood.net...
> Robert Dewar wrote:
>
> > E. Robert Tisdale wrote:
> > > If you can find an Ada compiler which emits byte code for a
> > > JVM,
> > > you may be able to convince them that they should implement
> > > safety critical applications in Ada instead of Java.
> > > Otherwise, you should probably just forget about Ada.
> >
> > JGNAT from Ada Core Technologies is such a system ....
> >
> > Robert Dewar
> > Ada Core Technologies
> > www.gnat.com
>
> I only found a short announcement
>
>     Ada Core Technologies and ACT Europe
>     announce JGNAT Professional release 1.1a. JGNAT Professional,
>     the Ada 95 development environment for the Java platform,
>     is a port of the GNAT Ada 95 technology to the Java environment.
>     JGNAT supports the development of both applications and applets.
>
>     JGNAT comprises a compiler generating Java bytecode
>     fully compatible with Java virtual machines
>     conforming to Sun's specification (JDK 1.1 and above),
>     and a set of tools to aid in developing Ada programs for the Java
platform.
>     Contact sales@gnat.com for more information or contact:
>
> at
>
>     http://www.gnat.com/
>
> by clicking on
>
>     Announcing JGNAT Pro release 1.1a
>

Re: Ada vs. C++ in defense projects

[John Griffiths]

Question : has the (any,all) JVM been validated for safety critical
applications?

TTFN John

"Robert Dewar" <robert_dewar@my-deja.com> wrote in message
news:8u2o3d$hft$1@nnrp1.deja.com...
> In article <3A04C67F.FEB1D90A@netwood.net>,
>   "E. Robert Tisdale" <edwin@netwood.net> wrote:
> > If you can find an Ada compiler which emits byte code for a
> > JVM,
> > you may be able to convince them that they should implement
> > safety critical applications in Ada instead of Java.
> > Otherwise, you should probably just forget about Ada.
<snip/>