comp.lang.ada
 help / color / mirror / Atom feed
From: Richard Riehle <richard@adaworks.com>
Subject: Re: Silicon Valley techies suit up Army with sleeker gear
Date: Tue, 12 Feb 2002 08:53:42 -0800
Date: 2002-02-12T16:46:46+00:00	[thread overview]
Message-ID: <3C694896.6B4AA199@adaworks.com> (raw)
In-Reply-To: 3C677CF2.2060806@mail.com

Hyman Rosen wrote:

> Richard Riehle wrote:
> > This is not a language issue.   It is a competency issue.
>
> Heh. You know I'm eventually going to have to quote this
> back to you :-)

Most software debacles of this sort can be traced back to
poor requiremenents plannning.    The failure of Raytheon,
as described in the article, sounds like a failure to ask,
"What problem are we trying to solve and what is the
context in which we are trying to solve it?"    If one is
solving the wrong problem, no collection of development
tools is going to make a difference.

It seems the engineers from Silicon Valley took the time to
ask the fundamental questions.    They knew what problem
they wanted to solve.   Their choice of development tools,
once started, might have been better.   Windows CE may not
be the most reliable operating system but it does the job,
for now.   Let's see how it holds up in the field after it has
been on the job for a longer period of time.

My original point was that, Ada was never the culprit here. I
have seen developers blame Ada for their own incompetency
before.   When one large company mismanged its contract to
build an ATC and failed to meet its contractual goals, it found
it easy to blame Ada.  Pure silliness, but they needed some
excuse for their failure.   Meanwhile, other companies have
done excellent ATC software using Ada.

Of course, your point, not too subtly lying beneath the surface
of your warning (I did see the smiley) is that language would
not matter.   On this we disagree.    The more experience I
have with C++, and the more I study it, the more convinced
I am that it is more error-prone than Ada.  While I agree that
one can (underscore can), build reliable software with C++,
it is more difficult to ensure the dependability of the final
product.  We will have to agree to disagree on this point.

Language does matter once one has decided what problem is
to be solved.   However, if one has made poor choices in
the problem analysis stage, no  engineering expertise, programming
language, or development method will save them from creating
a dumb solution.

Richard Riehle







  reply	other threads:[~2002-02-12 16:53 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-02-11  1:11 Silicon Valley techies suit up Army with sleeker gear Ken Garlington
2002-02-11  1:55 ` Jim Rogers
2002-02-11  6:07   ` Richard Riehle
2002-02-11  8:09     ` Hyman Rosen
2002-02-12 16:53       ` Richard Riehle [this message]
2002-02-11 15:33     ` Marin David Condic
2002-02-11  6:09 ` David Starner
2002-02-11 13:27   ` Jim Rogers
2002-02-11 18:34     ` David Starner
2002-02-12  1:42     ` Warner Bruns
2002-02-12  2:32     ` Ken Garlington
2002-02-12  5:58       ` Jim Rogers
2002-02-12 12:07         ` Larry Kilgallen
2002-02-12 13:13           ` Jim Rogers
2002-02-12 18:21           ` Marin David Condic
2002-02-12 12:27         ` David Gillon
2002-02-13  5:04           ` tmoran
2002-02-11 14:09 ` Preben Randhol
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox