In message <[EMAIL PROTECTED]>, Justin Erenkrantz writes:
>Okay, let's do this: If you're on this list ([EMAIL PROTECTED]) and you are 
>already a committer to some other ASF commons project and would like to help 
>the ASF Commons by lending your insight and help us to manage this endeavor, 
>I'd like to solicit your self-nomination for Commons PMC.  Your stated 
>interest in joining the PMC and prior experience elsewhere within the ASF 
>would be suitable justification for me to nominate you to the rest of the PMC.

I mulled this over for a bit, trying to decide if I had the time to
participate, and I've decided I do.  The reason I'm interested is because
both code libraries I've donated to Apache (technically, the user/developer
community donated one of them) would probably have been a better fit
in an Apache Commons (had it existed at the time) than Jakarta, despite
the fact that Jakarta projects use the code.  The reason they wound up
in Jakarta was likely because Jakarta had to some extent become the
de facto umbrella for non-XML Java projects.

Now that Apache has expanded the scope of projects it hosts, it's possible
to narrow the scope of each of the projects.  Whereas before we would have
debates about whether a build system or a regular expression/text processing
library was in scope for Jakarta, now the build system can become its own
top-level project and the misfit libraries, at their communities' options,
can move into Apache Commons.  In the past I was opposed to rejecting
useful subprojects just because they weren't neatly in scope, but now that
there's an appropriate place within the ASF for just about every project
(e.g., ultimately subprojects can become top-level projects), I don't
think it's a bad thing to try to place subprojects under more appropriate
top-level projects.  And yes, I understand that the big question is how to
decide what is appropriate.  I tend to disfavor purely language-based
categorization.  However, the Apache Commons PMC only needs to establish
what's appropriate for Apache Commons.  From the APR example, there's
at least the implication that if a reusable library becomes large enough
by some definition (a large enough community, enough subcomponents, ...)
that it would be better served by promotion to top-level project status.

In any case, with regard to "prior experience elsewhere within the ASF,"
to the extent that I'm responsible for the original code, I've donated
Jakarta ORO and Jakarta Commons Net to the ASF (although both subprojects
have since seen contributions from many developers; Commons Net more so
than ORO).  I served on the Jakarta PMC a couple of years ago and am doing
so again.  Recently, my main activities have consisted of code maintenance
(reviewing and applying patches, responding to bug reports, etc.),
encouraging code contributions to expand the developer community of the
projects I'm invovled with (Commons Net has been successful in this regard),
and accidentally stirring up discussions about how to handle "mature"
projects on [EMAIL PROTECTED] :)

I've written rather much more than I intended, but have said rather less
than I thought.  In the end, I've always been more of a library/tool
builder than an application builder, so the place I think I can do the
most good in the ASF is Apache Commons.  But we have a bootstrapping
problem (there is as of yet no Apache Commons code) and I'd like to
help get the project off the ground.

daniel


Reply via email to