Re: Government Policy on Ada Acquisitions

[mfeldman@seas.gwu.edu (Michael Feldman)]

In article <34kef8$l9f@jac.zko.dec.com>,
Bevin R. Brett <brett@ada9x.enet.dec.com> wrote:

[this is the only point I'm going to respond to]
>
>Ada'83 was too hard to learn
>----------------------------
>
>It has taken 10 years to figure out that an Ada'83 subset is easier to learn
>than C or Pascal or Cobol.  This realisation would have been valuable in 1982,
>but instead we pushed all the exotic features...

Who are you referring to as "we"? We in the universities, some of whom have
been teaching CS1 and CS2 courses with Ada for ten years (names on request)
have known this all along, and acted on it. We've known how to teach Ada
and done so quite well. Unfortunately, the industry gave us little
cooperation in helping us spread the word (compiler/tool pricing, etc.),
but we've been there all along, teaching Ada in increasingly large
subsets in each subsequent course. Industry tended to label us as
insufferably "academic".


[well, OK, I'll respond to this one too.]

>Ada'9X is too late and too complex
>----------------------------------

>C[++] is a poor language, but when it is the only language that supports 99% of
>the bindings and reusable components available, it is going to be a very
>common language - which in turn means it will be the language used for
>bindings and reusable components...

I have a really wacky, fuzzy-headed academic sort-of idea.

How 'bout if all the Ada companies who are sitting and bitching about
the lack of tools, and fighting GNAT at every turn, 

- adopted GNAT as their core compiler (Can't make money doing an Ada 
  94 compiler? Take the free one, then!), ported it to every machine
  in sight;

- turned their creativity to making the world's best GNAT-compatible tools 
  (which can damn well be proprietary), and competing with other on
  _that_ battlefield; and

- pooled their compiler development funds to ensure that GNAT will
  blow the doors off every C++ compiler in sight? (That is, sign
  contracts with ACT, or spin off another company like ACT.)

Make GNAT _the_ standard Ada 94 compiler, sorta just like DOS. Remember 
Mister Gillette: give away the razor; soak 'em for the blades. Compete
with the tools instead, and the bindings, and the safety-critcal
runtime libraries, and...

Nah. Too academic. Right?

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5253 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
"Pork is all that stuff the government gives the other guys."
------------------------------------------------------------------------