Making a better textbook (was Re: The Deitel book)

[Dave Brueck]

On Wed, 6 Nov 2002, Dave Reed wrote:

> I've been using Python for my own projects (from short scripts to
> 25,000 line apps) for almost three years now and can't imagine using
> anything else right for anything that wasn't extremely computationally
> expensive. After attending the panel, I talked my other colleagues
> into using it for our CS 1 course this fall. The only thing that
> bothers me about Python for teaching is the lack of enforced private
> members for classes. As an experienced programmer, I can live with
> that because I know better, but I don't know whether the students will
> believe me when I tell them it's a bad idea to directly access them
> outside of classes :-)

Who cares if they believe you though, and why should they anyway? Part of
becoming a good programmer is getting bit by taking shortcuts (over doing
it the "right" way), learning from it, and moving on - and it's a great
thing to learn through experience that the prof might actually know what 
he's talking about! :)

> I also suspect this issue may prevent other universities from seriously
> considering Python for introductory courses.

Yeah, but in reality it shouldn't. Up-and-coming programmers routinely
ignore what we consider to be even the most basic best practices. I mean,
what first year (or any year, for that matter!) languages force people to
use good variable names, limit their use of global variables, avoid magic
numbers, and stay away from creating 1000-line functions?

Anyway, it was exciting for me to read your observations so far from using
Python at the university level, especially the idea that more
problem-solving time is being spent in "algorithm space" than "syntax
space" (which shouldn't surprise me since that's the experience Python
users tend to have). I'd love to hear more of what you learn as time goes
on - please keep posting!

-Dave