PEP scepticism

[Andrew Kuchling]

"Tim Peters" <tim.one at home.com> writes:
> It's too glib:  the canonical example is CPAN, and is there some reason CPAN
> is *not* "boring" in the Perl world but is in the Python world?  Doubt it.

Not at all; CPAN has had bits which were broken for a long time, such
as the scripts submission, which I think was broken for a few years.  
However, the primary catalog feature kept working, probably because:
        1) someone wrote it initially, and not much subsequent maintenance
           was needed.

        2) lots of people used it, so if it broke, the resulting screams
           ensured that someone kept it running.

I suspect that an ambitious person could jumpstart the catalog into
existence, even if they ignored the catalog SIG completely.  Write an
initial implementation that's reasonably good, seed it with data from
Parnassus or wherever, and push it hard for a few months; if enough
people begin adding their new modules, then it'll just bootstrap
itself into indispensability, much as Parnassus did.

> The real question is why someone else hasn't popped up to do these things.

That's another good one; I don't know.  Perhaps people think you have
to be a serious wizard to contribute, but that's really not true;
there's lots you can do without being a genius who rewrites ceval.c at
the drop of a hat.  People who wanted to gain experience on a free
software project would be much better starting off on Python, which
you can write docs or write Python code or hack C code, than by
initiating yet another small application that never gets finished.
I've wondered if a document on how to work with the Python dev team
would be a useful thing, and if it would encourage more people to help
out.

--amk