David,
Nice add to "OFBiz Contributors Best Practices", especially the roles/responsibilities of
committers. That way, folks like us will know that when the doctor refuses to give us rat poison,
it means the good doctor is doing his/her job. So, I believe committers can now safely and
conveniently issue a "please read this that this, patch rejected, but thanks, please come back
again!", without driving contributors away?
But that said, it's not easy to go through that document. I know most of the comms tools for OFBiz
by now (ML, JIRA, etc), so I dig that document even with casual glance through. But most folks
(even I, for easy protocols I can follow when sleepy) will need a step-by-step for working with
and/or contributing to OFBiz. Eg, 1. write issue to ML for discussion, 2. propose solution, 3.
initial "vote/discussion" phase over, 4. put on JIRA, etc (assuming you don't want to clog up JIRA
with non-issues). It will also be easier if we had a template for such procedures, eg "write [RFC]
or [RFImplement] or [RequestForWhatever] in subject heading".
I guess what's missing in ML is the structure in JIRA (enhancement type, bug
type issues, etc).
I know, it's not easy to do the above. Writing docs by drawing from bits and pieces in our
experience here and there is easy; further organizing that into structured flows is additional work.
We'll get there, I believe. :)
Anyway, nice work. Once it's common knowledge (put billboard ads in every country if necessary,
kidding) that committers' jobs are delicate and critical, contributors won't easily be turned off
when their patches get rejected for bad formatting/protocol/etc.
Now, if I can just get my doctor to prescribe "more work" to fix "bad
eyesight". Hmm.
Jonathon
David E. Jones wrote:
On Jan 20, 2007, at 8:38 AM, Jonathon -- Improov wrote:
Jacques,
Are you the only one reviewing the submissions? Any way the rest of us
can help? I see the long "customer service-like" thread at
https://issues.apache.org/jira/browse/OFBIZ-552 . Or perhaps if you're
a committer, you should be relieved of many other duties?
This is always a fun topic... and has been discussed in a few recent
threads on the mailing lists.
First off I agree that Jacques has been helping a lot and in spite of
some mis-steps (more about that later), Jacques has been helping a lot,
and it's really great, and people submitting patches that he gets
through should thank him.
So why are things this way? Why are there so many open issues? Why can't
committers keep up? Why don't we just add more committers? Why don't you
just make ME a committer so I can get my stuff in?
The answers, my friend, are in... the following documents:
Contributors Best Practices: http://docs.ofbiz.org/x/r
Committers Roles and Responsibilities: http://docs.ofbiz.org/x/mQ
I just revised and updated these documents as a few things in them
weren't quite correct (like the process for adding a committer), and
there were a few questions, including parts of this one, that I think
weren't adequately addressed.
After reading through those documents you should have a better
understanding of how things work, or don't work. By definition if there
is a back log of issues, there is a shortage of volunteer-time. It's
that simple. Hopefully this is mostly caused by people not knowing where
or how they can help, and now hopefully those documents address that
much better.
-David
