Re: Anybody in US using ADA ?

[Jerry Petrey <"jdpetrey<NOSPAM>"@raytheon.com>]

Mike Silva wrote:

> Richard Riehle wrote:.
> > "John R. Strohm" wrote:
> >
> > > I *THINK* that Lockheed-Martin Fort Worth is using Ada for F-22 and F-16.
> > > You might try looking at their web sight (http://www.lmco.com) and seeing
> > > what they are looking to hire.  (Most of their effort appears to be trawling
> > > for resumes for  Joint Strike Fighter, which is in C++.
> >
> > This will eventually turn out to be one of the all time stupid software engineering
> > decisions, one that will end up with lots of people blaming each other for a
> > series of cost-overruns, buggy software, and probably a lot of regrets.
> > However,  it is a decision that needs to be experienced so the managers can learn just how
> > hideous C++ is for this kind of software.   One can only hope they do learn that
> > lesson and no one will get killed due to a software accident because of it.
> >
> > Richard Riehle
>
> I have a number of questions about this:
>
> 1) How is such a decision (choosing the language for a major project)
> made, especially when the language chosen has not been proven in such
> applications?
>
> 2) Is there any sign yet of cost-overruns, buggy software, or regrets?
>
> 3) How does C++ compare, safety-wise and reliability-wise, with
> earlier languages used in such applications, e.g. Jovial?  (I *know*
> how it compares with Ada!)  To put it another way, how big a leap
> backwards is the use of C++ for these applications?
>
> Mike

Having worked for Lockheed and a number of other defense companies in the past, I can comment on
question 1.
Usually the managers who make the decisions on such things as languages, tools, processors, etc.
don't have a clue what is the best choice.   They are not particularly interested in things like
reliability, maintainability, safety, or quality; rather, what they think will be fast and cheap
and make them look good in the short term (the old �You guys start coding, while I go upstairs and
see want they want us to build�).  They tend to listen to the engineers they perceive as being
knowledgeable (in many cases these are relatively young engineers who have done a good job of
kissing up and winning the favor of the managers).  These engineers, in turn, are thinking of what
will be the best for their career and what they are most familiar with.  These are the guys we have
to convince that Ada is the right choice, but it is very difficult to do since they are often
ignorant of Ada and firmly believe that due to the popularity of C++, it is the right choice for
new applications.  Many previous posts have suggested ideas for improving Ada�s acceptance -
cheaper and better compilers and tools, bindings, etc.  These are all good, but most of all we need
to get Ada more into the education system and promote the successes of Ada applications that are
done.   And, of course, do more Ada applications where we have the control to do so � at work or at
home.  Having Ada available on more commonly used platforms
would certainly be a big step as well.

Jerry
--
-------------------------------------------------------------------------------------
-- Jerry Petrey
-- Senior Principal Systems Engineer - Navigation (GPS/INS), Guidance, & Control
-- Raytheon Missile Systems             - Member Team Ada & Team Forth
-- NOTE: please remove <NOSPAM> in email address to reply
--------------------------------------------------------------------------------------