commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Phil Steitz <>
Subject Re: [DISCUSS] Karma necessary for Release Manager roles
Date Tue, 23 Jun 2015 20:29:50 GMT
On 6/23/15 12:55 PM, Gary Gregory wrote:
> On Tue, Jun 23, 2015 at 12:30 PM, Stefan Bodewig <> wrote:
>> On 2015-06-23, Gary Gregory wrote:
>>>> sebb wrote:
>>>>> I think it would be useful to allow non-PMC members to act as
>>>>> release manager.  This means relaxing some of the existing technical
>>>>> restrictions.
>>> Just be clear, that would mean that any Apache Committer could release
>> any
>>> Apache Commons component.
>> ... could prepare a release and call for a vote.  It would still by the
>> PMC vote that made it a release.
>> This is not what Sebb suggested, aynway.  He talked about re-purposing
>> the now unused Commons LDAP group for non-PMCers that are allowed to cut
>> releases.
>>> On Tue, Jun 23, 2015 at 11:49 AM, Thomas Neidhart <
>>  wrote:
>>>> I think that every committer shall be able to do a release for a
>>>> commons component after it has been discussed on the mailinglist.
>> I agree, we'd still have PMC oversight.
> Well, no, because once you are in Nexus, you can release, so if any Apache
> Committer can push the release button, that's not PMC oversight.

You need to distinguish between access control and oversight.  Until
Nexus came along, there really never was anything preventing any
commons committer from publishing release artifacts.  To my
knowledge, we had just one "unsanctioned" release pushed to maven
repos (by mistake back in the wild m1 days where it was relatively
easy to inadvertently push artifacts) and we never have had a
committer not follow the process of cutting RCs and pushing only
successfully voted artifacts to the mirrors / maven repos.  I don't
personally see the need to use access control to prevent committers
from circumventing the release process.

I am +0 for just letting any ASF committer RM commons releases.  I
agree with Thomas, though, that help from others for new RMs is good.

The +0 vs +1 for me has to do with supporting the release.  Now that
we don't vote and offer committership to ASF committers (automatic
commit), there is no acceptance of any kind provided by ASF
committers who contribute here.  That's fine for CTR patches, etc;
but I would like to see a little more when it comes to a release. 
Joining the PMC is probably a bigger commitment than necessary; but
it works4me to signal a connection to the project.

> Gary
>> The middleground suggested by
>> sebb would work for me as well, but I don't see why it would be
>> necessary.
>> Stefan
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail:
>> For additional commands, e-mail:

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

View raw message