Re: Some Ada Books.......

[mfeldman@seas.gwu.edu (Michael Feldman)]

In article <964.282ba00a@vger.nsu.edu> g_harrison@vger.nsu.edu (George C. Harrison, Norfolk State University) writes:
>In article <3183@sparko.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:

>> Of the current CS1 books, I prefer Skansholm to Volper/Katz. 
>
>My problem (and my student's problem) with the books you mention that I have in
>my library (I don't have your D.S. text.) except for Cohen's text is that one
>wonders WHAT an Ada program really looks like.  Too many texts parallel the LRM
>or are so weak that they are either too difficult or too elementary even for a
>CS1 course.  

You didn't mention which books you're referring to, so I'll just give some
general comments for the net.

The Skansholm and Volper/Katz books are both targeted to CS1, and while I
prefer the former to the latter, both books are very conscious of pedagogy
and presentation order, do _not_ follow the LRM, and concentrate - in the
early chapters, at least, on CS1 topics. These authors all, in my opinion,
know how to teach to novices. Volper and Katz are both first-year teachers
in the Cal State system; Skansholm is a Swede.
>
>I supplement our own Ada Programming course with about 50 of my own
>programs/packages/etc. and do not provide my students with a general IO package
>to hide text_io.  This course parallels our CS2 course for about 2/3 of the
>semester - then we take off into generics and tasking.

That sounds about right. Whether to use an IO package that "hides" Text_IO
or bring it out in the open is surely a "religious issue"; in the end, the
students will have to learn Text_IO, so I don't think it's a big deal.

One of your problems is that your students already know some programming, so
a straight CS1 book may be too patronizing to them. 
>
>Most texts stress the mechanics of Ada or leave out a lot of the mechanics 
>in favor of the "gee wiz" of Ada.  I need a text that has both.

Well, actually I think you need one that concentrates on the techniques of
algorithm development and good design, independent of the programming 
language, then plugs in the right stuff from Ada wherever that makes sense.
(The success of the Koffman books over the years makes me pretty sure that
he's been doing something right; I hope that my overhaul of his material
for Ada will continue the trend - end of plug).

I find myself wondering why the course is "Ada programming" and not a
straight CS topic that happens to use Ada as its language. Isn't that
where we should be heading?

Mike