What ever happened to Ada

What ever happened to Ada

[Noam Kloos]

[-- Attachment #1: ada.txt --]
[-- Type: text/plain, Size: 1655 bytes --]


Was: Ada on old and Simple systems

What happened to the original philosphy of Ada usage?
I asked a quite general question how to use Ada programs on 8088
and 80286 machines.
I received response in several ways like why don't I buy a 80386 or an
offer to buy an old Ada compiler for 8088 machines.

It is just that i was testdriving the Gnat for Windows95 the first time
and simply compiled the hello.adb. Then tried to run the executable on
an old XT 8088 machine and got the message : 'this program cannot be run
in MSDOS mode'.
This seems to me a bit absurd and analog to over bureaucracy.

I hear programming Ada on an archaic system like the 8088 running DOS is
very slow and hard to use with tasking.

I don't see Ada to be used only for the high power abilities like
concurrency. My Idea of the power of Ada is using a wide range of
platforms with using and reusing a common software library so
maintenance of the combined informationsystems at hand can be optimized
and simplified.

To me it is not so the issue of getting the most out of Ada. When
writing code for less advanced systems and devices there would need to
be a less advanced way of programming too, but at least it can be tested
and maintaned and simpulated on a host computer.

I think the ideal situation is having a main development system
generating code for the target systems with consideration of less able
systems and devices.

I think this is how it was ment to be and maybe was at DoD in the old
days, however it may be another dream that has not come true.

I still see inflexibility and a too close focus on trends like java, the
web and high performance machines.

Re: What ever happened to Ada

[Tom Griest]

Noam Kloos <noam@3wis.nl> wrote in article <328848A2.53FC9822@3wis.nl>...
> Was: Ada on old and Simple systems
> 
> What happened to the original philosphy of Ada usage?
> I asked a quite general question how to use Ada programs on 8088
> and 80286 machines.
> I received response in several ways like why don't I buy a 80386 or an
> offer to buy an old Ada compiler for 8088 machines.
> 
> It is just that i was testdriving the Gnat for Windows95 the first time
> and simply compiled the hello.adb. Then tried to run the executable on
> an old XT 8088 machine and got the message : 'this program cannot be run
> in MSDOS mode'.
> This seems to me a bit absurd and analog to over bureaucracy.

When you have something called: "Gnat for Windows95", why do you
think that it should generate code for a platform that cannot
run Win95?  It is not a cross compiler after all.
 
What would really be absurd is to have a Pentium Pro processor that
could not address over 640K of memory, which would be the limitation
if your restricted the GNAT/Win32 code generation to the programming
model of IBM PC/XT.  There is no reason why you couldn't have a
GNAT cross-compiler target the PC/XT... except economics.  The fact
of the matter is that these machines are beyond their useful life.
Except for the possible special case (like using them as a terminal)
they usually cost more to operate (if you include the value of the 
operator's time) than it takes to upgrade them.

What IS reasonable is to take an OLD Ada compiler and be able to
compile the SAME PROGRAM using the OLD compiler for the OLD PC/XT
and a new compiler for the new Win95.  This can be done, and it seems
that this is all that is needed to satisfy your "dream" of Ada.

-Tom

Re: What ever happened to Ada

[Robert Dewar]

Noam Kloos said

  "It is just that i was testdriving the Gnat for Windows95 the first time
  and simply compiled the hello.adb. Then tried to run the executable on
  an old XT 8088 machine and got the message : 'this program cannot be run
  in MSDOS mode'.
  This seems to me a bit absurd and analog to over bureaucracy."

Why on earth would you expect a 386 executable to run on an 8088? These
are radically different arhitectures, you might as well have tried running
the executable on a Mac or a Unix machine!

It would certainly be possible to make a version of GNAT that generatd
executables for an 8088. It just has not been done. If someone is interested
and has a commercial requirement for such a product, it might well get done,
but to think that it is absurd that 386 executables do not run on an 8088
is very curious!

If you are expecting someone to present you with a free Ada 95 compiler
that runs on an 8088, I don't think it will happen. There simply is not
enough interest (we have seen ZERO interest in such a product) for someone
to invest resources in such a project.

The advice that you get a 386 represents a suggestion that may be much
less expensive for you than paying for an 8088 compiler to be developed :-)

Re: What ever happened to Ada

[Fergus Henderson]

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

>Noam Kloos said
>
>  "It is just that i was testdriving the Gnat for Windows95 the first time
>  and simply compiled the hello.adb. Then tried to run the executable on
>  an old XT 8088 machine and got the message : 'this program cannot be run
>  in MSDOS mode'.
>  This seems to me a bit absurd and analog to over bureaucracy."
>
>Why on earth would you expect a 386 executable to run on an 8088? These
>are radically different arhitectures, you might as well have tried running
>the executable on a Mac or a Unix machine!

It sounds like Noam Kloos might have been happier if his XT had just
crashed, rather than getting a sensible error message.  At least then
he wouldn't have thought the problem was due to "over bureaurocracy".

Perhaps this explains the success of C ;-)

--
Fergus Henderson <fjh@cs.mu.oz.au>   |  "I have always known that the pursuit
WWW: <http://www.cs.mu.oz.au/~fjh>   |  of excellence is a lethal habit"
PGP: finger fjh@128.250.37.3         |     -- the last words of T. S. Garp.