Re: How Ada is failing socially

[jls@netcom.COM (Jim Showalter)]

>     If Ada is so great, and the US government is spending so many billions
>on Ada software development, why is the marketplace for Ada tools and
>libraries non-existent?

Because nobody ever went broke underestimating the taste of the American
public? (P.T.Barnum)

Because a great number of people have a knee-jerk reaction to anything
the DoD (or Big Government) ever does, facts be damned?

Because the DoD has been rather idiotic in its approach to broadening
acceptance and use of the language by not spending some sliver of those
billions of dollars to subsidize purchases of Ada compilers and tools by
schools, endowing chairs of software engineering at universities, etc etc etc?

>     The bulk of their products are language compilers, language tools and
>language libraries. In the Summer '91 issue, I made the following rough count
>of products by major languages:
>			Ada          -       10
> 			Assembly     -       32
>			Basic        -       65
>			C            -      289
>			C++	     -      111
>			Clipper      -       38
>			Cobol        -       15
>			Fortran      -       53
>			Modula-2     -       11
>			Pascal       -       56

One should always be careful to normalize data, lest one mistake an apple
for an orange.

First of all, Ada is generally still used more on large
projects than small ones, since it is on larger projects that its inherent
support for large-scale software engineering really shines over other
languages that lack such features (any language supports conditional
tests, but how many support macro-scale decomposition and enforcement
of interfaces?)--to quote P.J.Plauger, convener and general secretary
of the ANSI C standards committee: "Above 100,000 lines, you probably
should be writing in Ada.". Give this large-project focus, it is
not tremendously surprising to me that PC tools for Ada are few and
far between. One might well ask the counter-example: how many tools
for C++ for large projects are available? Last year the answer was: 0.

Secondly, there is no indication in the above counts as to the QUALITY
of the tools available.

Third, perhaps some of those other languages NEED more tools than Ada.
Lint and its enhancements certainly comes to mind...

>For some reason, the software market does not think that there is much money
>to be made with Ada products, and probably for reasons that have nothing to
>do with the technical efficiency of the language.

Again, the programming-in-the-SMALL software market isn't investing much
in Ada products. Check out the programming-in-the-large market sometime.

>Companies entering new software development projects are going
>to lean towards those languages for which there are many tools and companies
>offering products, even if the language is lower in quality (i.e. I might not
>think C or C++ is perfect, but I do like the wide variety of tools available).

AH! I've been just WAITING for someone to make this point for months, since
I have a few thoughts on this. The issue comes down to short-term vs long-
term gain. In the short term, the greater availability of tools for, say,
C++ (and the seemingly easier migration path from C to C++) might well lead
to a decision to use C++. But what are the LONG-term consequences of making
that decision? If an organization is truly committed to making substantial
improvements in its software development process, starting out with the
implicit objective of doing so without spending any money or making any
waves is pretty short-sighted. I know of many organizations that have tried to
upgrade their software development process, and the ones that were the most
successful were the ones that treated succeeding at it as a STRATEGIC corporate
objective, worthy of time, money, attention from all levels of management, and
even the pain of the transition. The ones who accomplished the least significant
results were the ones who invested in tools and training only begrudgingly,
clung to old standards and practices, refused to allow key people to spend
the TIME it takes to learn a new paradigm, etc etc etc.
-- 
*** LIMITLESS SOFTWARE, Inc: Jim Showalter, jls@netcom.com, (408) 243-0630 ****
*Proven solutions to software problems. Consulting and training on all aspects*
*of software development. Management/process/methodology. Architecture/design/*
*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    *