Python for everything?

Luis M. Gonzalez luismgz at gmail.com
Fri Jul 1 09:04:39 EDT 2005


Let me express it with an exaple (by the way, sorry for my bad
english):
Suppose you are planning to build a house.
You have two choices:
- Option one: Buy several thousands bricks, doors, tiles, windows,
etc... put them all together according to the blueprints and build your
home.
- Option two: Get loads of sand, concrete, wood and raw materials, make
your own bricks, make each door and windows with your prefered kind of
wood, design and make your tiles, etc.. and after a long, long time,
create a unique, one of a kind house.

Option one would be like using Python for creating your programs.
It gives you all the built-in functions and high level data structures
with which you can create a solution for almost any kind of problem you
might want to solve. You don't have to spend time allocating memmory,
using pointers, creting your data structures from scratch, etc, etc..
because it's all there already available for you, and all these time
consuming, error-prone tasks are very well hadled by python itself.

Option two is like using C:
It gives you full control in the creating on your program, but 90% of
the times (or more) you might think "do I really need to go through all
this hassle?". Sure, it will be faster but, "do I really need this
extra speed?".

I think that with the current processors, speed and performance is less
an isue than it was a few years ago.
Most of the times, developing time is paramount. And even if you need
very good execution speed for a given task, you can still code an
extension in c or c++ and keep all the benefits of a high level
programming language such as python.

And if these arguments still can't convince you, be adviced that
there's a very ambicious and exciting project called "Pypy" which aims
to create a high performance python implementation, and it is being
developed very succesfully.
They say that the secret goal is being faster than c... and I'm not
sure but I think that there's a tentative hard line set to december of
2006. Check it out: http://codespeak.net/pypy/index.cgi?news

Cheers,
Luis




More information about the Python-list mailing list