Dirk-Willem van Gulik wrote:
On Mon, 23 Jun 2003, Steven Noels wrote:
Stefano's insightful post got me carried away to run some stats on
members & projects: http://blogs.cocoondev.org/stevenn/archives/001008.html
I've always stopped short of doing just this; and more kept things limited
to a pie diagram and postings/#of commits.
This as it mostly shows 'today' rather than the members body which grew
over time and is effectively lagging. I.e. you are looking at data which
tells you more about history than about the future. And that todays future
is tomorrows history.
Actually, the data tells you the results of history, not the path that
history took to get there, which is just as important.
Please comment if you care, but keep the thread on community (or
cocoon-dev). I'd love to hear your opinion.
My main interpretation is
-> We are tremendously dynamic in terms of ratio's and
relative numbers; things turn upside down regulary.
-> xml and java are 75% of the activity; the 'old school'
has dropped below 20% now (Ignoring PHP here).
-> Despite the enourmous influx of java and xml the ASF
as a whole is growing significantly slower than the internet.
Is this good or bad? I don't think that we can grow at anything close
to internet rates right now.
-> Documentation is growing even slower; even including
translations.
-> Organisationally xml and java are still lagging behind;
but have been catching up (though the catch up has slowed down
somewhat due to a much larger influx from the old school
side; and that influx is by average younger than the proposed
influx from xml and java (in terms of lines of code and/or years
of activity on *MORE* than one project).
In the case of XML, this stems partially from the manner in which the
project was formed. As was pointed out at ApacheCon this fall, both
Jakarta and XML were kind of experiments. This year a substantial
amount of effort has gone into trying to change the results of those
experiments -- namely the effort get projects to leave the umbrella and
become top-level projects. Note that for XML this happened in exactly
one case -- Coccoon. All the other projects in XML decided to either
stay in XML or move to ws.apache.org, another umbrella PMC structure.
-> Java (and to a lesser extend xml) is _actively_ under
represented and produces less orgaisational/infrastructure/legal
people than one would expect given the current relative number of
existing java/xml folks in organisational positions. That may
be a cultural thing.
I think that we have multiple subcultures under the ASF umbrella, due to the way that the umbrella projects were formed. Whether you like that or not, I think that is the reality. I know that I personally would have remained fairly isolated had I not attended an ApacheCon in 2001.
-> In the java, and to some extend the xml world, we have much, much
much more code which was only touched 1-4 times by <= 2 people
over time.
-> the java world seems to need amazing number of indians (or
committers) relative to lines of codes or bugs fixed. And seems
to see more isolated pockets of people than the xml and other
parts of the ASF.
My impression on this is that the folks at jakarta have been more free
(at least compared to projects in XML) with commit rights. I don't know
if this is actually the case, but it is my perception.
Just for thoughd (and the above was taken from a 2002 count; it may be off
now).
Dw
