Thanks for the feedback, Craig.
Excerpting and reordering from your mail...
> It would be good to consider formalizing the governance process
> for the benefit of other projects.
Yep, I agree. From one of my previous emails...
>>>>> The process should be consistent across all GlassFish projects, so
>>>>> I'll put together a better fleshed strawman and will run it by
>>>>> people...
and then:
>>>>> OK, since we are doing this for the first time, I propose this
>>>>> (strawman) protocol:
>>>>>
>>>>> * An existing committer (the 'advocate' - better name?) proposes a
>>>>> new committer (the 'candidate')
>>>>> * All other committers should consider whether the new candidate
>>>>> has shown technical & team abilities needed to be granted commit
>>>>> status
>>>>> * Committers should vote.
>>>>> * Advocate tallies the votes and reports back to the DEV alias.
>>>>>
>>>>> We will need to figure out how many +1 vs -1, etc, etc. But I
>>>>> suspect this is not going to be an issue for Wonseok.
Back to your comments... I think they are intended as feedback on the
strawman proposal, right?
re: +1, -1, +0, -0
There is quite a bit of variance on rules in Open Sourced projects.
From benevolent dictatorship all the way to full democracy. Some
examples, not intended to be representative...
https://glassfish.dev.java.net/public/GovernancePolicy.html
http://www.eclipse.org/projects/dev_process/
http://httpd.apache.org/dev/voting.html
http://www.apache.org/foundation/voting.html
http://www.mozilla.org/mission.html
http://www.mozilla.org/owners.html
http://www.genunix.org/wiki/index.php/OpenSolaris_Governance_Draft_02
Etc...
I'm a bit worried about "early optimization/binding" of the process
before we have more real experience in our community. We could copy
wholesale something, but I think a governance has to be well matched to
the community it serves, so my proposal is to:
Define a strawman that is "good enough" and define a review and refine
process that we can use to use real experience to evaluate and evolve
that strawman into a more permanent governance. OpenSolaris took about
a year to get theirs done. Maybe ours will take less time, we will see.
> In my experience, tallying votes with annotations [not legally allowed
> to vote] or similar language discourages community building.
Humm... Do you mean we should count all votes? Or that we should list
all votes and only count the committers? Or something else?
Counting all votes seems impractical as a rule. The whole principle of
having committers is one of delegating decisions to them ("they know
better"), so using the committers list seems the most natural approach.
I could list votes differently if you think that would make a
difference, but I think somebody has to list them and track them or we
will miscount / get confused. Specially in high mail count situations.
Does this help?
- eduard/o
Craig L Russell wrote:
> Hi Eduardo,
>
> Just a couple of comments.
>
> A vote of +0 or -0 should actually be counted as a vote. +0 means I
> don't know or care enough about the issue but I'm happy with the
> proposal. -0 means I know about the issue and I'm not convinced it's a
> good proposal but don't have the energy to actively oppose it. The
> absence of a vote should not be counted as a 0.
>
> In my experience, tallying votes with annotations [not legally allowed
> to vote] or similar language discourages community building. It's good
> that the processes be well understood so we know at the end of the vote
> whether it passed or not, but calling out voters doesn't have the effect
> we typically want in the community.
>
> It would be good to consider formalizing the governance process for the
> benefit of other projects.
>
> Regards,
>
> Craig
>
> On Nov 14, 2006, at 11:36 AM, Eduardo Pelegri-Llopart wrote:
>
>> Thanks, Tom.
>>
>> I see a few more came in. Here is my tally:
>>
>> rochelle +1
>> Jielin +1
>> tom +1
>> [Peter K +1 -- but he is not a committer, right?]
>> guy +1
>> chris +1
>> andrei +1
>> gordon +1
>> Markus +1
>> marina +1
>> [Craig R +1 -- but he is not a committer, right?]
>> Michael +1
>>
>> that is 10 +1, two known 0 (vacation & new to project), 0 -1.
>>
>> I think that Mitesh and Sahoo have not voted, right? They are in
>> india TZ, so I think it would be appropriate to give them a chance to
>> vote overnight.
>>
>> We still need to decide what "-1" means :-), but so far, it seems we
>> can decide that separatedly.
>>
>> - eduard/o
>>
>>
>> Tom Ware wrote:
>>> Of the 14 current committers, I have recieved 9 votes in support, and
>>> 0 votes against. Of the non-voters, 1 is on vacation and 1 is new to
>>> the project and I would not expect to see votes from either of them.
>>> I believe that is adequate support to move forward.
>>> -Tom
>>> Eduardo Pelegri-Llopart wrote:
>>>> I have listed all the committers at the TopLink Essentials page. See:
>>>>
>>>> https://glassfish.dev.java.net/javaee5/persistence/index.html#committers
>>>>
>>>>
>>>> Based on this proposed protocol, Tom is responsible to report back
>>>> on everybody's vote. Votes are not private. We will discuss,
>>>> separatedly and in a public forum, how the GlassFish community
>>>> should handle arguing the merits of a given candidate, but I am
>>>> fairly confident that we will not have that problem in the case of
>>>> Wonseok and we don't want to impede progress while we figure out the
>>>> process.
>>>>
>>>> thanks,
>>>> - eduard/o
>>>>
>>>> Eduardo Pelegri-Llopart wrote:
>>>>
>>>>> OK, since we are doing this for the first time, I propose this
>>>>> (strawman) protocol:
>>>>>
>>>>> * An existing committer (the 'advocate' - better name?) proposes a
>>>>> new committer (the 'candidate')
>>>>> * All other committers should consider whether the new candidate
>>>>> has shown technical & team abilities needed to be granted commit
>>>>> status
>>>>> * Committers should vote.
>>>>> * Advocate tallies the votes and reports back to the DEV alias.
>>>>>
>>>>> We will need to figure out how many +1 vs -1, etc, etc. But I
>>>>> suspect this is not going to be an issue for Wonseok.
>>>>>
>>>>> The process should be consistent across all GlassFish projects, so
>>>>> I'll put together a better fleshed strawman and will run it by
>>>>> people. Sometime after the flurry of this next Monday (tomorrow) :-)
>>>>>
>>>>> - eduard/o
>>>>>
>>>>> Tom Ware wrote:
>>>>>
>>>>>> Hi All,
>>>>>>
>>>>>> I propose we grant commit access for the entity-persistence module
>>>>>> to Wonseok Kim.
>>>>>>
>>>>>> Wonseok, has been contributing high quality code on a regular
>>>>>> basis to the project and also been providing valuable feedback to
>>>>>> other particpants. The quality and quantity of work he has been
>>>>>> providing indicates to me commit access should be granted.
>>>>>>
>>>>>> Please reply either to the persistence list or directly to me by
>>>>>> the end of day on Monday (November 13) I will consider
>>>>>> non-response to indicate support.
>>>>>>
>>>>>> Thanks,
>>>>>> Tom Ware
>>>>>>
>>>>>>
>
> Craig Russell
> Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
> 408 276-5638 mailto:Craig.Russell_at_sun.com
> P.S. A good JDO? O, Gasp!
>
attached mail follows:
I had sent this to DEV_at_glassfish.dev.java.net. I had assumed that all
groups were represented there but, just in case, here it is.
Jim described them as "Mom and Apple Pie"... :-)
- eduard/o
>> I have written down an early draft of Principles for Project
>> GlassFish. See:
>>
>>
http://blogs.sun.com/roller/page/theaquarium?entry=draft_of_principles_for_project
>>
>>
>> Comments via comments on blog, comments on Wiki page, or in this mail
>> thread.