[Chicago] 2 talks

[Kumar McMillan]

On Thu, Nov 11, 2010 at 7:41 AM, Martin Maney <maney at two14.net> wrote:
> I've kept this somewhat contrarian rant sitting in a tab for a while,
> because on the one hand I like pretty much everything about "Python
> 3000", but on the other hand in the real world, it feels about the way
> Mackall describes it:
>
>  http://lwn.net/Articles/411251/
>
> (...actually, I think I started writing this with another page in mind,
> one which I do not seem to have on tap, which went on some about it.
> The part I am forced to agree with is that in theory, Python 3 is
> great, but in practice little of the 3rd party code I rely upon runs
> there.  Still.  The migration plan appears to have been built on good
> intentions and moonbeams, or managed to overlook the thick web of
> interdependencies that actually exist.  Or something.)

Yeah, it's a pretty bad chicken and egg problem -- no one wants to
port until all their dependencies have ported and so begins the spiral
of doom.  I actually have been looking at the state of the top level
packages that most modules depend on and most of them do support
Python 3 by now even if it's in some forked bitbucket repo or
wherever.

So...what can we do to help everyone migrate their module we all
depend on to Python 3??  That's what I'll talk about tonight; tox will
be a key player here.  Note that a lot of work and a lot of thought
has gone into 2to3.  I've used it to covert several modules and it
generally just works out of the box.  I think a lot of people don't
realize this.

-Kumar