does anybody earn a living programming in python?

[Magnus Lycka]

walterbyrd wrote:
> If so, I doubt there are many. 

Depends on what you compare with. I'm pretty sure there are thousands
of people working as Python programmers, and many more using it as a
smaller tool in their work. Of course this is small compared to Java
or C++.

In the US, it seems a lot of companies in the animation industry, such
as Dreamworks and Industrial Light & Magic regularly look for Python
skilled people.

In my vicinity, Göteborg with surroundings in Sweden, I know of
several places where people work with Python. I'm sure the amount of
C++,  Java or VB programmers are magnitudes bigger, but there are also
a lot more people eager to fill those positions.

My employer, Jeppesen Systems AB, has dozens of Python programmers,
and we train our customers to use Python for customization and 
integration etc. So, we know that a significant number of the world's
largest air lines and some significant railroad companies have trained
Python developers (we trained them) and use Python for customizing our
products, and for e.g. integration with legacy systems.

Just a few hundred meters from us, there is AB Strakt, which was more
or less a pure Python company last time I looked. The pharmaceuticals
company Astra Zeneca uses Python in some of their research, and I've
seen people from other companies in the vicinity asking Python questions
on various mailing lists etc. Python is certainly used in Chalmers, the
technical University here. This is in a region with way less than one
million inhabitants. I'm pretty sure we have >1 Python programmer per
10 000 inhabitants. Let's say that's above Swedish average. I still
think there is at least 300 Python programmers among the 9 million
people here. Although we're just 0.15% of the worlds populations, my
experience is that we make and use something like 1% of the high tech
stuff in the world, so my standard extrapolation technique would yield
30000 python programmers globally.

> I wonder why that is?

Well, it takes time to change things, and many decision makers aren't
very well informed. Maybe they are more likely to choose something
which is marketed and supported by some large commercial entity. The
rise of Linux and open source is changing that, but it takes time.
If you look at languages introduced the last 15 years or so, I think
Python is one of the more popular. The languages I can think of right
now, that have appeared since 1990 or so, and received a significant
mindshare are C#, Java, JavaScript, Perl, PHP, Python, Ruby and Visual
Basic.

Of these, C#, Java and Visual Basic have serious commercial backing
which promoted them heavily for certain applications, and provided a
lot of convenient tools. Those languages are also closely modeled
syntactically on a predecessor, C++ for Java, Java for C#, and VB is
just a new BASIC dialect, which makes the transition to start using
them easier, at the expense of retaining stupid aspect of old languages
that we'd rather get rid of. I.e. getting started seems more important
than getting it right. That should help in the short run, but be less
good in the long run.

Perl and CGI was the best available tool when the big web boost
happened, and PHP has taken over that role. They are very popular in
a particular niche (but Perl seems to be in decline) but like
JavaScript, they are niche tools.

Ruby is the language most like Python I guess, and it's nowhere near
Python in popularity, even if RoR has given it a boost. As people have
noticed, Python is often used together with C or C++, and it seems to
me that Java programmer might prefer Ruby to Python. This might be
because Java and Ruby are both purely OO, and might fit people who think
OO is everything, while Python and C++ are multi paradigm programmers
that share a different mindset.

So, to summarize, Python is probably the most popular general purpose
programming language invented since 1990 which didn't have multi
billion dollar backing. Now, it has more backing, with Google investing
in it, for instance by hiring the creator and other important Python
developers. Python 2.5 has advanced it even further, and it seems to
me that Python 3.0 will be a much smoother ride than Perl 6 turned out
to be for the Perl community. IronPython and PyPy are also very
interesting projects.

Python is a generic programming language, so I don't think we need some
kind of killer application like RoR. On the other hand, I think a really
smooth IDE with a convenient GUI builder could help making it more
widely spread "among the masses".