ARPA blows $100 million chance to help Ada

ARPA blows $100 million chance to help Ada

[Gregory Aharonian]

     The August 30, 1993, issue of Electronic Engineering Times, page 20
has an article titled "Ada promotes commercial CAD-tool use" in which it
starts out with:

	"The Pentagon wants to use commercial CAD tools to sharply
	reduce the time needed to get embedded signal processors
	into the field.  So it will spend nearly $100 million over
	the next four years to get the costly tools into the hands
	of defense electronics companies.  ARPA is investing the
	time and money to develop a network for commercial design
	tools needed to speed prototyping.  Mark Richards is the
	ARPA contact point.

        Lockheed Sanders is getting $42.4 million, and Marietta $53.4
        million, and according to ARPA, after four years, industry
        will pick up the ball.  "Commercializating it means the
	program is self-sustaining by industry", Richards says.

     In two ways, this is a great example of how the DoD's leading software
research agency absolutely despises the Ada programming language.  First,
it seems that when it cares about something, ARPA is willing to intervene
into the private sector.  Two examples are this CAD expenditure, and its
attempts over the past two years to influence the future of the parallel
processing industry by dictating which computers its contractors could buy.
So when it cares about something, ARPA is willing to spend BIG time.
Apparently not so with Ada.

     Second, ARPA currently has no plans to have some of this $100 million
spent on Ada, for example, having its contractors develop hardware CAD
design tools using VHDL that have integrated hooks to Ada, an easy task
convert the language similarities between Ada and VHDL.  After all the
embedded hardware being designed with these tools has to be programmed,
presumably in Ada, so integrating Ada and VHDL in these tools make sense
in itself, if not for helping to push the Ada language.

     What's the easiest way to kill Ada consideration outside the Mandated
world by an advocate of another language?  "Why waste time with Ada?  If
ARPA thinks the language is useless, shouldn't we also?".

    Since ARPA has $100 million to play with intervening in the commercial
markets, if the DoD is even marginally supportive of Ada, at least a few
million should be attached to some Ada/VHDL stuff.  IF.

-- 
**************************************************************************
 Greg Aharonian                                      srctran@world.std.com
 Source Translation & Optimization                            617-489-3727
 P.O. Box 404, Belmont, MA 02178