[Edu-sig] measuring python...

[Jeffrey Elkner]

Thanks a 10**6, Kirby!  I already planned to include Python's
commonality with other languages somewhere in the paper.  And I can talk
about it from personal experience.

I'm taking an AI course this semester and we are learning Lisp.  I was
struck by how many Lisp concepts I was already familiar with thanks to
my exposure to Python.  My classmates whose experiences were generally
limited to C, C++, and Java did not have this advantage.

I've printed out a few of the references and will begin looking at them
tomorrow.  The http://www.cise.ufl.edu/~jnw/COP5555/Lectures/02.html
site looks particularly helpful.

Thanks again!

jeff


On Sat, 2001-09-29 at 17:29, Kirby Urner wrote:
> 
> Fun quote:
> 
>    We often refer to Perl because it contains the extremes of
>    both good and bad, thus making for poignant examples. We
>    respect its power but also abhor some of its abusability.
>    It  can be called the "Bill Clinton" of programming languages
>    because it has some good ideas and is quite good at some
>    things, but it also has some serious integrity problems
> 
>    http://www.geocities.com/tablizer/langopts.htm
> 
> Relevant:
> 
>     Simplicity
> 
>     This is an attempt to keep the code simple. A common (but
>     imperfect) measurement is the number of "tokens" needed.
>     Tokens will be defined variables, objects, methods, keywords,
>     operators, specifiers, etc.
> 
>         http://www.geocities.com/tablizer/goals.htm
> 
> Note: this page part of a vast anti-OOP site, which is fun to
> explore if only because the guy is passionate about his
> anti-OOPiness.
> 
>         http://www.geocities.com/tablizer/oopbad.htm
> 
> (hahah:
> http://www.geocities.com/tablizer/xbasefan.htm
> I'm an Xbase fan too, although I use an OOPized version
> of it)
> 
> Also relevant:
> http://www.cise.ufl.edu/~jnw/COP5555/Lectures/02.html
> 
> I like the approach of this latter page, which doesn't make
> "readability" an atomic measure, but something higher level,
> with lots of component parts (code size might fit into it).
> 
> Readability seems related to the question "What makes a
> language a VHLL?"  Part of the answer:  built in data structures
> that already do a lot (e.g. Python's dictionary).
> 
> Final point:  one reason I'd give for why Python is a good
> first language and/or teaching language is that it has so
> much in common with many other languages.  In other words,
> I'd emphasize the similarities over the differences.
> 
> Python is very cosmopolitan and in this sense provides a
> broad spectrum of analogies, templates, experiences, which
> the student programmer can draw against, when learning
> whatever next language(s) (i.e. we should never assume
> that "first language" means "last language").
> 
> Kirby