At 03:44 PM 2/13/01 -0500, Michael Feldman wrote:

. . .

> > Yes, the validation suite.  This is a VERY key point and one that I should
> > raise more often.  But it just seems so natural to me that there SHOULD be
> > a validation suite, for EVERY language.  So, I tend to not mention it,
> > assuming others have the same view.
>
>Really?

Yes.  Rather naive, I guess.

> >
> > Why is our industry so far behind in their view of something so obviously
> > beneficial as a compiler validation process?  This confounds me.
>
>Well, we saw Phil Breashear's note about this - no C++ compiler could
>pass the validation suite.

Yes, a fascinating confirmation of much of what I say in my sucky software
lecture.

>Only in Ada do we care seriously about this, and I conjecture that the
>culture of caring about it derives from DoD caring so much about it in
>the 80s. Now that the culture is entrenched, it's (luckily) hard to
>shake it.
>
>Two more ancient-history anecdotes:

. . . etc.

You make a good point about the Ada world being the way it is "because of
the DoD".  However,  I consider the notion of "validation", perhaps in even
a stronger form than we know it for Ada, to be a natural, necessary
consequence of sound Software Engineering principles.  I also note that
virtually every other society-critical industry has, industry-wide, at
LEAST as strong a notion of "validation" (translated into whatever makes
sense in their domain).  This is closely related to, but distinctly
different from, the licensing issue.  Society (unconsciously) demands that
software, across the board, be validate before brought to market.  The day
will come when this "demand" will become VERY conscious, and heads will
start to roll.  (I, personally, would hope that software types will be in
the driver's seat of whatever mechanisms are put into place to identify
which heads.  :)

It is an observable fact that software developers, in general, are just
professionally irresponsible when it comes to things like "subjecting their
products to someone else's validation suite".  Although I don't pretend to
fully understand this phenomena, I do understand quite a bit of it.  There
are many "sources" of this attitude, of course.  But a lot of the
underlying "psyche" here is precisely the same one by which so many
software types become completely blind to the inadequacies of C-class
languages, and are more concerned to "be a geek", "do their own thing",
etc., than to be able to do good Math, Science, or Engineering.

This circumstance is entirely intolerable.

Anyway, enough of the soap box . . . for now.  :)

sro

S. Ron Oliver, semi-retired professor of Computer Science and Computer
Engineering.  www.csc.calpoly.edu/~sroliver

caress Corporation is proud to be the U.S. representative for Top Graph'X,
developers of high quality software components, using Ada.  For more
information, check out www.topgraphx.com.

Tired of sucky software! ?  Check out www.caressCorp.com and follow the
links to software sucks and The Oliver Academy.