incubator-libcloud mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gav..." <ga...@16degrees.com.au>
Subject RE: [libcloud] [VOTE] - Libcloud to become a TLP.
Date Sun, 28 Nov 2010 23:34:01 GMT


> -----Original Message-----
> From: ant elder [mailto:ant.elder@gmail.com]
> Sent: Sunday, 28 November 2010 7:31 PM
> To: libcloud@incubator.apache.org
> Subject: Re: [libcloud] [VOTE] - Libcloud to become a TLP.
> 
> On Thu, Nov 25, 2010 at 5:08 PM, Gav... <gavin@16degrees.com.au> wrote:
> > Hi All,
> >
> > This is a chance for the libcloud community and committers to
> > vote on whether or not you think we are ready to become an
> > ASF top level project. In other words, the community thinks
> > it has learned all it can from the Incubator, and that we feel
> > we are following the guidelines learned along the way.
> >
> > Please vote now, we have plenty of time until the next board
> > meeting is due in which a resolution would be required, so we'll
> > leave this open for 1 week to gather as many votes as possible.
> >
> > [ ] - +1 - yes please, we are ready to become a TLP.
> >
> > [ ] - -/+0 - Not that opinionated either way.
> >
> > [ ] - -1 - no , I don't think we are ready , because ...
> >
> > Note any -1 must be accompanied by a detailed explanation.
> >
> > This vote will close one week from today.
> >
> > Gav...
> >
> >
> >
> 
> I'm -1 for the reasons described at:
> http://mail-archives.apache.org/mod_mbox/incubator-
> libcloud/201011.mbox/%3CAANLkTim8ViOH3-D9TVbpUbgySTjWXns-
> RRv_4iSELw06@mail.gmail.com%3E

Well, let's bring those reasons here:

Quote: "... 

[1]I'm not sure I'm convinced there's nothing more to be learnt here.
Dealing with what to do with the Java impl would help at least
demonstrate the community can deal with an awkward situation, just
leaving it in the sandbox seems like a holding pattern to me.

[2] But even if thats not done i don't get the warm fuzzy feeling that the
libcloud
community has learnt the Apache Way yet. It would be nice to see not
just the same one person doing all the releases, all the committer
nominations, most of the commits etc, even with graduation there
hasn't been any real discussion that i've seen about what the
graduation requirements are and why they've been met. 

[3]If for some reason you were unable to continue here Paul i'm not seeing
much
evidence that the others would know all the ropes to continue. What do
mentors/others think about this?

..."

I've split them into 1,2,3 and address them here:

[1] - Paul has already explained this to my satisfaction. A sandbox area is
fine for
the java impl. The project need not have took it at all but they did. Mainly
one person
works on this area currently and is happy with its progress. The whole of a
projects
code need not be in a state of release for it to graduate from the
incubator. Accepting
the code and granting committer status and an area for this to be developed
shows
that the community dealt with the situation just fine.

[2] - One person doing the releases is fine for now. I know plenty of other
podlings that
graduated with only one RM - in some cases with only one release. I know
plenty of
Top Level Projects that have had the same RM for years. I am involved with
one such project.
Over time, other committers will learn to see and nominate prospective
committers, again not
a big deal IMHO and not something that other podlings have been delayed for.
Commit diversity: http://www.ohloh.net/p/libcloud/contributors?page=1 shows
16 committers - yes
Paul has done more than anyone else, but 316 out of 694 is less than half
total commits. Again
I can show you many many projects where one committer has more than double
the next person, and
especially when the project first starts out, that is just the way it is,
nothing to worry
about IMHO.
I am going to agree with you with regard to graduation requirements to a
point. It is up to the
Mentors I feel to point out and help address these requirements if they are
not being met. We
have been mostly hands off Mentors - which is fine, but it does show that
the project in the
main has been moving along by itself.

As both you and I have mentioned before, the only thing I can see that I
would like addressed
is the question of diversity - would someone who is not a mentor please edit
the document
at http://incubator.apache.org/projects/libcloud.html and put down your
employer if your
employer is paying you to work on this project, otherwise simply write
'independent' beside
your name. I will hold back my vote until this is done.

[3] - I don't want to speculate or penalise a project for something that
might or might not
happen in the future and especially not knowing the state of the project or
the knowledge
of the other committers at that time should it happen. It would be wrong to
hold a -1 to
graduation based on this.

--

So summary then Ant to your questions, I think overall I am satisfied with
the project and its
committers readiness to graduate. In places I think you are perhaps being a
little too strict
though I completely understand your reasonings. I guess I have a more
liberal view of the 
incubator as a whole, and comparing libcloud with other projects that I know
to have graduated
or failed, I feel they are generally in good stead, if a little lazy on the
paperwork.

If I see that the http://incubator.apache.org/projects/libcloud.html page
gets updated correctly
as I have asked before the deadline, then my +1 will be forthcoming.


Gav...


> 
>    ...ant



Mime
View raw message