Re: verdix kisses off Ada

[seas.gwu.edu!mfeldman@uunet.uu.net (Michael Feldman)]

In article <9575@verdix.verdix.com> brucej@verdix.com (Bruce Jones) writes:
>In article <1993May21.162150.20535@seas.gwu.edu> mfeldman@seas.gwu.edu (Michae
l Feldman) writes:
>>A dollar can only be spent once. If the resources aren't there to build bigge
r
>>Ada markets, how are they available to build C/C++ compilers? 
>
>Mike, its not just a question of dollars, the reason is that the resources 
>are different.  The resources needed to build a C/C++ compiler is a team 
>of highly skilled compiler engineers.  Verdix has such a team.  You can't
>just go buy these people, you have to build the team slowly over time.  

Oh, that's certainly true. But each person-hour can be spent only once.
Each hour invested in building a C or C++ compiler is one that's _not_
spent developing Ada stuff whose performance and reliability is such that
people who are spending their own money will buy it.
>
>I don't know what resources are needed to build a bigger Ada market, but then
>I'm just one of the compiler engineers. I am sure that it takes more than
>just money.  I suspect that GNU Ada and Ada 9x will do more to increase 
>the size of the Ada market than anything we vendors might do.

IMHO that's a copout. I have dealt with all the major Ada vendors for 10 
years.  Each time it's been suggested to them about how they could improve
the perception and reality of Ada in the world outside the mandate, they've
come back with lots of ways _someone else_ could do this.

There is no doubt that GNAT will improve accessibility to Ada for
companies and universities interested in getting started with Ada,
without paying the huge prices (relative to other languages) requested
by the Ada companies. This lets the Ada companies off the hook: they
are pushing the responsibility for Ada's future onto others - in this
case, NYU (which is building GNAT) and Uncle Sam (which is funding it). 

Ask any teacher what a vendor could do to help Ada along. You don't have
to wait for Ada9X, or wait for GNAT. We've been telling you for years
how you could build the grass-roots support for Ada by starting in the
schools, just like the hardware companies do. 

The vendors - without even trying - have responded with idiocy like per-seat 
pricing, per-seat support bills, and generally, unfriendly approaches. Asked 
why, they respond that they'd "go broke supporting schools", as though they 
didn't know that compilers used for teaching require almost no support.
(I have to violate my usual policy of not singling out vendors, because -
in my own 10-year experience - Verdix has been by far the worst offender here.)

The net has seen my diatribes on this before; I won't bore you again. But
if the vendors think there's nothing they can do but move to C++, they are
surpassingly myopic. Maybe you and I realize this, but the business types
running the companies just don't get it. We predicted years ago - well
before the C++ wave started, even - that their insistence on Ada "being a
small market" and taking no risk to expand it would guarantee that Ada
would remain a small market. 

We told them so; I can show you chapter and verse where we did. They just 
don't get it. Our prediction has, sadly, come true. And if you think that 
Ada9X or GNAT is miraculously going to change things, think again. Without 
a paradigm shift from the companies, nothing much will change. I'd like to 
be optimistic, but the growing investment of Ada houses in C++ tells me that 
I can't be. Where am I reading about Verdix putting its compiler engineers
to work building an Ada9X compiler that'll blow the doors off C++?

Steve Ziegler's response to the net, for all its warm and fuzzy wording,
is still saying that Verdix would prefer investing in C++ to investing in
really good Ada stuff that real, commercial, profit-oriented organizations
will buy with their own money. 

Wanna bet that Verdix' C++ compilers will be cheaper than their Ada compilers?

Enough.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman
co-chair, SIGAda Education Committee

Professor, Dept. of Electrical Engineering and Computer Science
School of Engineering and Applied Science
The George Washington University
Washington, DC 20052 USA
(202) 994-5253 (voice)
(202) 994-5296 (fax)
mfeldman@seas.gwu.edu (Internet)

"The most important thing is to be sincere, 
and once you've learned how to fake that, you've got it made." 
-- old show-business adage
------------------------------------------------------------------------