Python vs .Net

[Harry George]

Laura Creighton <lac at strakt.com> writes:

> > "Karsten W. Rohrbach" wrote:
> > > 
> > > Python is an interpreted, object-oriented, weak-typed, [...] programming
> > 
> > As Paul said, you misspelled "strong", but also forgot "and dynamically typed
> > ".
> > 
> > (See frequent and by now almost tedious past discussions in this group
> > as to why Python is most definitely not a "weakly typed" language.)
> > 
> > -Peter
> > -- 
> > http://mail.python.org/mailman/listinfo/python-list
> 
> If people around here would remember that the reason we have such
> discussions all too often is not that some people are _wrong_,
> and believe that Python is what you call a _weakly typed language_,
> but rather in their linguistic community 'strongly typed' ==
> 'statically typed' and 'weakly typed' == 'dynamically typed' things
> would be a lot more pleasant.  The confusion is inevitable, because
> there are one heck of a lot of computer science text books out there
> which teach precisely this.  

I'm not aware of these texts.  The technical (as in graduate computer
science) language and language design texts I've seen understand and
describe strong/weak, static/dynamic quite well.  It is the popular
press which mixes them up.  But I do agree it is worth pointing to a
FAQ whenever the issue arises.

> And, no, it doesn't mean that people
> like Karsten W. Rohrbach don't understand the difference between
> what Python does, and what Perl does -- but when they want to
> discuss this difference, they speak of whether the language does
> or does not support 'automatic promotion of types' -- which is
> completely independent of whether you have to declare them or not,
> and whether you can rebind them on the fly.
> 

My impression is that it is the silent rebinding which throws people
off, and gives the impression of weak typing.  Typos give this
effect.  In Prolog that is a std bug generator, and I've seen in in my
python code as well.

> Laura Creighton