forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tim Williams <william...@gmail.com>
Subject Re: Proposal for Forrest-Cocoon-Lenya commit access
Date Thu, 01 Sep 2005 12:57:19 GMT
Since I've learned most of what I know from reading words from you
guys, it surprises me that I come to a different conclusion.

As I understand it, committership is based on merit and one earns
merit only *after* consistently making good contributions
demonstrating that they've earned it.

I also, personally think that when they've earned merit, also should
come PMC membership (but that, I suppose, is a different discussion).

Ironically enough, it was Nicola who clarified this best in my own
mind with his Ant experience:
http://marc.theaimsgroup.com/?l=forrest-dev&m=112348437010165&w=2

So are we now saying that "merit" applies across the ASF?  That just
doesn't make sense to me given the diversity of projects here.  I mean
my own interpretation is that there are two parts to merit: 1)
Technical and 2) Community/Social.  I would concede in general that
the second applies to most if not all projects but it just can't be
said that technical merit does -- one could look at my c++ skills or
lack thereof for evidence.

If merit isn't at the foundation level, then it needs to be earned at
the project level. Earned, of course, being the operative word.  And,
how could it be earned without consistently good contributions. This
proposals gives blanket merit to folks that have not, with this
definition at least, earned it.  How does one reconcile these?  I can
only do this with by changing "committership==merit" to
"committership==(merit || trust)" but that seems like we're redefining
some longstanding ASF concepts.

Please note that I'm not saying that anyone on Lenya or Cocoon
couldn't earn merit (on the contrary, I'd concede that their Cocoon
knowledge would make it very easy to do so), just that they haven't
yet.

--tim

On 9/1/05, David Crossley <crossley@apache.org> wrote:
> There have been various discussions about this
> topic, some of them on various PMC mailing lists.
> There is no need for it to be private, so now it is
> in the open. The developer community needs to
> understand the background and the related issues.
> Private discussions do not enable the community
> to be involved.
> 
> We all, especially new Forrest committers, need
> to understand the ramifications of this.
> 
> The main effect that i see is that we would be
> opening up the Forrest repository to people that
> we do not know and are not familiar with.
> Is everyone happy with that.
> 
> This proposal is not about making people from
> these projects members of the Forrest PMC.
> The process of becoming a PMC member is up to
> each project. At Forrest, our process is different
> to Cocoon's and i don't know about Lenya's.
> 
> This proposal is only about convenience. Nothing is
> stopping anybody from sending patches to the other
> projects.
> 
> Here is the proposal. After a chance for discussion
> then we can go on to vote about it. Normally we don't
> need to vote on things, but we do in cases of changes
> in project direction and community composition.
> 
> Forrest commit access to Cocoon
> -------------------------------
> Cocoon PMC has recently asked Forrest PMC if it
> would be useful for our committers to have access
> to their SVN repository.
> 
> Cocoon commit access to Forrest
> -------------------------------
> Back in July 2003 before Forrest became a top-level
> project, we decided [1] to give Cocoon committers
> access to the Forrest CVS Repository. That is
> something that we need to re-decide, so now seems
> a good time. The reason that we need to revisit,
> is that the formality for creating Apache Forrest
> as a top-level project includes defining our
> project guidelines which includes deciding who
> gets commit access.
> 
> Forrest commit access to Lenya
> ------------------------------
> Lenya have recently decided to give Forrest
> committers access to their SVN repository. [2]
> 
> Lenya commit access to Forrest
> ------------------------------
> These projects are trying to collaborate on
> better integration. It might make that easier
> if Lenya committers had access.
> 
> Related issues
> --------------
> Just so that the newcomers know, normally the process
> of going from developer to committer happens entirely
> within each project. You do not get commit access to
> other projects (other than xml-commons).
> 
> This situation between the Cocoon-based projects
> is some new experiment.
> 
> As is the norm, the PMC of each project has the
> final say over their own codebases and communities.
> With the code, they can veto a change and roll back
> the SVN.
> 
> Even though these committers would have access,
> they need to pay attention to the desired direction
> of the project. If they don't understand that,
> then they should just contribute a patch.
> 
> Such access is enabled by "svn authorisation".
> For example, here is the current cocoon definition:
> [groups]
> ...
> cocoon=stefano,balld,... list of all cocoon committers
> ...
> lenya=andreas,alexmc, ... list of all lenya committers
> ...
> [/cocoon]
> @cocoon = rw
> @lenya = rw
> #-- begin Cocoon GSoC students
> [/cocoon/gsoc/mpfingsthorn]
> mpfingsthorn = rw
> ... etc.
> 
> See how they restrict access to certain parts
> for certain groups.
> 
> Forrest currently only has one "group" and that
> is all Forrest PMC members. We also have this
> new concept of committers that are not PMC members.
> We don't have any of those yet. When we do, they
> would be a separate group. For example, Lenya
> has a "lenya-soc" group for their GSoC people.
> 
> This proposal is only about the main groups.
> 
> References
> ----------
> [1] past decision about Cocoon access to Forrest
> http://marc.theaimsgroup.com/?t=105645457400003
> http://marc.theaimsgroup.com/?t=105738941000001
> 
> [2] decision about Forrest access to Lenya
> http://marc.theaimsgroup.com/?t=112551792600003
>

Mime
View raw message