Python Popularity: Questions and Comments

Justin Sheehy justin at iago.org
Thu Dec 27 20:43:01 CET 2001


"A. Keyton Weissinger" <keyton at weissinger.org> writes:

> OK. Here is the question I keep asking myself: If Python is as
> cool/easy/powerful/etc as I think it is (and scream from the rooftops at any
> opportunity), why do we not see it taking hold in a commercial sense?
>
> Yes, I know. Many of you will tell me all about either (1) that so-and-so
> installer (specifically RedHat) is all about Python; and that (2) XYZ
> company uses Python you think, but doesn't want to share their commercial
> advantage or some such.

I'm going to refer to one specific niche that does not answer this
entire question, but is very significant to it nonetheless.

A very large portion of software is never sold directly, and thus
doesn't ever get promoted in the fashion that I think you are looking for.

The software that I am talking about is developed for use internally
at a given company, either internally-developed or by contractors.
This describes a truly huge amount of software.  Companies generally
don't even think about advertising this sort of thing, as it serves
the internal (though possibly customer-facing) needs of that company.

There is no reason for most companies to even mention this type of
software externally unless they are contractors that produce such
software for other companies.

> And I will say, "Um yes. But you're missing my point."

Is your point that Python doesn't have a hype engine to match Java's?

If so, then I'll just agree quietly.  I clearly don't think this is as
big a deal as you do.

> Have you met the "average professional developer" lately? He/She is a Visual
> Basic "guru" of 3-5 (whopping) years experience. He calls open source "share
> ware" and states emphatically the following: "Why would I mess with THAT
> stuff? None of those open source cowboys code to 'Microsoft' standards."
>
> Does that make your blood boil? Mine too. But guess what? That guy will be
> there in a few more years. He will have been promoted because he got (an
> admittedly crappy) product out the door. And then he will spread VB across
> the world.
>
> Oh wait. That's already happened.

Sure.  There are also plenty of places out there that realize what a
disadvantage those companies deeply tied to VB (or whatever) have, and
use the technology that is best for the tool at hand.  Such places
exist, and a number of them thrive.

While the MS-driven world may well continue to be the giant for years
to come, there will continue to be a large market for those who are
able to get the job done as effectively or more so for not being so
tied to whatever the current vendor-driven hype or incentive may be.

> Do I have the know-how to dive into the Python source code and begin to
> address the real serious threading-in-real-world-situations, optimizations,
> and memory consumption issues? No. But I am willing to learn...

Good.  I'm glad to hear it.

However, while those items you name might be showstoppers in a few
cases, they are not locking Python out of serious commercial
development in general.  Things are not nearly as bleak as you think.

> I will always use Python where it makes sense to use. I just wish it made
> sense to use it in so many other places... And I think it can....someday.

> P.S. Please limit your flames to either tell me (1) this will never happen
> and here's why or (2) how I can help make it happen....

(No flames here.)

As an aside, I happen to currently be working (with a small group of
colleagues) on such a body of software as I described at the beginning
of this message.  I'm talking about a large system, involving a lot of
the buzzwords one might feel like throwing around: client/server
networking, encryption, serious performance/speed constraints,
provable (seriously) performance of certain central components,
interacting with a huge heterogenous network, components operating on
multiple very different operating systems, distrbuted database
management, various network protocols, large-volume data transfer,
close to real-time reporting and alerting, etc etc etc.

Well over 90% of the code being worked on for this project is in
Python.  When proposing my ideas and design, I was happy to be asked
what tools and languages were being used.  When I was asked "why
Python?" I was able to give a decent answer explaining why I had
chosen it for the bulk of the work.  That was the last time I heard
that question.

The only reason that I bring this up is to point out that Python is
already in use in very serious comercially-funded development.  You
might not hear as many advertisements about it, but it is there.

-Justin

 





More information about the Python-list mailing list