On Dec 28, 2003, at 10:48 AM, Ted Husted wrote:
----- Original message ----------------------------------------> From: Stephen Colebourne <[EMAIL PROTECTED]> To: Jakarta General List <[EMAIL PROTECTED]> Received: Sun, 28 Dec 2003 14:16:26 +0000 Subject: Re: [PROPOSAL] As it ever were (draft 2)
Since the PMC cannot delegate its responsibilities, the report would
have to be prepared by a PMC member, ideally one directly involved with
subproject. (A likely suspect being the DEV list moderator.)
Er, doesn't this just emphasize how broken this process is?
Not that I see. It formalizes what should have been done from the beginning.
This is getting strangely confusing. What we all have been talking about is getting all possible people from the Jakarta sub-projects on the PMC. In doing that, we have coverage of the codebases, and people will be responsible for things they are involved in, *and*, in the event they see something in a project they aren't involved in, can get involved.
I agree w/ Ted, partially. From his previous note, suggesting that specific reporting responsibillity falls to the -dev list moderator, I disagree that this should be requires as this limits participation (instead of 2 possible people, it's back to 1), and might make someone not want to help out via list moderation if it also placed additional requirements on them.
We tried to do it before, but we then run into the politics of whether the person making the report is the PMC "representative" to the subproject.
Right - given that any PMC member can report issues, the PMC chair simply summarizes what has been happening over the last month for the board. If subproject <foo> had nothing interesting going on (i.e. all is well), he or she would either just say that, or omit it alltogether, focusing only on the issues that arose.
The fundamental disconnect is that all of the committers should be on the PMC, because all of the committers are the decision-makers for one or more of our various products.
100% true.
geir
-PROPOSITION (2)-
[snip]
Work regarding a specific subproject can continue to occur on that
subproject's DEV list. PMC members (aka committers) following that list
can vote on its releases and other day-to-day affairs. So long as the
minimum quorum is met, the vote is legal and binding.
So, we are trying to delegate power to subprojects? Er, but we can't now can
we.
No. We are instituting a "minimum threshold meritocracy" for each product. The PMC members/committers who are working on a product, and interested in its development, and the ones who make the decisions about that product. That's how it works now socially, and that's how it should work legally.
So who can vote? 'Following the list' is a very vague term. Presumably I can
simply subscribe to struts-dev and then vote, never having even used struts?
It also seems highly dubious to say that a vote is legal and binding if most
of the PMC never knew the vote occurred.
As it stands today, any of the current PMC members could do exactly that.
And, this is also how it works in the Commons today. If I want to chime in on a product and start committing, other volunteers are happy for the help.
If you did subscribe to the Struts list and took an interest in the product, I'm sure we'd welcome your commits. You're an Apache Committer, and I'm sure you've earned your stripe. Not by trying to do harm, but by trying to do good.
The value of administrative [vote]s on the DEV list are often overrated. A veto must have technical merit to be binding. Malicious vetos are not valid. And, as you know, when someone tried to enforce their own will over the will of the community, the ultimate result (sadly) was a suspension of write access.
Under proposition (1), the significant events occurring for each
subproject would be reported to the PMC list, for the review of the PMC
at-large.
So the PMC is reviewing events already happened, not actively managing. Er,
sounds like the relationship between the board and a PMC to me.
No, the committers to each subproject are committee members. Most Apache projects practice a minimum threshold meritocracy. We don't expect every committer to be involved in every decision, or cast votes or opinion outside their area of interest or expertise. If three committers/members vote +1, we're good to go.
The PMC was not meant to be a legislative body: it's suppose to be the core group, the decision-makers, the active managers, the committers.
PMC membership is voluntary. Anyone can resign from the PMC at any
time. If a volunteer would like to be a committer, but not a PMC member,
then each subproject can then decide whether to support separate
committer and PMC member roles or not.
I would suggest that there is nothing in this proposal that will cause the
board members to have any more faith in Jakarta than they do now. And thats
because it changes nothing of significance.
It changes everything. It turns Jakarta from a place that is supposedly governed by an "other wordly elite" to a place that practice "minimum threshold meritocracy" -- both socially and legally. Today our social order is out-of-synch with our legal status. This proposal legalizes what already happens in practice.
* It provides a forum where ALL the decision makers can discuss oversight (not just a chosen few).
AND,
* It puts reporting in the lap of the decision-makers for each product, which ensures it stays on the *decision-makers* radar, and is not pushed up to some body that cannot possible oversee our products.
-Ted.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Geir Magnusson Jr 203-247-1713(m) [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
