incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Leo Simons <>
Subject Re: [discussion] Harmony podling to ask for vote for graduation
Date Thu, 19 Oct 2006 09:00:42 GMT
Last e-mail on this for the day, promised! :-)

Greg -- having many e-mails as threaded responses rather than a  
single one is useful if you have nested threading and you're only  
interested in part of a sub-sub-thread. Maybe gmail needs fixing :-P.  
In any case, here's your overview.

On Oct 17, 2006, at 6:44 AM, Geir Magnusson Jr wrote:
> I believe the active mentors (myself, Stefano, Leo and Dims) will  
> assert
> that the [harmony] community is healthy and very active.


> We'd like to help make this as efficient a process as possible -  
> please
> let us know if there are any issues that will prevent a successful  
> vote
> by the PMC on our graduation.  If there are no un-addressed issues in
> the next 3 days, I'd like to call a vote late on Thursday, October  
> 19th,
> so we can possibly be finished in time for the next Board meeting,
> October 25th.

Issues I saw raised:

== should have all legal ducks in a row ==

Check. Double check. Triple check. Pass. Best student in class.

== should not have a language implementation at apache at all since  
that's too big a project to fit with the rest of the ASF ==

With a harmony/incubator hat on, I disagree with that (since the  
stated goal of harmony is to be a language implementation!), and in  
any case I think it shouldn't be a factor in deciding on graduation.

With an ASF member hat on, I do continue to be concerned about the  
ASF its ever-increasing size, but that's not something that harmony  
itself can really address or is responsible for.

== should have done a release ==

I disagree with this. The release requirement is a (undocumented as  
required!) way to gauge that a project's developer community can jump  
through various hoops and be responsible about managing "big issue"  
things. This community has shown so already.

I think the fact that harmony has not done a "full release cycle" is  
actually a healthy thing for the project. The slowdown inherent to  
that kind of cycle for harmony will come because of needing to do TCK  
testing, and that very TCK testing will put more requirements on the  
project anyway than a normal ASF-style release process.

I think the kind of release process that harmony needs to have is  
very different from most existing ASF projects (for one, it needs a  
little server cluster to run tests on) and that any comparisons are  
therefore not very useful.

== So, where are we? ==

I don't think the "should have done a release" issue has been  
"addressed" (since harmony hasn't done one with all the bells and  
whistles, just snapshots). *I* don't think it needs to be addressed  
before graduating out of the incubator, in the case of harmony. It's  
not completely clear what "the incubator PMC" thinks about that topic  
(if anything, this shows how hard it is for the IPMC to make up its  
mind...we really do seem to have a pendulum opinion when it comes to  
this stuff). Me being me, I would just call a vote and find out.


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

View raw message