You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by J Aaron Farr <fa...@apache.org> on 2006/01/18 16:03:23 UTC
Mentor v Champion (WAS: Re: [VOTE] Changes to Incubator process(es))
I know the vote is essentially over, but the following thought
occurred to me this morning as I volunteered to support the OfBiz
proposal.
On 1/11/06, Davanum Srinivas <da...@gmail.com> wrote:
> [ ] - Any new proposal should have 3 ASF Members / Officers as mentors
> (without regard to affiliation)
Currently there's only one mentor, but there can be several champions.
The mentor has specific oversight responsibilities. The champions
are also expected to be involved but their specific role is to
demonstrate that the proposal has support within the ASF.
Did we really want to distribute the mentor responsibilities to three
people? Or was the intent to increase the required show of support?
My impression was that the intent was to show that there is a minimal
level of support among ASF members, not to dilute responsibility. In
this case, a more accurate process change would be to require (or
recommend) that proposals have three champions, not three mentors. If
so, we may also need to better define the role of champion.
--
jaaron
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: Incubator Roles revisited
Posted by "Noel J. Bergman" <no...@devtech.com>.
Bill Flood wrote:
> I thought it was a good question Leo.
Which one?
> I'm also wondering about the PPMC and the relationship to mentors,
> champions, etc.
The PPMC would be PMC members mentoring the project, plus others who are
voted on to help manage it. So your fisherman's committee analogy isn't too
bad.
> All good so far but I'd have expected that the incubator PMC would have
> asked the proposal authors to pair down the committer list to something
> manageable and form the PPMC from the group that offered the proposal
The PPMC started with Dims, James, et al, and their first order of business
is adding you and Ismael to the PPMC. At which point, I do expect that the
group of you will do precisely what you've suggested.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Incubator Roles revisited
Posted by Bill Flood <bi...@gmail.com>.
I thought it was a good question Leo.
I'm also wondering about the PPMC and the relationship to mentors,
champions, etc. I assumed the PPMC was the proverbial fisherman's committee
being taught how to manage the project of fishermen under Apache rules. For
Ode, Sybase asked James and Allen to mentor us and join the project. Dims
was also invited by James to mentor due to his level of energy :-).
Along came Ismael with Intalio's offering. All good so far but I'd have
expected that the incubator PMC would have asked the proposal authors to
pair down the committer list to something manageable and form the PPMC from
the group that offered the proposal (presumably James, Allen, me, and then
Ismael). At that point committer's would be added by the PPMC.
Another reading is that Karma has less to to with a project and is Apache
wide. That reading sounds like it could lead to problems for a podling.
I'll claim ignorance if my assumptions were off :-).
Bill
On 2/19/06, Leo Simons <ma...@leosimons.com> wrote:
>
> +1 to snipped stuff. You're on the right track.
>
> On Sat, Feb 18, 2006 at 07:15:23PM -0500, Noel J. Bergman wrote:
> > In addition, despite the "Mentor is a role undertaken by a permanent
> member
> > of the Apache Software Foundation" portion of the description, we have
> > previously elected non-Members to the Incubator PMC, and asked them to
> serve
> > as Mentors.
> >
> > Following from the idea that a Mentor must be a PMC member, do we have a
> > need to require them to be an ASF Member, as per the current document,
> or
> > can we more simply drive the rule from the fact that a Mentor must be a
> PMC
> > member, which is effectively how we have been doing it? Again, I get
> back
> > to saying that Mentors are PMC Members who are providing active
> oversight
> > and guidance. Does anyone see any effective difference, since PMC
> members
> > have equal votes, and the only binding ones?
>
> Not really.
>
> > Other than apmember karma,
> > which means access to private information and the ability to contribute
> on
> > restricted infrastructure, what would an ASF Member have that isn't also
> > vested in someone whom we have otherwise chosen to vote onto the
> Incubator
> > PMC?
>
> Codified level of trust.
>
> With ASF members, it seems we kind-of feel that its ok to have the PMC
> membership
> be "all but automatic" (these people are pretty much assumed to understand
> what
> direction incubation is going in, usually have prior experience on another
> PMC, or
> whatever it is that creates this 'trust' thing), whereas with other people
> it is
> not so clear. I think its fine to have non-asf-member mentors and
> non-asf-member
> PMC members, but I think that should be a concious vote from the existing
> PMC
> (like with other ASF PMCs) since that is another way to establish enough
> 'trust'.
> Eg the positive vote from the existing PMC here is this 'codified trust'.
>
> Make sense?
>
> (...)
>
> The implication here is that you don't get to be a mentor by adding your
> name to
> an incubation proposal unless
>
> a) you're an ASF member, and/or
> b) you're an incubator PMC member
>
> and we vote on this "people thing" seperately.
>
> Interestingly, these people things are invitation-only for things like
> committership and (pmc) membership, whereas with the incubator it seems
> we're on a
> road to invite-one-self kind of thing. Is this concerning? What is going
> on here?
>
> - LSD
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
RE: Incubator Roles revisited
Posted by "Noel J. Bergman" <no...@devtech.com>.
Leo Simons wrote:
> The implication here is that you don't get to be a mentor by adding
> your name to an incubation proposal unless
> a) you're an ASF member, and/or
> b) you're an incubator PMC member
Seems so. We would vote in public for the project, and in private for
mentors.
> these people things are invitation-only for things like committership
> and (pmc) membership, whereas with the incubator it seems we're on a
> road to invite-one-self kind of thing. Is this concerning?
Yes, in the sense that we want to avoid to potential embarrassment for
people who are not elected. Perhaps nominations should be kept to PMC
lists? Any other thoughts?
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Incubator Roles revisited
Posted by Leo Simons <ma...@leosimons.com>.
+1 to snipped stuff. You're on the right track.
On Sat, Feb 18, 2006 at 07:15:23PM -0500, Noel J. Bergman wrote:
> In addition, despite the "Mentor is a role undertaken by a permanent member
> of the Apache Software Foundation" portion of the description, we have
> previously elected non-Members to the Incubator PMC, and asked them to serve
> as Mentors.
>
> Following from the idea that a Mentor must be a PMC member, do we have a
> need to require them to be an ASF Member, as per the current document, or
> can we more simply drive the rule from the fact that a Mentor must be a PMC
> member, which is effectively how we have been doing it? Again, I get back
> to saying that Mentors are PMC Members who are providing active oversight
> and guidance. Does anyone see any effective difference, since PMC members
> have equal votes, and the only binding ones?
Not really.
> Other than apmember karma,
> which means access to private information and the ability to contribute on
> restricted infrastructure, what would an ASF Member have that isn't also
> vested in someone whom we have otherwise chosen to vote onto the Incubator
> PMC?
Codified level of trust.
With ASF members, it seems we kind-of feel that its ok to have the PMC membership
be "all but automatic" (these people are pretty much assumed to understand what
direction incubation is going in, usually have prior experience on another PMC, or
whatever it is that creates this 'trust' thing), whereas with other people it is
not so clear. I think its fine to have non-asf-member mentors and non-asf-member
PMC members, but I think that should be a concious vote from the existing PMC
(like with other ASF PMCs) since that is another way to establish enough 'trust'.
Eg the positive vote from the existing PMC here is this 'codified trust'.
Make sense?
(...)
The implication here is that you don't get to be a mentor by adding your name to
an incubation proposal unless
a) you're an ASF member, and/or
b) you're an incubator PMC member
and we vote on this "people thing" seperately.
Interestingly, these people things are invitation-only for things like
committership and (pmc) membership, whereas with the incubator it seems we're on a
road to invite-one-self kind of thing. Is this concerning? What is going on here?
- LSD
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: Incubator Roles revisited
Posted by "Noel J. Bergman" <no...@devtech.com>.
Berin Lautenbach wrote:
> I'd be happy for non-members who have met your criteria [with]
> one caveat - I'd prefer to have at least one member on each
> project to make sure there is someone in the podling that the
> membership as a whole have agreed is looking out for the wider
> interests of the ASF.
Is this expressing concern that we would elect someone to the PMC who is
more concerned about a specific podling, or a particular TLP, than the ASF?
I suppose that is a fair concern, but doesn't that indicate whom should be
elected to the PMC?
Interstingly, when you look over the bullet lists associated with the role
in http://incubator.apache.org/incubation/Roles_and_Responsibilities.html,
you may notice that we have enumerated important oversight issues, but only
one bullet even hints at the major responsibility of helping out with
community building. And none talk about technical contributions. I feel
that the former should be more explicit, but the latter is probably given as
much emphasis (or lack thereof) as it deserves.
Rich Bowen mentioned how he would like to be a Mentor, but doesn't know
Java. Mentors need the ability to maintain the required oversight, and
skill in community building. But I would rather have someone with those
qualifications than someone skilled in the technical domain. A Mentor
lacking the latter is likely to focus on community building, and helping the
community to make its own decisions -- perhaps using the Socratic method --
rather than inserting himself or herself into the midst of technical issues.
Helping the community to make collaborative decisions seems the right
approach to me, especially with more and more sophisticated projects.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Incubator Roles revisited
Posted by Berin Lautenbach <be...@wingsofhermes.org>.
Noel J. Bergman wrote:
<snip>
> As written, a Champion must be an ASF Member or Officer. Why? A Champion
> actually has no specific rights. Can anyone express a reason why the role
> should be restricted to a Member or Officer? One that comes to mind is as a
> filter, but realistically, anyone can ask a PMC to sponsor a project.
I don't think it's really relevant any more. The original idea was that
we wanted someone who was trusted within the ASF (thus Member or
Officer) to put forward the project within the organisation. Given the
increased oversite at the initiation stage, I think it's more than a
little redundant now.
<snip>
> Following from the idea that a Mentor must be a PMC member, do we have a
> need to require them to be an ASF Member, as per the current document, or
> can we more simply drive the rule from the fact that a Mentor must be a PMC
> member, which is effectively how we have been doing it? Again, I get back
> to saying that Mentors are PMC Members who are providing active oversight
> and guidance. Does anyone see any effective difference, since PMC members
> have equal votes, and the only binding ones? Other than apmember karma,
> which means access to private information and the ability to contribute on
> restricted infrastructure, what would an ASF Member have that isn't also
> vested in someone whom we have otherwise chosen to vote onto the Incubator
> PMC?
I always thought that using ASF Membership was more of a guide to say
"we know this person understands what is required within the ASF well
enough to be a sound mentor to this project".
So personally I'd be happy for non-members who have met your criteria
above with one caveat - I'd prefer to have at least one member on each
project to make sure there is someone in the podling that the membership
as a whole have agreed is looking out for the wider interests of the ASF.
Cheers,
Berin
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Incubator Roles revisited
Posted by "Noel J. Bergman" <no...@devtech.com>.
J Aaron Farr wrote:
> Davanum Srinivas wrote:
> > [ ] - Any new proposal should have 3 ASF Members / Officers as
> > mentors (without regard to affiliation)
> Currently there's only one mentor, but there can be several champions.
We should probably revisit
http://incubator.apache.org/incubation/Roles_and_Responsibilities.html and
remind ourselves of what it says, why, and what we really intend. And then
revise it as necessary to match the latter.
As per that document, the role of a Champion is to help bring the project
into the Incubator. As I read it, and in my own view, the Champion role
pretty much stops there (the last sentence of the description
notwithstanding, since we should really be focusing on the health of the
burgeoning community).
As written, a Champion must be an ASF Member or Officer. Why? A Champion
actually has no specific rights. Can anyone express a reason why the role
should be restricted to a Member or Officer? One that comes to mind is as a
filter, but realistically, anyone can ask a PMC to sponsor a project.
The Mentor role lists a large number of responsibilities, but they boil down
to:
[The] Mentor automatically becomes a member of the Incubator
PMC. A Mentor has specific responsibilities towards the
Incubator PMC
Turning this around, we have the comment I made earlier today:
a Mentor is an Incubator PMC member who is choosing to be
active and help in guiding the community
Without PMC membership, the Incubator specific notion of the Mentor role has
no real standing, which is why a Mentor must be ("becomes") a member of the
Incubator PMC. Mentors are PMC Members who are providing active oversight
and guidance.
We've established a couple of policies:
- All ASF Members who join an Incubator project as Mentor
and/or Committer shall be automatically entitled to
Incubator PMC Membership upon their own acceptance of
the responsibility, and subject to the normal notification
requirement to the Board.
- Any ASF Member may ask to participate on the Incubator PMC.
The Incubator PMC Chair shall submit the request to the
Board (cc to the PMC) to be acknowledged. At any time
between the Member's request and prior to the expiration
of the 72 hour period subsequent to the ACK, any current
Incubator PMC member may object to the automatic selection.
In the event of such objection, the selection will be
cancelled, and a formal vote will be called.
The second rule is recent, and renders the first redundant and unnecessary.
We have also relatively broad agreement, and should formally settle, on
having multiple (3+) Mentors per podling, in order to help ensure legal
oversight.
In addition, despite the "Mentor is a role undertaken by a permanent member
of the Apache Software Foundation" portion of the description, we have
previously elected non-Members to the Incubator PMC, and asked them to serve
as Mentors.
Following from the idea that a Mentor must be a PMC member, do we have a
need to require them to be an ASF Member, as per the current document, or
can we more simply drive the rule from the fact that a Mentor must be a PMC
member, which is effectively how we have been doing it? Again, I get back
to saying that Mentors are PMC Members who are providing active oversight
and guidance. Does anyone see any effective difference, since PMC members
have equal votes, and the only binding ones? Other than apmember karma,
which means access to private information and the ability to contribute on
restricted infrastructure, what would an ASF Member have that isn't also
vested in someone whom we have otherwise chosen to vote onto the Incubator
PMC?
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org