Mon, 21 Mar 2005 20:48:40 +1100
text/plain; charset=ISO-8859-1; format=flowed
>> I work supporting a DoD Agency enterprise standards program. We have a
>> request to evaluate the status of the Ada programming language. Could
>> you point me to information on any metrics that indicate the health and
>> status of Ada? For example, information on the revenue generated over
>> time for compilers and other Ada tools, the trend in the amount of Ada
>> code being delivered over time, etc.
Sorry, I'm afraid not.
I'm too busy working on a model compiler for X/T UML into SPARK (a
formally provable Ada-83 subset).
Anecdotally, Ada has been confined to reliable safety-critical systems
only, and mainly civilian ones at that. Problem domains where a failure
leads to a costly lawsuit rather than increased profits from "maintenance".
Defence manufacturers find it far more profitable to use other languages
(even if they cost more to do development in), as they make Big Bucks on
the maintenance and debugging after delivery. It's far easier to conceal
the inevitable errors when using other languages until quite late in the
lifecycle. They also get something sorta-kinda-working much sooner,
which looks good on the progress reports.
This built-in-quick'n'dirty unreliability is obviously a Bad Thing with
commercial aircraft avionics, but a huge cash cow otherwise, as the
customer has no choice but to pay through the nose after delivery to
correct fatal flaws. The unintended de-facto perceived abandonment by
the DoD of Ada has really impacted their bottom-line.
Alan & Carmel Brain
mailto:[log in to unmask]