On 12/06/2011 04:24 PM, Jonathan Protzenko wrote:
I think the main point of the discussion is to improve "the community".
If we really want to improve OCaml as a whole, then I think we can put
our efforts on better areas than patching the compiler.
I completely disagree with you (and this is rare enough!). The
discussion that Benedikt started is not about improving the community,
it is about improving the core system. Both are important issues, but
discussing them together will not help make any progress.
If someone wakes up a morning with a strong desire to help OCaml and
wonders how to contribute in the most efficient way, then it could make
sense to compare the relative benefits of improving the compiler vs. the
community vs. writing a book vs. etc.
But the situation is not like that: people work on different topics
according to what they enjoy or need. It happens that Benedikt has been
working on issues which he (and others) considers important enough to
deserve some attention. The question is now how to turn this work into
something useful for OCaml, and also how to avoid creating some
frustration amongst people who are willing to contribute on the core system.
That said, I'd argue to avoid creating a "community" fork.
It should be noted that while INRIA maintains its primary role in the
development of OCaml, the "core development team" is not entirely made
of people from INRIA. Historically, Jacques Garrigue has contributed a
lot to OCaml without being affiliated with INRIA. More recently,
several people outside INRIA have gained and used direct commit rights
to the OCaml SVN repository. Experiments and proposals are typically
carried on branches (which are publicly available), discussed amongst
developers, and then sometimes integrated in the trunk. This works
quite well, and I'm happy that Xavier continues exercising his
leadership to decide ultimately what goes into OCaml and what doesn't.
The problem is not so much the current process than the size of the
"core development team" and the fact that many of its members can only
contribute very little of their time to OCaml. I'd thus argue to
enlarge this group with more people who have demonstrated their skills
for working on the core system.
Alain
--
Caml-list mailing list. Subscription management and archives:
https://sympa-roc.inria.fr/wws/info/caml-list
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
Bug reports: http://caml.inria.fr/bin/caml-bugs
