incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel Gruno <>
Subject Re: [VOTE] Graduation of Apache Spark from the Incubator
Date Thu, 06 Feb 2014 18:50:40 GMT
On 02/06/2014 07:34 PM, Michael Joyce wrote:
> Let's try this again Marvin since you're jumping to the wrong conclusions
> here.
> If people are concerned about work on Github not being automatically
> mirrored to ASF mailing lists that's an Infra problem. They should go talk
> to them to help get it addressed, which includes the possibility of
> contributing code. I'm not saying that archiving project relevant
> discussion is Infra's job.

Then PLEASE PLEASE PLEASE stop saying "It's an Infra problem". People
are making it seem like the blame is on Infra when $project does
something outside the organization and it doesn't get recorded. That
responsibility is SOLELY on the project/PMC itself. If however you/they
are interested in figuring out how to pull in these off-site/off-org
discussions to the mailing lists, then they should talk to Infra and
work something out (we _are_ working hard on a solution, for the
record). Just stop saying it's an Infra problem - it's a project
problem, 100% an issue with the project, but Infra may be able to make
convenience archiving for them (I'm stressing *convenience* here, it
does not magically make doing all your work on GitHub legit).

Having said that, I have yet to see someone step up and tell Infra "Hey,
we need your help in doing this, how can we _collaborate_ on a
solution?". So...please do.

With regards,

 I figured that would be obvious to everyone
> here, but evidently I'm mistaken. Hopefully this helps to clear that up.
> Are you really curious as to why I would assume that basic integration with
> Github would exist when I'm aware of an already existing integration with
> pull request notifications? When there are hooks that exist for comments
> being posted to pull requests, I don't think there's a very large leap to
> assume that we would leverage that. That doesn't mean that I think it's
> Infra's job to archive for a project.
> A response like Jake's is exactly the kind of response that this discussion
> needs. Someone with an understanding of the system explaining exactly what
> the limitations are. That way, people like me (people who don't bother with
> the Github mirrors) can better understand exactly what they can do to help
> improve them in the future if they decide that's what they want to do.
> Thank you for always being helpful Jake, it's very much appreciated.
> -- Joyce
> On Thu, Feb 6, 2014 at 9:53 AM, Marvin Humphrey <>wrote:
>> On Thu, Feb 6, 2014 at 9:08 AM, Michael Joyce <> wrote:
>>> How is doing a review on Github any different than doing a review on
>> Review
>>> Board?
>> One is captured to Apache controlled channels and the other is not.
>>> If there's concern that work on Github isn't being adequately mirrored on
>>> the mailing lists then that sounds like an Infra problem to me.
>> It's not an Infra problem.  It's the problem of any PMC which fails to
>> ensure
>> that all of its communications are properly archived.
>> If anyone here is interested in contributing towards this feature, I
>> encourage
>> you to subscribe to the infrastructure-dev@apache list.  I also encourage
>> everyone to ponder carefully:
>> *   How to ensure that no information is lost when capturing communications
>>     in GitHub channels through notifications to our dev lists.
>> *   The impact of adding GitHub integration features on long-term ASF Infra
>>     labor costs.
>>> There are plenty of hooks [1] that makes this easy to do and, if I'm not
>>> mistaken, pull requests through Github are already supposed to mail dev@
>> .
>> That's right.  I've actually worked on the specific hook that does
>> that[1][2].
>>> I would assume that any comment on them should as well.
>> Why would you "assume" that GitHub comments are being mailed to dev lists?
>> They are not.
>> Every Apache PMC member is tasked with oversight of their project, and that
>> includes ensuring that all decisions happen on the dev list and are
>> properly
>> documented.  It is not enough to "assume" that Infra is taking care of
>> archival -- ensuring that the archival actually happens is the PMC's direct
>> responsibility.
>> Marvin Humphrey
>> [1]
>> [2] Thread on legal-discuss@apache:
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail:
>> For additional commands, e-mail:

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

View raw message