Suitability of Python for a Big Application?

[Martijn Faassen]

Olaf Appelt <tholap at compuserve.com> wrote:
> I'm new to Python, so take this with some salt.

> I like Python, but I don't think it's well suited for large projectes. Also
> it lacks in the speed department.
> Also, I believe that it's lack of type safety and access restrictions, which
> is a bonus for small projects beccomes a liability in large,
> multi-programmer projects.

This may be true, but is debatable, and..

[snip]

> A possible alternative might also be Smalltalk (again for flexibility,
> portability and speed), but again there's the problem with available
> programmer skills. If your team has to learn Smalltalk first, it's not
> realistic given the time frame.

While smalltalk implementations may be faster than Python, Smalltalk is
just as dynamically typed as Python, isn't it? And I didn't think it had
much of access restrictions as well. So unless you're recommending Smalltalk
for speed reasons, I don't see why it'd do better than Python in the large,
multi-programmer project domain? 

It's probably true there's more experience with Smalltalk in large multi
programmer projects. I vaguely recall anecdotal evidence that it can be
quite successful at this. If that's the case, your objections against
Python as regards to lack of type safety and large scale programming support
may be less strong, though.

To use Python in a large project you need programming (and design) discipline.
But such discipline is need with Java and C++ projects as well (*definitely*
with C++!); static type checking doesn't magically make large scale engineering
problems go away, though it can help, of course.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?