Python Popularity: Questions and Comments

[Alex Martelli]

"RPM1" <rpm1deletethis at frontiernet.net> wrote in message
news:u2nk47nvtu000c at corp.supernews.com...
    ...
> "full technical support"
>
> That's what gets thrown in my face whenever I mention Python.
> (And I mention it a lot).  The company I work for writes point
> of sale software in C and VB, because "there's support" and

VB is a commercial product.  C, like Python, is just a language,
of which exist implementations, both commercial and otherwise.

You can get many different implementations of C, and a few of
Python (the latter are typically "distributions" of the standard
implementation with packaging and support tweaks).  For several
of those implementations, you can buy technical support, "full"
(whatever THAT means) or otherwise -- this goes both for C and
Python, of course.

Firms like ActiveState (www.activestate.com) will typically be
overjoyed to sell you "full technical support" for Python on
the platforms they support.  See, for example:
http://www.activestate.com/Products/Enterprise_Solutions/Python.plex
for quite a few commercial offerings by ActiveState of Python
*Support* products/services, specifically.  I'm sure many other
purveyors of such products and services exist, and I hope relevant
URL's will be posted on this thread.

When you can get somebody to put this sort of thing in writing,
or say it in front of enough witnesses that they can't later
easily backtrack -- so that they've basically admitted that
if "there's support" then they HAVE to look at Python in depth --
you may then follow up by contacting the *SALES* organizations
of such suppliers as ActiveState.  Forget marketing: you need
hungry, aggressive *SALESPEOPLE* to ram change down the throat
of some organization that's clutching at straws to resist change.

Marketing is about product positioning on the market (features,
release timing, pricing models, etc), information flow from the
prospective-audience to the designers and vice versa, and, to a
VERY modest extent, general "public relations" and persuasion.
Don't confuse it with *SALES*, as we techies so often do -- that's
quite as bad as confusing very different tech roles, such as
software development and system administration.


> "customers won't pay for something made with freeware."

I wear Mephisto shoes, but I'm not going to claim that "customers
won't pay for software developed by somebody wearing Nike shoes".

As a non-sequitur, though, it's roughly on a par with this quote.

How will the customers know, and why should they care, what
brand of shoes (if any) the developers choose to wear?

How will the customers know, and why should they care, if (e.g)
your C++ sources were finally compiled/linked with free products
such as gcc, or for-pay ones such as commercial compilers?  The
same, obviously, goes for Python -- how will the customers know,
and why should they care, what kind of support contract YOU have
with ActiveState or other suppliers, if any?


> I think if Python wants to get "bigger" or more acceptable,
> it needs marketing.  The guy with the bag of money needs
> to say, "Oh yeah, Python, I've heard of that."  Customers

I'm not sure "Python" itself "wants" anything, just as I
doubt, say, "C", or "Java", have desires, feelings, hopes.

> ask us if we have a Java 'solution', they don't ask if we
> have a Python 'solution'.  That's got to change if Python
> 'wants' to grow more, (I don't know that it does).

Neither do I.  More to the point, this looks like a commercial
issue.  Is there money to be made by this sort of promotion,
which in turn does surely cost money?  If so, then it's more
of an issue of entrepreneurship -- get capital, start company,
invest, draw profits.  For typical Python *users*, well, if
their competition is using less-effective tools, why shouldn't
they just enjoy the resulting competitive advantage?  Production
costs aren't everything, of course, but in most endeavours they
do matter a lot to competitive positioning in the market.  In
many cases one goes to a LOT of trouble to try to have slightly
lower production costs than one's competition.  If in this case
my competition is willingly going to its own slaughter, should I
spend time and energy to reverse that?  As Clough put it, "Thou
shalt not kill, but needst not strive, officiously, to keep
alive".

Yeah, yeah, I know, this is not the prevailing ethos in Open
Source and more generally in the computer world -- we're all
missionaries out to save the world with our zeal and diligence,
particularly by ramming the "right" technologies down the
unwilling throats of everybody else.  For their own good, of
course.  Yeah, well.


>     It seems to me, (with my limited experience), that the
> most dedicated audience Python has is the scientific
> community.  I would start there.  Get some big chemical,
> pharmaceutical, or biological corporations to use Python,
> (by catering to their needs), and then they will in turn

Done, see http://www.python.org/psa/Users.html.  AlpgaGene,
Inc, NorthSide Physical Therapy, Thermo BioStar, Biosoft,
Caltech, are among the firms in this general sector who
not only use Python but are willing to be identified as
doing so on that page.

> support the language when they see the wonderful end
> product.  Then people like me can say to my boss, "look
> Dow, Johnson & Johnson, and ADM all use Python
> heavily, so there's nothing to be afraid of."

Why should such firms care about letting it be known
whether they use Python heavily, or Java, or C++, or
whatever else?  That's not their core business.  What's
in it for them?  Can you even FIND OUT what programming
languages are in fact used by the various departments
and fiefs of these large organizations?


> Maybe I'm wrong.  But it seems to me that Microsoft
> is ahead, not because of the high quality of their
> product, but because of there timing and marketing.

I think Microsoft's marketing has typically been truly
abysmal in most respects -- quite a match for the median
quality of most of their wares.  There are no doubt some
exceptions on both scores (e.g., the COM architecture,
and the .NET Framework, have some technically excellent
traits; marketing in both cases has done a horrid hash
of a job, particularly NOT managing to convey public
information about the technologies in question, and to
some extent, at least in COM's case, definitely NOT
steering product positioning according to market needs).

We'll see how their latest game-box offering fares, but
so far every one of their attempts beyond the desktop
has been substantially a failure, cross-subsidized by
profits on desktop operating systems, office apps, and
to a lesser extent other desktop apps (including games,
software development tools, etc).

This tells us nothing about marketing _technologies_
as opposed to products and services.  In that field,
Sun's very successful efforts at marketing Java may
be more instructive.  But, who has a billion dollars
available to throw at the task?  And how to they plan
to recoup the investment?  Sun has presumably judged
that opposing Microsoft's dominance has huge strategic
value for them: they don't make money directly out of
Java (not, by far, enough to make the huge marketing
investments profitable), but apparently they think
they're getting value for money through other revenue
channels.  What huge firm[s] might possibly make a
similar decision in the case of Python?  And why?


Alex