Jim Grisanzio wrote:
> Garrett D'Amore wrote:
>
>> The numbers speak for themselves.
>
> Numbers don't speak. They are interpreted and spun.
I'm not a marketing person, nor am I a lawyer, but I didn't think I
"spun" those numbers... Advocacy has 2x the number of CC grants of any
other group. That's a plain fact. It may have vastly more members as
well... based on numbers you cited it could be an order of magnitude
larger than all the other groups combined. That's beside the point, IMO.
>
>> What is the point of having so many CC's, if not to overwhelm the polls?
>
>
> You are assuming political motive where there is none. The point is to
> be involved. To have your voice heard. To participate. No one in
> Advocacy is thinking about "overwhelming" anyone at the polls.
I'm not presuming that there is actually any nefarious plot here. Only
stating that I'm not sure the balance is appropriate for the
organization as I see it. I'm sure there are other views.
If the only point is to have their voices heard within the confines of
the community, then ordinary contributor grants could work. As far as I
can tell, the only reason to have a CC versus a Contributor grant is the
additional voting privileges it confers in the larger audience. (Now,
some CGs also use CCs for voting internally, but I think that's a CG
internal matter.)
>
>
>> I'm not saying that this was the intention, but it certainly (to me
>> at least) can have that *appearance*.
>
>
> Using language like that may be good rhetoric, but it's harmful to the
> people in Advocacy who are just doing their jobs and want to be
> involved. Currently, they are a bit at the edge of the main community.
> That's changing now, though.
The ones at the "edge" are the ones I worry about... if they're not
involved in the main group, then why do they need voting rights in the
main group. This is no different, IMO, than the issue that Sun
employees who aren't involved in the OpenSolaris project are given
Grants. Neither situation is ideal. CC grants should, IMO, be granted
by the CGs to people who are expected to vote and represent the CG in
the larger main group. Its quite possible that CC grants are being
given for other good reasons, but maybe we (the project) need to
separate out CC grants from other forms of recognition. I don't know
what the right answer is... I'm just trying to get folks thinking about it.
I do like the ideas espoused by others, which is that perhaps we need to
split the engineering from non-engineering groups, to ensure that one
group can't unfairly dominate the other.
>
>
>> Part of the reason I'm concerned is that I've had a belief that we
>> were basically running as a meritocracy. I'd be a bit unhappy if
>> the core technical leadership of OpenSolaris (both in and out of Sun)
>> ultimately got steamrolled in any kind of election because the
>> Advocacy CG has several times the number of CCs as the CGs where the
>> actual "work" of developing (and not just coding, but also tools,
>> i18n, and docs) the product occurs.
>
>
> Has this happened before? No. In fact the *opposite* is true. Although
> Advocacy had more Core Contributors than any CG last election, it did
> *not* vote in any substantial numbers at all!
That suggests to me that the CC grants are not being well assigned. CC
grants should be given to contributors that care enough about
OpenSolaris to participate in elections. (This is one of the reasons
that in other CGs the CCs have to actively request or affirm the
grant.) I think being a CC should carry with it a responsibility.
Again, I think maybe the Advocacy group is using CC grants to fill an
"acknowledgement" role rather than a governance role. Perhaps it should
review internally whether some other form of acknowledgment that doesn't
carry the governance implications is more appropriate.
If enough Advocacy members didn't participate, they alone could
potentially harm the group just by ruining the turnout and thus
invalidating the election.
> Your speculation on this issue is baseless. Also, I can just as easily
> say this: I'm concerned that the ON engineers running OpenSolaris will
> steamroll the users since more engineers votes. In fact, it would be
> interesting to see what CG offered the most votes in the last
> election. I doubt it was Advocacy.
I'm not saying they have or will do that, only that they *can*. The
power is there. I'd rather no one group had that much control over the
project.
>
>
>
>> I'm not demanding anything, only *asking*, in the hopes that an
>> explanation would allay my concerns.
>
>
> Cite your evidence to support your concerns.
I have *concerns*, not accusations. My concern is that too much power
is given to one group. Whether it chooses to use that or not (and if it
does whether it is for the good or ill of the project) is entirely
beside the point. No one group should have that much power, IMO.
>
>> The defensive posture that you seem to be taking suggests to me that
>> maybe my concerns are justified. Are they?
>
>
> Cite your evidence to support your concerns.
My evidence is just the numbers and the defensive posture you've
taken. Again, I'm not *accusing* that anything bad has intentionally
happened, only looking for a statement, that "no it hasn't", and
pointing out that there is a possible problem lurking here.
The Advocacy Group is not on trial, despite your intensely defensive
posturing.
>
>> Well you won't tell me, and that's fine ... but in the absence of a
>> response I'm left to make my own conclusions.
>
>
> Your conclusion is baseless without any evidence.
That may be. But you've not really done anything to help me see
otherwise. I can only conclude based on the facts I have at hand. The
facts I have at hand are that grants in the other groups are hard to
come by, backed by real contribution to the project (usually in the form
of code), and the Advocacy Group has a huge number grants (likely with
little "overlap" with other groups), and with CCs who contribute in ways
that I apparently can't understand.
I was seeking enlightenment, not punishment. I don't think you
understand me here.
>
>> (Whether those conclusions mean anything in the broader scope is an
>> entirely different matter. Right now, with the membership the way
>> it is, it would be very hard for any group to push the project in a
>> direction that the Advocacy group didn't like.... )
>
>
> All evidence thus far on the OpenSolaris project suggests the exact
> opposite. The Advocacy CG is largely made up of User Groups, and most
> of them are not that active and only loosely connected to
> opensolaris.org.
If that is so, then why do those folks have CC grants? Do they need to
be acknowledged in the larger group for some reason? Do they even want
that acknowledgment? To what end? Again, my concerns are strictly
governance related.
>
> Also, the OpenSolaris Membership itself is not that active in terms of
> governance. Most CGs are not actively led, and the Members have
> clearly decide to not lead the community. That's their choice, of
> course, but it goes a long way to supporting the notion that the
> OpenSolaris community has a leadership problem. I don't entirely
> believe that, but I'm leaning in that direction lately.
Most of the groups are busy doing real work, and a hands-off approach
has worked well for the communities in which I'm involved (CC in three
of them, participant in several others). I do support the notion that
when it comes time to vote, CCs have an obligation to vote, and apart
from the intentional abstainer (which is itself an active political
decision) should be voting. Just failing to vote out of lack of
awareness, or lack of interest, is not a good excuse, IMO.
>
>
>> Btw, I posted originally with the view that maybe, just maybe, there
>> really is a good reason for there to be so many CCs in that group. I
>> was hoping for a clarification, and frankly wasn't expecting the
>> defensive posture.
>>
>>>> Are the CC's in that group *really* core contributors, or are they
>>>> just people who occasionally pipe up on mailing lists?
>>> Please define "real" and point to the approved definition of "real"
>>> and why you feel you can make that assessment for another CG.
I don't feel I can make that assessment. I'm asking the Advocacy Group
to think about it. I do think "contribution" as used to determine CC
status needs to be more than casual alignment with a group. The exact
details of how a CG decides that is of course up to the group, but I
hope that all groups have CCs who have done something substantive to
improve the project, and aren't just bystanders, or people that have
installed OpenSolaris and posted once on a mailing list. Again, its
not a clear boundary, but IMO CC status should be something to be earned
non-trivially... folks who get it should have demonstrated that they
have put a significant amount of effort into the project.
>>>
>>>> I've lived under the belief that core contributors should be people
>>>> who are very actively helping the community achieve its goals.
>>> We are. And we have demonstrated that. Clearly.
>>
>> It isn't clear to me that all 48 core contributors have contributed
>> so heavily to warrant a CC grant.
>
>
> I'm not sure how that's your business.
It isn't, really, except that when the Advocacy Group has twice the
political clout of any other group (and possibly more than all others
combined), then it becomes a governance issue.
I also am somewhat concerned as a CC myself.... I had to do real work to
earn that CC status. In my case it has been an ongoing and substantive
involvement with the groups that have given me CC status... in the form
of code, continued help/discussion on lists, active code review, giving
presos, ARC participation, etc. And much of this goes way beyond my day
job as a Sun employee. If anyone can trivially obtain a CC grant in the
Advocacy group, it certainly devalues the meaning of what I (and others)
have worked hard to earn.
>
>> It isn't clear to me, quite frankly, as an external viewer, that
>> there is that much work that the Advocacy group has done.
>
>
> Did you read my last mail where I listed some of work we are doing?
I did, but only afterwards. Yes, I understand that there is a lot of
activity. I don't know how many people are involved at the CC level
though. And again, I think maybe Advocacy is using CC grants for
purposes other than governance. I don't know for sure, of course. But
I'm asking the Advocacy Group to evaluate this for itself. And please
do think about it, rather than just blowing the concerns I've raised off.
>
>
>> But then again, I admit fully I don't have a full view into how the
>> Advocacy CG identifies CCs. I *hope* that CC grants require some
>> active form of contributorship and not just participation. (Just
>> because I participate in e-mail discussions on the crypto lists, for
>> example, doesn't mean I should be a core-contributor for that CG.)
>
> I wouldn't know. I'm not involved in that CG. How could I possibly
> comment with any credibility whatsoever?
I'm just talking generally. Random participation in discussions
shouldn't, IMO, qualify anyone for CC status. If it did, I'd have more
like 10 CC grants. Instead I think real contribution, in the form of
action, should be required. (Action meaning writing code or content,
moderating lists, organizing large events, etc.) And I think CC status
should be given to those that have demonstrated by their deeds an active
interest in the CG, and also understand the governance duties that come
with it.
>
>>>> In most other communities this would probably be achieved by actual
>>>> code contributions. In the Advocacy group, I'd guess this would be
>>>> work like writing PR content, managing web forums, or maybe running
>>>> User Groups.
>>>>
>>> You "guess"? That sounds pejorative. I hope you didn't mean it that
>>> way. What is "writing PR content, managing web forums, or maybe
>>> running User Groups" mean anyway?
>>
>> I don't know... I'm not a marketing person.
>
>
> Marketing is only part of Advocacy. A small part, actually.
Okay, well then I don't really understand anything about Advocacy at all.
>
>
>> But to me advocacy suggests the sort of things that that marketing
>> folks might do. And I'm not trying to suggest that those things
>> aren't important. But I don't think they are so much vastly more
>> important that the group should have 2-4x the clout of the other
>> large groups in OpenSolaris.
>
> The group has more Core Contributors because it has more members. It's
> that simple. We are drawing form a larger pool of people.
I understand that. And maybe there are even a larger number of people
in that group that participate with as much enthusiasm and action as
warrants the CC status. That is the question I was asking to have
answered originally... At some level I still wonder if this concern is
addressed.
The other thing is that *users* and *system admins* and even
*developers* who are just using OpenSolaris, rather than contributing
towards it (and contribution means much more than just coding, I do get
that!) are not Core Contributors, or even just Contributors. Just
because I installed OpenSolaris on 500 workstations, and am a huge fan
of the project, doesn't qualify me as a Core Contributor, IMO. I
should be able to stand up and list my contributions to the project to
justify my CC grant. (In my case, I had to do that for each of my 3 CC
grants.)
>
>
>> I'm also not certain that setting up a blog or some other kind of
>> posting forum really should grant "core contributor".
>
>
> Why are you so pejorative toward Advocacy?
I'm not. But I don't think you understand my concerns.
>
>> However, some folks certainly have done a lot of work (Ben Rockwood
>> comes to mind, for example. As Dennis Clarke, etc.) to help spread
>> adoption and evangelize OpenSolaris, and they deserve to be
>> recognized. Are there 48 such individuals? I'm not sure.
>
>
> Well, it's ok not to be sure. You are not involved, so naturally you
> can't be sure.
>
>
>> Put another way, if I set up a small "Temcula OpenSolaris User's
>> Group", would I deserve a core contributor grant from the Advocacy
>> Group? I hope not. Now if I helped coordinate several large user
>> groups, did a bunch of associated mail list moderation, etc, then
>> yeah, I can see it would make more sense. I'm just hoping that there
>> is sanity in the Advocacy Group. (I'm not saying that there isn't
>> such sanity...)
>
>
> Sanity? You say you're not saying it yet you said it. First we were
> out there gerrymandering to fix the election and now we are insane. Ok.
I never said gerrymandering was occuring, only that it might have the
*appearance* of occurring. Anytime any one group has that much
political clout, and can arbitrarily give itself more (by assigning more
CC grants) it certainly calls into question whether some kind of
controls on political power are appropriate.
And, I specifically said I didn't claim there wasn't sanity... only
asking the Advocacy Group to reassure me that it was indeed there...
that they weren't giving out CC grants willy-nilly. And you still have
not answered that question.
>
>>> Actually, Advocacy sponsors the BeleniX list, which is directly
>>> about coding (within the context of a distro that is largely run by
>>> a thriving user group that has other very technical contributors).
>>
>> The BeleniX distro should have its own CG, IMO.
>>
>>> It also sponsors the trademarks list, which is about
>>> marketing/branding and that is a specialized activity. It also
>>> sponsors the main news page on opensolaris.org, the only CG to
>>> maintain one of the main site pages. It also sponsors a new
>>> mentoring project, which is about getting new people involved in
>>> coding. It also sponsors 56 user group projects, which involve
>>> users, sys admins, students, professors, and engineers of all types
>>> in a dozen or more countries.
>>
>> Users, sysadmins, and students are not "core contributors".
>
>
> Wrong. The Constitution says otherwise.
Pardon me. I think maybe I should have more specifically said that
Users, Sysadmins, and Students don't qualify as CCs on that basis
*alone*. The CC grant has to be earned due to contribution. If those
individuals are contributing at an appropriate level, then of course a
CC grant is appropriate.
>
>> Someone running/coordinating a mentoring project for those people
>> might deserve a CC grant though. (I guess its kind of a gray area
>> for whether or not actual mentors would get one.)
>>
>>> Advocacy CG members have been busy presenting at conferences and
>>> other events around the world for quite some time now, and those
>>> people have generated quite a lot of content -- the vast majority of
>>> which is technical. Advocacy CG members have also been meeting with
>>> press and analysts around the world to deliver technical and
>>> community content. The Japanese OpenSolaris Community runs
>>> installfests led by engineers. The Bangalore Community maintains
>>> BeleniX and puts on bug-fixing sessions led by engineers via the
>>> request-sponsor program. The German Community ran the first and only
>>> OpenSolaris developer conference (Berlin first, Prague next) and the
>>> papers presented were all technical except for one (mine). The China
>>> Community runs coding contests and engages students at universities
>>> to the tune of well over 100,000 people at this point. Other
>>> communities are contributing technical translations in collaboration
>>> with the portals and the Internationalization CG. And it goes on and
>>> on. I can't even keep track anymore. Do you consider any of these
>>> items to be second-class contributions?
>>
>> No. But I don't consider the *sum total* of them to be more
>> important than the other work that takes place elsewhere.
>
>
> Of course it's not. Who said it was?
The number of CC's effectively makes it politically so. Again, my issue
is strictly governance.
>
>
>> Not everyone who writes code for ON is an ON CC. Likewise, I don't
>> think everyone who participates in the Advocacy group should have a
>> CC grant.
>
>
> There are more than 4,000 people in Advocacy. Obviously, not all are
> Core Contributors.
OK.
>
>
>> Out of curiosity, does anyone use Contributor (not Core) grants? It
>> seems like there is an opportunity to use them here.
>
> Yes.
>
>> Also, some of those activities belong in other groups, IMO. For
>> example, there are already communities for Internationalization,
>> Approachability, etc.
>
>
> Many people participate in multiple CGs. That's a good thing. That
> will continue and grow, actually.
>
>
>>>> Perhaps we should consider a constitutional amendment to limit the
>>>> number of CC's that a community can nominate (set at some number
>>>> high enough to reward those groups with greater participation, but
>>>> not so high that any one or two groups can dominate an election? 10?)
>>> Is this designed to keep Advocacy down? If so, please try that. It
>>> will do wonders to help me motivate the Advocacy electorate. Why are
>>> you suggesting that we punish people who are simply doing good work
>>> under the rules we all voted for? Why are you suggesting that we
>>> disenfranchise the *largest* segment of the meta OpenSolaris community?
>>
>> I'm not suggesting that we punish them. I'm suggesting that Advocacy
>> is itself not 3-4 times more important than any other activity
>> involving OpenSolaris.
>
> Why do you keep saying that?
Because as it stands, that group has much more potential political power
of any other CG. (Well 2-3x anyway.) Whether they exercise it is
besides the point.
>
>
>> Advocacy by itself doesn't write documentation, perform translations,
>> or develop new features.
>
>
> Advocacy people have been involved in all those activities in one way
> or another.
That's fine. And those people should get grants from the appropriate
groups then. I hope Advocacy doesn't give me a grant just because I
wrote a networking driver.... that grant should come from a more
appropriate CG (in this case Networking or Device Drivers).
>
>> Your "threat" really does help me make my point.
>
> Not a threat. A promise.
Well, if Advocacy is going to hand out CC grants, then it should make
sure that the CCs it has are involved enough to actually vote. If they
aren't, well then what is the purpose of the CC grant?
>
>> Put another way. If Advocacy simply decides to boycott the
>> elections, then it maybe mathematically impossible to get a quorum.
>> I don't think any group should have that kind of power.
>
>
> To the contrary, Advocacy has very little power and that's very easily
> documented. It didn't vote in large numbers last time around and it
> shows little interest thus far this time around as well. I'll have to
> work harder.
Maybe CC grants are too easily given out then? CC grants should come
from parties in the group who *want* to participate in elections.
Anyway, my concern is that the power is there. Whether it is wielded or
not is not relevant to my concern.
>
>
>>>> Or perhaps we should just abandon the distinction between CC's and
>>>> ordinary Contributors, and open the vote to the great unwashed
>>>> masses? I'm not sure... but I am wondering if the bar set to
>>>> become a CC in some groups (e.g. Networking or ON) is set
>>>> substantially higher than in others?
>>>>
>>> I'm all for making voting easier and eliminating the distinction
>>> between "Core" and "Contributor". It's confusing. And it is keeping
>>> the voting population of Advocacy artificially low.
>>>
>>> The bar is not higher for Networking or ON. It's just different. We
>>> have to allow for the notion that OpenSolaris is more than ON.
>>> Actually, it doesn't matter if we allow for it or not. It has
>>> already happened. Also, we have to realize that over time, ON will
>>> be a small percentage of the entire OpenSolaris community, but that
>>> doesn't mean ON loses any influence of ON's activities. Advocacy
>>> will never attempt to run ON. We only want to run our own CG so we
>>> can help OpenSolaris grow. We trust that ON can run itself. We
>>> expect it, actually.
>>
>> But in the context of *OpenSolaris* (not ON nor Advocacy), the
>> Advocacy group inherits though its CC grants such an overwhelming
>> vote that the leadership in other equally important communities can
>> trivially be made politically irrelevant. That isn't the kind of
>> meritocracy in which I want to participate.
>>
>>>> The other wrinkle in all this is that some communities have
>>>> considerable overlap in their CC membership. E.g. many of the ON
>>>> CC's are probably also CC's in other groups. It would be an
>>>> interesting data point to measure "voting loyalty" for each CG,
>>>> where each CC gets one vote, which is divided equally amongst all
>>>> the CGs in which they participate. CCs that belong only to one CG
>>>> contribute 1 to that CG. CCs that belong to two contribute 0.5 to
>>>> each of those two CGs, and so forth.
>>>>
>>>> The resulting graph may yield some surprising data about how fairly
>>>> (or otherwise) weighted the election is likely to be.
>>>>
>>>> Of course, I'm operating here under the premise that we actually
>>>> *want* all of the CGs to have a roughly equal say in important
>>>> group-wide matters... that may itself be an entirely false premise.
>>>>
>>> They should not be "roughly" equal. Just equal.
>>
>> If the community groups are equal to one another, then no one
>> community group should have an unfair advantage in the voting. A
>> group with several times the number of voters of its nearest
>> community has the ability to totally overwhelm/disenfranchise any of
>> the other communities it owns.
>
>
> Perhaps we should limit all CGs to, say, 100 people. Then each can
> choose CG's from the same pool. Then everyone will be the same.
Huh? I don't understand this. Did you mean CC's? I don't understand
your statement/suggestion.
>
>
>> In US history, there are several problems and different views that
>> have led to things like the electoral college, the bicameral
>> legislation, etc. Maybe what we need here is some kind of bicameral
>> legislative body. I dunno. Right now I don't like the idea that
>> Advocacy group could pretty much take over the entire organization
>> even if all the other CCs banded together to fight it.
>
> You have Advocacy taking over the entire community and you still can't
> cite any evidence to suggest that Advocacy has any power whatsoever.
The facts are in the numbers. Just because the Advocacy group is
ineffectual is totally beside the point!
> In fact, I've given evidence to the contrary. When it votes, it will
> have power, but it can not express that power over other CGs and
> certainly not over core engineering issues.
Actually it can.
Imagine if it decides to propose an Amendment to the constitution
requiring any new feature to be presented to at least two User Groups
before it integrated. It has the voting power right now to push such
a crummy rule through. And that would have engineering impact. Again,
the ability is there, regardless of whether it would ever so be used or
not.
By the way, I'd far rather see the Advocacy have 10-20 effective
participants (folks who vote actively), then 40-50 CCs of whom only
10-20% vote. I suggest that the Advocacy Group may want to reexamine
how it gives out CC grants... they should be given to people who are
committed enough to the project to stand up and be counted.
I also like the suggestion to break the group into two parts with
separate governance, one part being the folks who are working on the
engineering portion, and one part being the folks who make up
communities like the advocacy group. The demographics of it suggests
that it may not be possible to remain equitable for both sides otherwise.
-- Garrett
