TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy


Options: Use Classic View

Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Content-Transfer-Encoding: 7bit
Sender: "Team Ada: Ada Advocacy Issues (83 & 95)" <[log in to unmask]>
From: Dave Wood <[log in to unmask]>
Date: Thu, 5 Oct 2000 19:10:30 -0700
Content-Type: text/plain; charset="iso-8859-1"
MIME-Version: 1.0
Reply-To: Dave Wood <[log in to unmask]>
Parts/Attachments: text/plain (81 lines)
----- Original Message -----
From: <[log in to unmask]>
> Except for people like myself, who see standards slipping, and the quality
> software engineering fast eroding...

This presupposes that quality software engineering ever existed
in the first place.  You can't slide down into the muck if you
never emerged from it in the first place.  (This is the collective
"you", not you personally.)

Having spent 20-odd years in the business prior to my escape
(ah, but like any good junkie, I can't wean myself from this
mailing list!  Ada rules!) in just about every setting from academic
to military, from commercial avionics to consumer products,
from end-apps to compilers, there seems to me one constant:
software engineering on an organizational level is a pipe dream.
The best software engineering always seems to come at the
individual level.  Everyone on this list is, no doubt, a fine software
engineer, but I've never seen an organization that could
consistently take it beyond the Program level (at best).

Like the illegal drug trade (excuse the ongoing metaphor), you
can only effectively combat it at the demand-side.  If there is
demand, there will be supply, one way or another.  There is no
demand for top-quality software except in very unusual
circumstances, therefore there is no supply of it.  Most of the
consumer-level software you buy is junk.  Most?  Maybe ALL
is more accurate.  As long as the market keys on lowest cost
and fastest draw to the next release, there will not be quality
software engineering beyond the individual level (or if that individual
is a little higher up, maybe the project level).  Beyond that,
economics take over, driving who is promoted to higher-level
power positions and therefore making future organizational

Who knows, maybe the world is well served, on balance, by
the un-ending supply of junk software.  I remember when owning
my own compiler was an impossible dream - those things cost
thousands of dollars.  Now you can get 'em for free.  Ada
certainly didn't drive that - C did.  The fact that you can get
your hands on free or cheap Ada compilers today is a direct
result of the trends created out of the C world.  So if you want
to take a positive view, thanks to C, it's now possible to put
Ada in the hands of more and more people.  Apart from languages,
cheap junk software has made it possible to have a huge array
of really highly productive tools on every desktop.  MS Office
is a fabulously powerful set of tools (compare to what was
possible in the 70s or 80s!)  No matter what you think of the
specifics of the tools themselves or of the company that provides
them, the implicit productivity is just enormous.  That's played
no small part in the productivity revolution that the world
economy has enjoyed.  It's all true for other low cost tools as
well.  The Corel tools are pretty cheap and powerful.  Sure,
they're buggy and clunky, but you can still do a lot with them.
Outside of the consumer domain, there are endless examples
of buggy software in specialty apps or built into embedded
systems, that by virtue of low cost and quick TTM put great
power into the hands of many.

Ada was a great experiment in government-funded quality
software that just fell on its face and has never fully recovered.
It was an attempt to create a demand by fiat and to create
the required supply by virtue of corporate welfare.  Like just
about any other socio-economic engineering project, it was
doomed from the start, really.

The language is great and there are some really good tools out
there, but the general programming market just isn't going to
embrace it the way some people would like.  Ever.  That just
gives a better opportunity to the rest of you.  Maybe all you
Teamers ought to gather up in some neutral site, say Nebraska
or New Zealand, and start your own software house.  Whether
or not you use Ada, at least there's probably a pretty hefty
number of savvy developers who understand and will practice
sound engineering techniques.  If only you can do it in an
economically viable way, the world would be your oyster.

-- Dave Wood