Re: Is GNU-Ada self-insulting?

Re: Is GNU-Ada self-insulting?

[Gregory Aharonian]

>This begs a question in my mind:
>	How much flaming would be going on if the DoD had chosen
>	to *pay* a commercial company to develop a PD Ada compiler
>	(or god-forbid, a defense contractor!)?

Wake up.  Remember the Navy's effort with either Softech or Intermetrics
to develop the Ada ALS Ada Language System.  Talk about millions down the
drain.  The point in general is that the DoD should not be funding products
or fostering technologies (the waste of STARS), but instead should rely on
the free markets it is defending.

>Looks like having FSF, who do make some excellent software, perform
>the PD Ada projects is a good business decision to me.

As someone else has pointed out, GNU Ada most likely will undercut sales
for the few Ada vendors still in business.  It's tough enough to run a
business when the government competes with you, but it's worse when the
government funds your competitor.  In both cases, it shouldn't.

Greg Aharonian
-- 
**************************************************************************
Greg Aharonian
Source Translation & Optimiztion
P.O. Box 404, Belmont, MA 02178

Re: Is GNU-Ada self-insulting?

[Paul Robichaux]

In <SRCTRAN.92Dec1223112@world.std.com> srctran@world.std.com (Gregory Aharonia
n) writes:

>Wake up.  Remember the Navy's effort with either Softech or Intermetrics
>to develop the Ada ALS Ada Language System.  Talk about millions down the
>drain.  The point in general is that the DoD should not be funding products
>or fostering technologies (the waste of STARS), but instead should rely on
>the free markets it is defending.
 
{ Disclaimer: I'm a former ALS/N developer and Control Data employee. }

Greg, I don't think you're giving a full and accurate picture here.

ALS was originally an Army project. The Navy decided to take advantage
of the ALS effort, such as it was, and recycle large portions of ALS
into a Navy variant- ALS/N, which was targeted at three embedded
systems (AN/AYK-14, AN/UYK-44, and AN/UYK-43) and the VAX.

Softech had the original contract and developed the first versions of
the ALS/N tools. Softech _lost_ the contract and Control Data
Government Systems won it during a recompete.

When I left last August, we had four validated compilers against ACVC
1.11. We had _two_ failed PIWG tests and several hundred satisfied
Navy and contractor users.

The Navy funded ALS/N for the right reason: it supported embedded
computers of which they were the sole user. Newer systems (Phalanx
Block II comes to mind) are increasingly using off-the-shelf hardware
(MIPS R4000s and the Verdix compiler in this case.) At the time ALS/N
was started, though, the 14/44/43 systems were in need of an Ada
toolset.

-Paul
-- 
Paul Robichaux, KD4JZG              | May explode if disposed of improperly.
Mission Software Development Div.   | Printed on recycled phosphors.
New Technology, Inc.		    | ** PGP 2.0 key available on request **