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
Sender: "Team Ada: Ada Advocacy Issues (83 & 95)" <[log in to unmask]>
Date: Thu, 8 Oct 1998 12:23:51 -0700
Parts/Attachments: text/plain (131 lines)
> From [log in to unmask] Thu Oct  8 12:04 PDT 1998
> [said Ron]
> >
> > On a more serious note.
> >
> > Some of the bad points that I've heard about Ada have been:
> "you've heard..."?
> I think you are doing us all a disservice to broadcast "stuff you've
> heard" that is, at worst, false, and at best, quite out of date.

You hear why somebody or company won't use Ada just as you hear why
somebody else or some other company does use Ada.  My experience has
always been on established projects or projects where those decisions
are made at a (much) higher level.

> >
> > 1)  The time it takes to compile.  Granted, the Ada compilers
> >     generally take longer to compile programs than C/C++
> >     compilers (for the same amount of code) but they also do
> >     alot more work (there are more compile time errors in Ada
> >     than C/C++ and few, in my experience, runtime errors.)
> This might have been true 10 years ago.
> With today's compilers, on today's fast platforms, differences in
> compilation time are negligible. Have you compiled anything with
> GNAT recently?

I don't dispute that the technology has improved.  These arguments
seem to have a life of their own and, regardless of their validity,
do comprise some of the reasons used against Ada.

> >
> > 2)  The language dosen't "support" (take your pick)  as part of
> >     the language.  These are normally binary, pointer, or memory
> >     operations that are machine specific and fall into the
> >     infamous 'Chapter 13'.  Much of this was originally supposed
> >     to be addressed in the Ada 9x specification.  I've been out
> >     of the Ada programming area for the last few years so I can't
> >     say if they have or haven't addressed these issues.
> But you work for Aonix. In my opinion, you should try to find out
> some facts before you write stuff like the above. It shows a
> certain ignorance about your own products that is not very becoming.
> It's tacky just to pass others' BS along to wider audiences.

Wrong division.  We don't use Ada at this location.

> >
> > 3)  The language is too large, or too verbose, etc.  This was always
> >     (IMHO) a fictional argument.  The language is so much like Pascal,
> >     so regular, and so well documented that what seems like it should
> >     work normally does work and if it works one way doing one thing it
> >     works the same way doing something else.
> C++ is less verbose?

No, I was listing one of the traditional arguments.  I think C++ is rather
obscure especially when compared to Ada.

> >
> > 4)  The cost of the compilers is too high is another complaint.  This
> >     gone away do to the efforts of some very nice people writing several
> >     freeware/public domain Ada compilers and environments.
> GNAT is NOT freeware, and NOT public domain. Again, I think it behooves
> you to take just a few minutes to understand just what a Free Software
> compiler is. In the Unix community, they refer to Linux as "open-source."
> I think that best describes GNAT as well. Words are important; please
> learn to use them properly. Let's work to reduce the BS content of
> this list and this industry.

Would you classify gnat as shareware?  I'm not aware of anyone asking for
money for using gnat and I thought it was under the GNU copyleft protection.

I also make a distinction between freeware, which anyone can use without the
author requesting a fee, and public domain, which places the ownership of the
source code in the public domain (though that may not be entirely accurate.)

> >
> > One of the major reasons that I see preventing the use of Ada by
> > companies is their desire to hire people off the street and put them
> > directly to work programming (note I don't use the term 'Software
> > Engineering' here...).  The people I ran into at IBM that had 15 to 20
> > years there had all been sent to company schools that took 6 to 9 months
> > (before they went to their assigned jobs.)  When I started I had 2 half
> > days of inprocessing and started working on their systems the first day.
> Yes, this is the great shame of the computer industry today.
> >
> > The companies I've worked for (or interviewed with since then) have all
> > wanted you to have the skills before starting to work for them and none
> > seemed willing to invest the time and money for training their people.
> > This is just my experience, I hope other people have had different
> > experiences.
> I agree. The way to get around this, in the Ada case, is to make sure
> lots of people know Ada. Bug your local colleges to teach more of it.
> Point them to
> for a quick education on why this might be good for them.
> This is a vicious circle: students and teachers ignore Ada because there
> are "no Ada jobs." Employers ignore Ada because "there are no qualified
> people." Both are half-true. There are plenty of people, and a reasonable
> number of jobs. Matching them up is apparently quite hard.
> >
> > Ron
> Michael Feldman
> ------------------------------------------------------------------------
> Michael B. Feldman -  chair, ACM SIGAda Education Working Group
> Professor, Dept. of Electrical Engineering and Computer Science
> The George Washington University -  Washington, DC 20052 USA
> [log in to unmask] - 202-994-5919 (voice) - 202-994-0227 (fax)
> ------------------------------------------------------------------------
> "Whaddya mean, the Air Traffic Control system won't run
>  without a Microsoft browser??!!!"
>    -- political cartoon, Washington Post, Jan. 3, 1998
> ------------------------------------------------------------------------
> Ada on WWW: or
> ------------------------------------------------------------------------