On Thu, Oct 29, 2009 at 6:00 PM, Mark Summerfield <[email protected]> wrote:

> Are you saying that you _plan_ to make PySide incompatible with PyQt in
> the future?

No no - Renato does not speak for the PySide project but only for
himself. I'm sorry if you took it that way. As for myself, I have no
place representing PySide either, as that's the task of the nominated
project maintainer, Marcelo Lira. I only can give my own opinions and
indications where the Nokia-led efforts will be directed, but
personally I have a strong interest to "disown" the project and make
the major API design decisions completely with public processes
(hence, PSEPs). So, there is no hidden agenda. :-)

This being said, I'll give my opinion on the actual matter. I think we
should take a pragmatic stance on the compatibility and not promise to
follow PyQt blindly and indefinitely but to make explicit decisions
for any new API features, whether they mimic PyQt or not. Of course,
the value of compatibility should not be be underestimated, and
deviations should be done only after grave consideration and for
extremely good reasons. And of course, we should definitely reserve
the right to add new features not present in PyQt. So, the
compatibility issue goes both ways.

Of course, we have planned to start the work on the fabled new API at
some point, and there we might (or likely will) break the API, but
that doesn't mean we would stop working on the compatible API at the
time.

To summarize my opinion: Although compatibility shouldn't be
undervalued, every API design decision should be a result of the open
PSEP process, be it to follow PyQt or to implement new features.

Cheers,

ma.
_______________________________________________
PySide mailing list
[email protected]
http://lists.openbossa.org/listinfo/pyside

Reply via email to