PythonLabs Team Moves to Digital Creations

[tim_one at email.msn.com]

[Tom <nospam at nospam.com>]
> Some of this sounds good, but some is a bit scary.

Indeed, not unlike life itself <0.9 wink>.

> In particular:
>
>   [Guido]
>   We will be spending part of our time on core Python development
>   (including Jython and Mailman) and part of our time on Python
>   infrastructure improvements that also benefit Zope.
>
> If Python itself is good, and is being actively developed, then
> other companies and individuals will develop 'infrastructure'.
> For example, it appears that WingIDE, Boa, and ActiveState's
> Visual Python IDE will all be ready in the near future - this
> is evidence of the momentum that Python currently has.

Infrastructure refers to basic services, like roads, sewers and power
lines in your home town.  Whiz-bang IDEs aren't the idea here; they're
at the other end of that scale.  Things like library feature set and
internal implementation tradeoffs are on target, and, yes, we will make
changes in those areas that benefit Digital Creations products (but we
did before too!  the lines here are grey).  When such changes become
part of the std Python distibution, they should benefit everyone (much
as repaving a public road may benefit everyone in your town, although
probably benefits you most directly if the road happens to be one you
travel daily -- DC will naturally be most interested in improving their
own section of town before others, and, yup, we've agreed to devote a
substantial part of our paid time helping them do exactly that).

> However, if work on the core of Python were to slow, or if it's
> direction was guided by the interests of one particularly product
> (ie. Zope), then this might discourage the development of all
> other 'infrastructure' projects, and would hurt the Python
> community.

Python's direction will still be set by Guido.  Rights to our Python
(etc -- Jython/Mailman) work will be assigned to the PSF (to Guido,
until the PSF exists), establishing a strong legal wall between Python
(etc) and DC's (or anyone else's) particular private interests.

DC isn't a charity, but neither is Beopen.com, and while CNRI *is* a
charity (according to the IRS), Python work there too was strongly
influenced by the availability of funds for specific projects.  So the
potential problems you fear have always existed; the question is
whether Guido can continue to balance competing forces in ways that
leave almost everyone "happy enough".  DC isn't erecting new barriers
here, and-- to the contrary --is enthusiastic about pursuing genuinely
helpful new ideas (like the PSF).

Anyone else with a special interest to advance is free to contribute
the Python changes they want to the PSF too, via SourceForge.  Guido
will continue to Pronounce on such changes based solely on his view of
what's best for Python.

Note that there have always been more requests for library and language
and doc and implementation changes than can possibly be accomplished!
The really dramatic thing that's changed this year is opening up the
development process by moving it to SourceForge and attracting about 30
developers:  Guido's direct team is much less vital than it used to be
in getting core Python work done (for proof, look at who contributed
the major Python 2.0 features).  Python's development process "in the
public interest" is much stronger today than ever before, and the move
to DC doesn't hurt it.  Again, to the contrary, DC explictly supports
good ideas to make it even stronger (like the PSF).

nothing-new-to-fear-and-less-old-ly y'rs  - tim


Sent via Deja.com http://www.deja.com/
Before you buy.