Re: Not really unsigned and not trash.

[cjl@iuvax.UUCP]

>    First, I am not inexperienced in Ada.  As a member of the
> team which has developed one of the largest running projects yet
> written in Ada, a product which has broken both the DEC and Data
> General compilers over and over, I probably have a better feeling for
> the language than most.
> 
> last thirty years.  I teach a course in "Comparative Programming
> Languages" here and allocate portions of the semester trying to
> sincerely show the Ada's features and then let them draw their
> own conclusions.  While their conclusions are the conclusions of
> innocents, they usually feel it is a verbose hodgepodge formed by
> a committee who wanted it all but didn't know what all was available.

  I am currently teaching "Programming Languages" in my school too.
I shared Robert's experience that slow compiler tends to degrade the
software coding. However I don't quite understand why Robert said
that "Ada is a verbose hodgepodge formed by a committee who wanted
it all but didn't know what all was available".  I hope this topic 
can be furtherly explained.

  According to my experience of teaching, the feeling about Ada is
quite different. When students learned Ada, it is just too difficult 
for them to go back to Pascal,C etc. While PL/I is generally
criticized for being a monster created by a committee, Ada is quite
different. Every feature is aimed to support certain software engineering
principle. Without learning software engineering, it is too difficult
to understand Ada. So I just dropped the traditional approach to
teach programming languages by language comparism. Instead, the
class is centering around the topic of software engineering. The
other languages ( such as Modula-2 and C ) is only mentioned and compared 
when they support different approaches to solve the same problem.
So at the end of class, students learned more about software engineering
than the language comparism. But the tradeoff is paid off especially
when students start to feel uncomfortable with other languages
because they learned to design programs in a more structured and 
modularized manner but they find weak support in other languages.
And I think that is a better approach for teaching P.L. in a limited 
amount of time.  (The book I chose is Habermann's "Ada for experienced 
programmers" for senior students with Pascal as their mother tongues.)

  Ada is big. Big and slow compiler has certain effect on the program 
development. If present compilers remain slow or some features are
proved to be too expensive, the definition of Ada Jr. may be
necessary. So it would be beneficial if Robert can tell us why
he thinks  Ada is wrong IN DETAILS according to his experience.
In addition detail and concrete discussion may be less offensive than
quick and general conclusion.

C.J.Lo
Dept. of CIS, IUPUI
UUCP : ...!iuvax!cjl
ARPA : cjl@Indiana@CSNet-Relay