Re: C++ vs. Ada -- Is Ada loosing?

[seas.gwu.edu!mfeldman@uunet.uu.net (Michael Feldman)]

In article <16269@goanna.cs.rmit.oz.au> ok@goanna.cs.rmit.oz.au (Richard A. O'K
eefe) writes:
>
>My current impression is that Ada textbooks tend to be more accurate in
>the claims they make about what is or is not valid Ada, and tend to have
>a higher level view of the software process, than C books.  Is this an
>illusion, due to my knowing C relatively better than I know Ada?
>
This is not an illusion. Ada books were - in the bad old days - full of
errors because nobody had good compilers around to test their code with.
Now that everyone has correct compilers (as least as far as validation
can show them correct), we can test our stuff.

My data structures book (1985) had lots of errors in the code, but these
were either _algorithm_ errors or "version control" problems where the
pieces did not fit together correctly because of author and publisher
carelessness. None were "Ada errors" or nonportabilities.

My freshman book (1991) had a lot of typos when it first came out, but
these were in the _text_ - misspellings and inconsistencies in the
English, tables, blah blah. Lots of UW students got to point out these
errors during my year in Seattle; I paid $1.00 for the first report of
each error.

There were a few algorithm errors and other quibbles that had nothing to do
with Ada. And ONE nonportability: there are two programs in which I made
a bad assumption about float-to-integer conversion, so the programs will
give different results depending on whether the float is truncated or rounded
where the fractional part is in the middle of the interval. This is 2 
programs of 180, and the error was my own stupidity.

I tested (nearly) everything under at least 4 compilers, and other
adopters of the book are using any number of other compilers that I
don't have access to. I have yet to get a report of a portability bug
in any one of them; indeed, I am not getting bug reports at all.
The book includes a diskette of all the programs, so nobody ever has to
endure a keying error to use them.

The purpose of this is not to hype the book, but just to point out
how easy it is to develop Ada code that'll really be compiler-independent,
given today's compilers and programs that don't use a lot of Chapter 13
stuff (you wouldn't find that in a freshman book!).

You might quibble with an author's coding style, or choice of algorithm,
but these days you'll have to work very hard to find a language error
in today's Ada books.

Mike Feldman

PS - My $1.00 per error still stands.