On Wed, 3 Aug 2016 07:09:50 -0700, Ralph Goers wrote:
On Aug 3, 2016, at 3:46 AM, Gilles <gil...@harfang.homelinux.org>
wrote:
On Tue, 2 Aug 2016 23:59:23 -0700, Ralph Goers wrote:
On Aug 2, 2016, at 4:00 PM, Gilles <gil...@harfang.homelinux.org>
wrote:
Most PMC members seem to wish that CM becomes dormant.
I totally agree with you that it would be good to have _that_
clarified.
I have no idea how that is your takeaway from all the discussions.
The point I have tried to make is that CM needs a community of
committers, not just you.
The current situation is that there is only me (with commit
privilege).
Without me, all development activity would have stopped more than
8 months ago. [That's an observational fact.]
There are volunteers for following up on my proposals but, indeed,
their statements seem to count for nothing.
Isn't it awkward that people like Rob Tompkins, formerly interested
in contributing to CM feels compelled to find "something else" to do
here in order to get noticed, and perhaps later (or perhaps never)
get the authorization to do what he really intended to in the first
place?
There isn’t enough to do in CM to warrant becoming a Commons
committer? I am pretty sure that isn’t true. People don’t have to
work
on different things, they just have to work on enough and participate
enough to have someone propose them as a committer. BTW - did I miss
your nominations for committers to help on Math?
Nominations? Based on what?
Everyone here has the same information as I do about the would-be
committers.
Would you vote positively at this point?
They want to work on CM but that would imply that I am the sole relay
to the repository, for reviewing, commenting, checking, committing,
With current CM, this includes bits of code on which I'd rather not
comment anymore...
If this PMC intended to discourage contributors, that would be a
nice way.
My position has always been that having
discussions about what to do with the code is a waste of time when
you
are the only committer doing anything.
IMHO, you get things upside down (as did the CM team all along):
people come to contribute because they are interested in the code
(be it to add to it, up to completely overhaul it, from time to
time).
Where the project is heading to is a fundamental aspect for
deciding whether one wants to contribute.
As an example, Artem Barger is interested in improving the
"machine-learning" package.
As it happens, I'm also interested in that part of CM. Why
should we have to carry the burden of the pack of codes left
behind by the forkers and _literally_ waste our time maintaining
something that either we don't use or needs thorough refactoring?
The extracting of modules would make it clear to users and
would-be contributors what is currently being worked on and what
is in need of maintainers.
But the Commons PMC does indeed "prefers" a monolithic and
_dormant_ CM.
Bull pucky. That is just your interpretation.
No, it's not just an interpretation.
Several people, not contributing to CM (Oh, again, I said it)
said they "prefer" a single component. When asked to explain
their rationale, there is just "feeling".
So if they won't "feel" like voting for a release, it's disrespectful
to ask people like Rob or Artem to work on these codes.
I can understand that not everyone is interested in those
components, but that there is only a minority (i.e. me) here
interested in them should prompt for action, not idly waiting
for the matter to go into oblivion. That would the role of
a useful PMC.
James tried to do something.
Jochen tried.
I tried.
Everything blocked by a diffuse "feeling".
I have no problem with
whatever the community wants to do.
If that were true, you could have said that the newcomers who
want to work on a revised CM are welcome to do so, and the
output of that work would normally be adopted by Commons
(unless it's proven crappy of course).
Moving Math to the incubator
would have allowed you to have a much lower barrier to add new
committers, but you didn’t really want to discuss doing that.
This is plain false.
Incubator PMC people said that it was one-of-a-kind situation,
noting that the incubator's usual task was to create an Apache
project around an existing community, not to discuss how to
create one.
Bull pucky again.
Send that to the concerned people.
The Logging PMC (of which I am currently the
chair) moved Log4cxx to the incubator to try to build a larger
community. All the committers had disappeared but we had people on
the
mailing list saying they wanted to work on it. So we moved it to the
incubator where they got commit access. The Logging project is much
like commons where when you have commit access you can work on any
subproject, so we had the same reluctance to give unproven people
commit access.
Again, wrong, wrong, comparison: logging is limited scope, "math"
is not. We go to the incubator to do _what_?
That's a real question. If Jochen has answers...
Gilles
Ralph
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
