spark-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <>
Subject Re: [VOTE] Designating maintainers for some Spark components
Date Fri, 07 Nov 2014 05:27:59 GMT
On Thu, Nov 6, 2014 at 7:28 PM, Sandy Ryza <> wrote:

> It looks like the difference between the proposed Spark model and the
> CloudStack / SVN model is:
> * In the former, maintainers / partial committers are a way of
> centralizing oversight over particular components among committers
> * In the latter, maintainers / partial committers are a way of giving
> non-committers some power to make changes

I can't speak for CloudStack, but for Subversion: yes, you're exactly
right, Sandy.

We use the "partial committer" role as a way to bring in new committers.
"Great idea, go work >there<, and have fun". Any PMC member can give a
single +1, and that new (partial) committer gets and account/access, and is
off and running. We don't even ask for a PMC vote (though, we almost always
have a brief discussion).

The "svnrdump" tool was written by a *Git* Google Summer of Code student.
He wanted a quick way to get a Subversion dumpfile from a remote
repository, in order to drop that into Git. We gave him commit access
directly into trunk/svnrdump, and he wrote the tool. Technically, he could
commit anywhere in our tree, but we just asked him not to, without a +1
from a PMC member.

Partial committers are a way to *include* people into the [coding]
community. And hopefully, over time, they grow into something more.

"Maintainers" are a way (IMO) to *exclude* people from certain commit
activity. (or more precisely: limit/restrict, rather than exclude)

You can see why it concerns me :-)


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message