Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Kafka >> mail # dev >> committer and pmc requirements


Copy link to this message
-
Re: committer and pmc requirements
Where Jay has set the bar for now sounds good to me.  Yeah
review-before-commit!

I think the long term goal of committers == PMC is the right one.  While
I am not sure if they have it set as formal policy it seems common on
projects I have observed for new committers to become PMC members $n$
months later.  Being a solid active committer for (say) 3 months is good
enough for me.  But keeping the initial committer/PMC votes separate
lets us keep flexibility to adjust the committer requirements without
worrying about PMC ones.

+1 on stating a formal emeritus policy up front (but keeping a
relatively long timeout, other projects can easily consume a quarter).

On 03/14/2012 12:14 AM, Jay Kreps wrote:
> Hey All,
>
> One thing suggested to us by our Apache mentors was to formalize the
> criteria for becoming a committer and pmc member. Different projects have
> different criteria in this regard. What are people's thoughts in this
> regard?
>
> FWIW, here are my thoughts. I would suggest we hold a high bar on technical
> capability but a fairly low bar on level of contribution. I think 3+
> substantive patches plus an interest in ongoing involvement should be
> enough. I think this is more appropriate to a young project such as
> ourselves. Since we do blocking code reviews for both committers and
> non-committers this doesn't put us at too much risk of weak code creeping
> in. I don't have any thoughts on what would be a good standard for PMC
> membership. It might also be a good idea to make people "committers
> emeritus" after 6-12 months of inactivity. I have found on past projects
> that committers tend to accumulate to the point where a substantial portion
> of people are not active which is probably not the right thing for Apache
> since things are decided by voting.
>
> What do others think? Any past experiences of this "done right"?
>
> -Jay
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB