re: verdix kisses off Ada

[dog.ee.lbl.gov!network.ucsd.edu!swrinde!emory!ogicse!verdix!sfz@ucbvax.Be rkeley.EDU (Stephen Zeigler)]

Verdix Corporation has been described as "kissing off Ada."  As Vice
President of Compiler Products for Verdix, I would like to correct
the record.  The following posting was made to comp.lang.ada:

> >From the May 6, 1993 issue of Washington Technology, page 3:
> 
>     Verdix Corporation, a leading supplier of software engineering
>     products using the Ada programming language, is attempting to
>     sell its security products division.  The firm is looking to
>     refocus resources on its core software programming business,
>     particularly efforts in products for the C++ computer language,
>     which has eclipsed use of Ada outside of the Pentagon.
> 
> 
>     What will it take for the DoD to see the light and drop the Mandate?
> For all Ada compiler companies to fail and/or refocus on C++? For all Ada
> components vendors to fail and/or switch over to C++?  For all of the
> services to pay lip service to the Mandate? For DoD software developments
> costs to go through the roof?   WHAT??????
> 
> Greg Aharonian
> -- 
> **************************************************************************
> Greg Aharonian
>  Source Translation & Optimiztion
>  P.O. Box 404, Belmont, MA 02178

The above quoted article seems to have left a misimpression, at least
with Mr. Aharonian.  Verdix continues to invest
enthusiastically in its Ada tool technology, which supplies a complete set
of development tools to a wide range of customers, for most popular hosts
and targets.  Verdix Ada technology is also supplied by OEMs such as Sun
Microsystems and Silicon Graphics.

Verdix does internal development in C, C++ and Ada, but generally has found
that Ada is the most cost effective language for new development.  We follow
similar guidelines to the DoD: if an internal module is to be more than
30% recoded we will use Ada.  We have records of all changes ever made in the
development of Verdix products; those records show us that for the lifecycle
Ada seems to be about 2x or better for cost effectiveness than C.  C++ numbers
are too early to show meaningful results.  In general we expect C++ to be best
for integrating known "objects" when writing relatively concise new programs,
while Ada and later Ada9x are better for realtime, for precise and for
extensively new programs.

Our customer base seems, for the most part, happy with our products and
with Ada.  We are happy that our Ada business has shown consistent growth and
profitability for the last 6-7 years, thanks to those happy customers - Verdix
does most of its marketing by word-of-mouth amongst its potential customers.

The purpose of the Ada mandate has been to encourage use of a new technology
through its long introductory period, when many understood that such complex
new technology would require substantial investment and time.  Many new
technologies receive support for decades before they become cost effective.
The Ada mandate is there because the DoD realized that decisions on its
complex, extensive software should not be made on the basis of making code
easy to write, but rather making code easier to maintain.  The commercial
world has different goals; in most industries it is better to get something
in the market quickly than it is to be as concerned with long-term maintenance,
since a maintenance stream is a gold mine to the supplier.  Thus some of the
important motivators for Ada are weakened in the commercial world today, with
exceptions for code needing high reliability such as airplanes.

Even so, Verdix is patiently building ever-improving Ada technology with
the expectation that competitive forces and natural business evolution will
increase the use of Ada both inside and outside of the DoD.

Ada provides several important facillities not available yet in C/C++.
Verdix is introducing VADSmp and DADS this year: VADSmp lets Ada tasks take
advantage of shared memory multiprocessing, while the more ambitious DADS
(Distributed Applications Development System) does the same for Ada users
atop homogenous distributed systems.  Both products perform multiprocessing
transparent to the Ada programmer:  the same build/debug tools are all
there just as for VADS.  These product lines provide significant, indeed some
have said "revolutionary", new capabilities to Ada users, in production
quality form not available in C, C++ or other mainstream languages.  Verdix
is seeing commercial interest in Ada already, and with these new products
we hope that Ada usage can accelerate.

Our C++/C efforts are equally serious.  They represent our conviction that
languages are like tools in other industries - it's not good to try to hammer
with a saw or to saw with a hammer.  Our company is dedicated to making
compatible tools, so that our customers are free to choose the right tool for
the right job.  Our clients should be able to mix Ada, C and C++, while using
the power of DADS and VADSmp.
Organization: Verdix Western Operations, Aloha  OR
Keywords: