Re: Ada The Best Language?

["Marin David Condic" <marin.condic.auntie.spam@pacemicro.com>]

"codesavvy" <codesavvy@aol.com> wrote in message
news:5be89e2f.0107181237.4ab3594@posting.google.com...
>
> I didn't say that Ada had nothing to offer, just nothing that is
> substantially better than C++.

Well, having used both languages, I'd beg to differ. I think that Ada offers
a lot more than C++ in the way of safety and correctness. I think its syntax
and semantics are more regular, clearly defined and less subject to
erroneous use. I think Ada provides for multi-threaded applications in a
portable way. I think its better suited to realtime systems and systems
requiring high integrity and/or long life. I think that makes Ada
substantially better than C++ - depending on what your definition of
"Substantially" is.

There are any number of other advantages Ada has over C++ - if you're
interested, there are websites that will start turning up resources for you.

>
> You must have missed my other post where I stated that Ada 95 is an
> excellent language and may actually be better than C++.  However, I
> don't feel that developer productivity is significantly enhanced with
> Ada 95 as opposed to C++.  I have asked if anyone knew of a study
> where the conclusion was that Ada 95 increased productivity and so far
> I've only been chastised.  Perhaps if we can agree on what is meant by
> "significantly more productive" we could carry on a rational
> discussion.  We would have to agree on the context of productivity and
> how it is measured.  I offered up a strawman in another post.

I had a ten year long study of defects and productivity relating to Ada when
I was in a past life. (I was the stuckee for Metrics.) While the study was
not comparing Ada to C++, it was comparing it to other languages that we
used in developing realtime control systems. Over ten years of use with Ada
showed that we doubled productivity and reduced our error rates by a factor
of four. Would C++ have done the same? Nobody will ever know in this
instance because it was not done, but my guess is "No". Why? It doesn't
provide the kinds of things that Ada provides that we attributed our success
to. (Whenever you reduce errors, productivity goes up because you aren't
fixing bugs - you're developing new code. Hence, our error analysis was
telling us the kinds of things we *weren't* messing up because of Ada's
safety features and I see no similar safety features in C++ that would have
caught these things.)

There may be other studies. I'd start looking at the AdaPower website and
follow links to other things like the AdaIC, etc. Studies *do* exist
indicating productivity boosts. But I think more significance should be
given to error reduction (at least for some domains) because of its
reduction of liability as well as its contribution to productivity. (Think
of it this way: Suppose you could develop code equally as fast in Ada and
C++. Suppose your Ada product has fewer bugs when it makes it to the field.
The more buggy C++ product damages your company reputation, reduces sales,
maybe increases warranty costs (think embedded systems) etc. The Ada product
does the same thing and got to market at the same time, but now doesn't
dammage your reputation, creates happier customers, etc. Don't you win this
way?)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/