hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stack <st...@duboce.net>
Subject Re: [DISCUSS] HBase as Apache top-level project?
Date Thu, 01 Apr 2010 14:56:38 GMT
On Thu, Mar 18, 2010 at 5:00 PM, Jonathan Gray <jgray@facebook.com> wrote:
> Annoyance has really not gotten us anywhere.  And I don't think it matters to those
in Hadoop whether we are a TLP or SP, they will not (or should not) be offended if we break
off.  Do you think they would take us (or our patches) less seriously if we were a TLP?


> What has pushed things forward is continuing to make HBase better so that more people
want to use it.  A larger community and involvement from larger companies will help push
Hadoop changes aimed at HBase, especially when those companies are Hadoop contributors.


> If we do think we can get some HBase committers onto the Hadoop PMC, and we think that
this will make a material difference in outcomes for us, then my opinion may change.  Today
I don't really think the issue is whether we are on the Hadoop PMC or not... my understanding
is that big decisions are not voted on for a majority, if someone votes against it then it
is tabled.

The quoted rule where long-time Hadoop subproject committers become
Hadoop PMC members may not actually hold.  Or, to put it a another
way, efforts at trying to take advantage of this rule have run into
resistance, understandably so -- as in, how does hbase expertise
entitle a committer to hdfs commit rights? -- and I do not intend to
push it further.

So, that leaves the Jay Booth "stay and be annoying -- in a good way"
opinion, outstanding as a reason not to move.  My current thought on
this is that the work involved will be the same regardless -- i.e. the
patch making, JIRA bashing, and consensus building -- whether we're
under the hadoop umbrella or up on a TLP perch.


View raw message