Long Live Python!

[Peter Hansen]

Alan Green wrote:

> Until it reaches the mythical "mindshare critical mass", Python is
> going to have to take over the world just one project at a time. For
> those of us who want to work in Python in the typical IT shop, this
> means doing some advocacy.
> 
> To be an effective advocate of Python, you need to:
> 
> a) deeply understand what the people you are advocating to understand,
> b) have the respect of the people you are advocating to, and
> c) deeply understand why Python is better at the kinds of problem
> being solved, and be able to express it.

> I plan for the process to take 12 months before expecting to see
> significant results. And remember: it might not work, but then again,
> there is no fun in a guarantee.

Well said (and having done something like this I can tell you
it is quite true).

In my case, I had the great fortune to have a "perfect" pilot project,
which would have taken significant time in another language but
which in Python was completed inside of a month by a co-op student
who had never heard of the language before.  (One week to learn it
well enough to start in earnest, one week to develop the 
requirements in enough detail to begin, one week to write the bulk
of the code, one week to test and polish prior to the demonstration.)

I strongly recommend picking an appropriate pilot project and
using this to help with the three items above.  Watch your
audience closely during the demonstration (or whatever equivalent
happens to be handy), carefully manage scope creep to prevent
the pilot project from becoming an extended prototyping nightmare
(thus ensuring completion on time and gaining the respect of your
"advocatees"), and watch carefully to learn just where Python is
extremely effective and where it is just better than average,
and where it doesn't quite perform as well as some other option.

Almost everything I know about Python in terms of its suitability
for different tasks and its problem areas I learned during this
short project (over a year ago).  This is not a language which
hides its inner nature from view.

It helps to have management which is either so preoccupied with
other problems or so forward-looking and courageous that they are
willing not only to go ahead with the pilot project, but willing
to accept the results and take a path away from the traditional
approaches.  (I have to admit I had more the former kind than 
the latter...  :)

-- 
----------------------
Peter Hansen, P.Eng.
peter at engcorp.com