[python-advocacy] "Python in the Enterprise" PowerPoint deck?

[Roy Smith]

> I'm starting work on a PowerPoint slide deck on why Python should be
> used at my company. Before I go hog wild on this, I would like to ask
> everyone what would be the best presentation that is already out there
> to start from or should I just start from scratch?

I'm not sure if you're talking about a startup deciding what language to
go with, or getting a mature organization to add Python to their toolset.

Having gone through several cycles of introducing Python to a mature
organization, my experience is that if you try to convince them to use it
you'll fail.  Everybody will have 47 reasons why it's a bad idea.  I don't
want to learn a new language.  Perl is just as good (or better).  We can't
use it because of the source code hiding problems.  It's not supported on
all our platforms (or, at least, not already installed).  We don't like
(or understand) the licensing.  Whitespace is stupid.  And so on.  As you
answer each one, another will pop up.  It's a losing proposition.

On the other hand, if you just go ahead and start using it for tools that
are useful but not on the critical path, you will have an easier time of
it.

For example, when we rolled out IPv6 support, we were having loads of
problems with our development, build, and test machines having messed up
IPv6 configs (not surprising, since this was all new stuff to our IT
guys).  So, I wrote something in Python which checked for the 10 or so
most common misconfiguration errors.  It did things like verifying that
you could look up the IPv6 address for a host, that this host had AAAA
records in the DNS for itself, that you could open an IPv6 socket and
connect to it, that various (OS-specific) patches were installed, and so
on.  Around the same time, we were fighting fires with machines having
bogus system clocks, so I threw in some trivial NTP client code (using an
Active State recipie) to do a sanity check on the system clock.

Nobody *had* to use it, so I didn't have to convince anybody to allow
Python to be used.  Eventually, people learned if they used the tool, they
would same themselves a lot of pain.  When it got to the point where
managers were asking, "Why didn't you use the config check tool?", it was
a done deal.