Re: Software Engineering in Florida

[Ehud Lamm <mslamm@mscc.huji.ac.il>]

On Mon, 8 Nov 1999, Robert Dewar wrote:

|<We tout Ada as a software engineering language>
|
|I hope not, this is a useless idea, 

It is indeed dangerous. When I ask students questions that compare (let's
say) Ada95 and Java solutions to a creatin issue, they usually say "well
obviously Ada is better.... This is an Ada course, and the course is
titled 'Ada and Software Engineering'...."

I always tell them that (a) this is not a valid argument in anycase (we
might call this fallacy an "Ad Languagem" instead of "Ad Hominem"... (b)
Many times it is plain wrong and (c) that it is entirly possible, and
worthwhile to study SE issues in C, for example, beacuse the setting is
deifferent, methods are different, and some issues that ARE solved in Ada
must be creatively solved in C. 

An anecdote.(But please don't take this to seriously)
We had a question that asked "What methods can you use in C/Pascal to make
your code more flexible to changes?"
Most students gave the obvious "comments, constants, subroutines" answer.
But one gave me a list of C "tricks" like using "void *", now this may not
be very elegant and generics in Ada are nicer etc. - but it is indeed a
common approach in C. The student received a bonus plus a comment asking
him to make sure he knows the implications of this trick on his systems
reliability etc.

SE is not a language issue - even though languages are tools that can help
or interfere with the production of good software. There is no one-to-one
mapping, and all languages will have SE issues.

Ehud Lamm mslamm@mscc.huji.ac.il
http://purl.oclc.org/NET/ehudlamm <== My home on the web 
Check it out and subscribe to the E-List- for interesting essays and more!