commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <>
Subject Re: [plea] get back to work
Date Tue, 23 Dec 2003 11:31:39 GMT
On Tue, Dec 23, 2003 at 10:59:02AM +0000, Danny Angus wrote:
> > I'll say again - the Board could avoid this damage by
> > choosing to ACT rather than REACT.
> It is extremely unlikely that the board will become involved in this unless
> either the PMC request it or it directly affects the board's business. The
> right people to decide the future of Jakarta commons are the Jakarta
> commons commiters.

While it might be nice to request help from "The Man", I seriously doubt
that will happen. As Danny put it: this is something for Jakarta to work
out, not for the Board to step in and deal with.

Generally speaking, the Board has two positions:

1) the Project is not damaging the ASF (but note it might not actually be
   making progress, but the ASF doesn't *require* forward progress)
   --> leave it alone and let it work out whatever issues it feels it has

2) the Project *is* damaging the ASF

   --> shut it down

There isn't a lot of middle ground for the Board to work within because it
doesn't know the Right Answer for the given community. What would happen
if the Board started thinking it *did* know and began to interfere with
the operation of the Projects throughout the ASF. Chaos. Revolt. :-)

That's my personal position with my Director's hat on. I'll posit that it
is also pretty close to the rest of the Board's position. Also note that
the Board is pretty conservative. Things have got to get pretty bad before
we take an action such as (2). I couldn't even conceive of shutting down
Jakarta; that kind of action would be more likely for smaller or newer
PMCs rather than something like Jakarta. (although I will say that things
like the cornerstone contribution peeve me immensely)

That said, the Board *will* convey directives PMCs about concerns it may
have. For example, a lot of the current traffic on the Jakarta PMC and the
action on achieving better oversight is because the Board said it didn't
feel that the PMC was providing effective oversight over such a large
codebase (hundreds of committers and how many hundreds of thousands of
lines of code?). Further, that the PMC in its current form is incapable of
doing so. Expanding the PMC might do it, but the simple fact is that
Jakarta is composed of many communities; many of them are more than
capable of handling themselves as a TLP (much as Stephen points out).
Shifting those will go a long ways towards validating that oversight is
possible and is present.

How does J-C fit into this? What I see is a loosely knit group of
developers each working on one or more components. There is definitely a
feeling of community, but the codebases and the people working on them
form discrete subgroups. As an easy example: the codec component really
only has two committers on it. I'm not sure that constitutes a problem,
but it also seems to indicate that the community != interested developers
which means that those developers cannot necessarily do what they want
(or believe is best) with the code that interests them.


Greg Stein,

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message