Allow AB to choose its own chair #223
Comments
|
The link is for the chairing of the AC meeting, not the chair of the AB. I think you meant https://www.w3.org/2018/Process-20180201/#ABParticipation? There's no requirement to have the CEO chair the AB - in fact, it was only as I joined the AB that that became true (prior, the appointed chair was Steve Zilles). I agree that it's probably sub-optimal that the CEO is the chair - although I think it's very useful and important that the CEO participate in the board (which isn't explicit in the Process; the Chair could allow them, but it's not a "right" of the CEO). I'm not sure I'm supportive of the AB selecting the chair; there are a lot of term length/continuity tradeoffs. I might suggest the better solution would be that the Team appoints a chair that is not the CEO, but the CEO is invited. |
|
I recommend allowing for co-chairs of the AB.
|
|
It would be good to explicitly enabling co-chairing, yes, as that's become the norm in WGs. |
|
I support the change and the suggested addition of co-chairs. |
+1 to this suggestion. |
By "the change", you mean specifically the AB selecting its own chair? |
Yep, fixed. :)
I strongly prefer the AB selecting its own chair, since it's in the interests of the AB to choose one that best helps it make progress on the issues. It's not clear to me what motivation the Team has in that respect. Certainly the Team can make recommendations. But the members of the AB have the most motivation to choose a good chair to make the best use of their time and effort, and will not feel the need to make that choice e.g. out of deference to upper management (which are the people the AB is supposed to be overseeing). As for term length/continuity, I'll observe that we at W3C tend to default to continuity in the choice of chairs, so while the AB could change chairs after every election, it seems quite unlikely that they will. :) And since as proposed, the AB can choose a chair who is not a current member of the AB, if the chair is not re-elected but the AB finds them competent and helpful, it can choose that person to chair the next session anyway. |
|
Made PR #225 corresponding to this proposal.
|
frivoal
added
Needs Review
Agenda+
|
I disagree with making this change at this point in time. First some history. When I joined W3C, my instincts were exactly the same as Fantasai's. Namely, that it would not be a feature for the CEO (full disclosure - that's me) to be the Chair. Many people urged me to be a Chair, but I declined. Three years later, pressures increased for me to take the Chairing role. I was told that we would get more work done, it would be more directed to the needs of the consortium, and it would be more impactful - if I chaired. So I agreed. I have now chaired for many years. My self-assessment is that in these years, we have gotten more done. So my reason for not wanting to change the process now is that I would like to first understand what problem we are solving; and whether we have tried to solve the problem in the context of the current setup. In the description of this issue, I see a vague notion that some people believe that the AB would be more effective with a different chair. I do not see any specific example of a problem. I do not want to replace an effective system with an ineffective system. But, as I said at the beginning, my initial instincts were the same as fantasai's. So if there were a problem that this would be solving, I'm open to the discussion. But we should discuss the problem and its potential solutions first. |
|
The problems I see that having an elected chair might solve are:
|
|
@michaelchampion I think all of your points are reasonable points in theory. In theory, the CEO controlling the agenda of the AB is unhelpful. But I'm asking about problems in practice. For example, the way that I address the issue of "controlling the agenda" is by sending out agendas well in advance; asking for comments; and making sure there is time for everyone's agenda items. I've done that for topics that were important for the Team, less important for the Team, and even uncomfortable for the Team. |
|
@jeffjaffe |
|
I also think that we should consider a co-chair. Most W3C groups have 2 chairs now, and it seems to benefit the chair and the groups. It eases the burden on the chair, allows for time away (vacation or otherwise), and can allow co-chairs to split responsibilities (formally or informally). |
|
Agree with @michaelchampion and @TzviyaSiegman. I think that you, @jeffjaffe, should be free of the need to remain an impartial agent of the agenda so that you can be an active advocate of W3M's interests. And I think the AB should have a chair who will be able to actively manage the agenda in support of the interests of the AB at large and to more actively drive discussions to conclusion, so that the AB can make swifter and more definitive progress on the issues. You are more hampered in driving issues to conclusion than a more independent chair would be, and less able to serve as a neutral switchboard since you are one of the key inputs. We want you be a strong advocate, and we want the chair to be a strong creator of neutral, but effective discussion space.* I don't doubt that the AB has been more productive since you took on the role of chair. It means the previous chair was not a very effective chair, and you are a more effective one. But I think the AB's workload becomes more and more critical, and we need to do the best we can... which includes finding the most effective chair for the AB that we can. Opening up the chairing role to finding the best chair, rather than pegging it to the same person as the CEO, broadens the possibilities to get there, and I think removes some of the self-cancelling effects of having both roles be occupied by the same person. * This is in fact one of the reasons why I am not chair of the CSSWG. And personally, I find it very convenient that Alan and Rossen are effective chairs who are not me. :) |
|
Good food for thought. I actual don't think that the CEO's role on the AB should be an active advocate of W3M's interests. The CEO's role on the AB, imho, should be - like all other AB Members - wearing the AB hat while sitting on the AB. That is, representing what he/she believes are the best interests of the web and the consortium. At the AC meeting, I was asked the question whether the WHATWG partnership might cause us to lose members. I gave several answers; one of which is that our job was not to Lead the W3C to its Full Potential, but to Lead the Web to its Full Potential. If leading the Web is suboptimal for the W3C - so be it. A recent example of not optimizing for the team. That's not to say that W3M's voice shouldn't be at the AB. I often express W3M's opinions on certain topics; recently W3M's views about a NomCom. (Correspondingly, at W3M, I also often express AB opinions as well.) I agree that it is a challenge to both have opinions and also be the neutral creator of discussion space. So, certainly food for thought. |
|
I agree that co-chairs is something we should employ in the AB. I also agree with @fantasai’s statements below.
Natasha Rooney | Technologist, Web and Internet, W3C & IETF | GSMA | nrooney@gsma.com<mailto:nrooney@gsma.com> | +44 (0) 7730 219 765<tel:+44%207730%C2%A0219%20765> | @thisNatasha | Skype: nrooney@gsm.org<mailto:nrooney@gsm.org>
On Nov 6, 2018, at 22:19, fantasai <notifications@github.com<mailto:notifications@github.com>> wrote:
Agree with @michaelchampion<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fmichaelchampion&data=02%7C01%7Cnrooney%40gsma.com%7C1e2e07489d5a4ccc128608d64435eadc%7C72a4ff82fec3469daafbac8276216699%7C0%7C0%7C636771395702195232&sdata=UEQVu2oSX%2BVmHQ2cwCMG5KYZFWFRFCqhaipH%2BrcQKz4%3D&reserved=0> and @TzviyaSiegman<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FTzviyaSiegman&data=02%7C01%7Cnrooney%40gsma.com%7C1e2e07489d5a4ccc128608d64435eadc%7C72a4ff82fec3469daafbac8276216699%7C0%7C0%7C636771395702195232&sdata=DZzmi%2BxjHFaLhiZMLiJuvzuzRXmZHUJKJ7EZop7UBIA%3D&reserved=0>.
I think that you, @jeffjaffe<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fjeffjaffe&data=02%7C01%7Cnrooney%40gsma.com%7C1e2e07489d5a4ccc128608d64435eadc%7C72a4ff82fec3469daafbac8276216699%7C0%7C0%7C636771395702205242&sdata=L11t6kIP24wiwFDzslkFEnNJ991ZuK7slNXTr9mrros%3D&reserved=0>, should be free of the need to remain an impartial agent of the agenda so that you can be an active advocate of W3M's interests. And I think the AB should have a chair who will be able to actively manage the agenda in support of the interests of the AB at large and to more actively drive discussions to conclusion, so that the AB can make swifter and more definitive progress on the issues. You are more hampered in driving issues to conclusion than a more independent chair would be, and less able to serve as a neutral switchboard since you are one of the key inputs. We want you be a strong advocate, and we want the chair to be a strong creator of neutral, but effective discussion space.*
I don't doubt that the AB has been more productive since you took on the role of chair. It means the previous chair was not a very effective chair, and you are a more effective one. But I think the AB's workload becomes more and more critical, and we need to do the best we can... which includes finding the most effective chair for the AB that we can. Opening up the chairing role to finding the best chair, rather than pegging it to the same person as the CEO, broadens the possibilities to get there, and I think removes some of the self-cancelling effects of having both roles be occupied by the same person.
* This is in fact one of the reasons why I am not chair of the CSSWG. And personally, I find it very convenient that Alan and Rossen are effective chairs who are not me. :)
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fw3c%2Fw3process%2Fissues%2F223%23issuecomment-436429831&data=02%7C01%7Cnrooney%40gsma.com%7C1e2e07489d5a4ccc128608d64435eadc%7C72a4ff82fec3469daafbac8276216699%7C0%7C0%7C636771395702205242&sdata=c4cz0qcOrn8B94nriXmqT%2Fdct5Cie0UZm0MirLEE%2Fho%3D&reserved=0>, or mute the thread<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAB1RxlWwz20CrBbxeucGn9Sfi-9P7saSks5usgrvgaJpZM4X4iPh&data=02%7C01%7Cnrooney%40gsma.com%7C1e2e07489d5a4ccc128608d64435eadc%7C72a4ff82fec3469daafbac8276216699%7C0%7C0%7C636771395702215247&sdata=zcn6u1PGS3xQvOb7ovTlTmGIhSmtDZ4qoHVBZAhVxk4%3D&reserved=0>.
.
|
|
I think this is a worthy discussion of a point of governance principle, but I am concerned that it's not going to get enough consideration if we try to do something in Process2019. |
|
@dwsinger I disagree, I think if the AB wants to maintain Jeff Jaffe as the sole chair of the AB, certainly it can do so with this change, however by not making the change we constrain what the AB would be able to do for the next year. |
|
Formally insisting the AB elect its chair, and changing the existing chair, are entirely orthogonal. While I support a change, I agree with @dwsinger that we don't need to predicate Process 2019 on making it, and I think there are aspects that need to be clarified. In particular, having thought some more, I would suggest that we make it so W3C appoints the chair, on the recommendation of the AB, for a 3-year term to begin when nominations for AB election open. I believe that neither the CEO nor serving AB members should be chair, for essentially the same reasons. This balances stability of chairing, giving the AB a reasonable lead time to decide what to do, and clarity before people nominate for election. |
|
@fantasai Currently there's nothing that prevents the AB member from asking for a chair change. Formalizing it in the process2019 or not does not seem critical. That said, given that WGs/IGs have a 'non-team chair' practice (not mentioned the process either), it should be the same for the AB. |
If W3C should always follow the recommendation of the AB, then there's no need for an intermediary, and the AB could pick the chair directly. Since that's not what you're asking for, presumably you have some situations in mind where the W3C should not follow the recommendation of the AB. Can you elaborate?
That seems suboptimal: it allows for an AB on its way out to impose a chair to the incoming one. When they're mostly the same, that's alright, but if a large number of members got ousted and replaced by new ones, having the new ones be stuck with the old ones' choices seems undesirable.
Even if it is only advisory, the AB supervises what the team and CEO do. This is not true in the other direction. Some individual AB members may be too strongly partisan or opinionated to be effective brokers of consensus, but the inherent conflict of interest isn't the same. The AB isn't an actual board, but it bears some similarities to one. It is often argued that the CEO shouldn't be the chair of the board. I've never heard the argument that the chair of the board shouldn't be a board member. |
A 3-year term provides for stability over freedom for each new AB. It also makes the commitment to chair a clearer one, less subject to the vagaries of an individual election, which I think will be important in attracting a good chair rather than settling for one whose primary virtue is being available to do the job. The outgoing AB are probably not the best AB we could have, but I they have more experience of how the AB works and what it does, and while they may bias for poor existing practice I think the 10 months of being on the AB provides insight that enables them to make a good choice. Personally, I strongly value institutional memory, which is held by people, and maintained by not turning over all the apple-carts at once. |
Chairing is a distraction from participation and vice versa. That is the same principle on which the CEO should not be chair. Effectively the chair is a member of the board - just not one of the 9 elected members whose primary purpose in being there is to represent the members, and who should be unambiguously the people whose advice is being given. In the pathological case, this also stops a member of the AB being pushed to chair as a means of muting them. (Which is incidentally how I cam down against b locking a chair and an AB emmber from the same organisation). |
Actually, I doubt the team will ignore the recommendation of the AB. The primary rationale is a formalism, because only every 3rd AB recommends a chair under my proposal. There is a tension in the arrangement. If the AB picks a chair the Team (or CEO at least) simply cannot work with, there is the possibility to ignore them. Equally, if the team does ignore AB recommendations, it is a fair bet they will get a lot of pressure to do that again, so it is a precedent to avoid if at all possible... |
|
The current Process does not allow both the CEO and a chair who is not the CEO to be on the AB, though, so it still needs fixing. Minimal changes here would be: diff --git a/index.bs b/index.bs
index 15020b4..69ee86b 100644
--- a/index.bs
+++ b/index.bs
@@ -580,9 +580,8 @@ Advisory Board (AB)</h3>
<h4 id="ABParticipation">
Advisory Board Participation</h4>
- The [=Advisory Board=] consists of nine to eleven elected participants and a Chair.
- The [=Team=] appoints the Chair of the <a href="#AB">Advisory Board</a>,
- who is generally the CEO.
+ The [=Advisory Board=] consists of nine to eleven elected participants and the CEO.
+ The [=Team=] appoints the Chair(s) of the <a href="#AB">Advisory Board</a>.
The team also appoints a <a href="https://www.w3.org/Guide/staff-contact">Team Contact</a> for the [=AB=],
as described in <a href="#ReqsAllGroups"></a>.This makes the CEO ex officio part of the AB and allows the Team to independently choose a chair (or co-chairs). This allows the AB to experiment without having to bend the rules of the Process. |
|
The AB can invite people to meetings. For a decade or so it invited Steve Zilles, with the explicit request to chair for them. We have invited a scribe for about 2 decades. If the appointed chair is not the CEO, the AB can still invite the CEO (that is my marginally preferred model for how it happens, but I don't think it is a very important issue). If we make a change I agree we should reflect it in the process, but I think we already have sufficient latitude to experiment until then. |
|
+1 that we don't need to "fix" the process to allow a non-CEO chair and a CEO chair to both attend the AB meeting. |
|
I think @fantasai is heading in the right direction on this, but agree with @chaals' comment that the AB has always invited participation from other people. Proposed variation of Fantasai's original and more recently amended text (sorry for the lack of correct markup): "The Advisory Board consists of 9 to 11 participants elected by the Advisory Committee. The Team appoints the Chair(s) and Team Contact for the Advisory Board. The appointment of the Chair(s) MUST be consistent with 3.1.1 Conflict of Interest Policy. The Advisory Board MAY invite other people to participate in Advisory Board meetings at its discretion." |
|
If we redefine the AB without the chair as an ex-officio (in principle voting) member, I think we should clearly note that the CEO will generally be invited to attend AB meetings and should expect to be available for all meetings. (@LJWatson's proposed text seems to imply such a decision and @fantasai's doesn't, but the rest of @LJWatson's text are improvements I like that aren't contingent on any such decision). I don't have precise wording for this because we have not taken any such decision. |
dwsinger
added
the
Agenda+
|
OK, so depending on whether the Chair(s) are granted voting rights even if not elected members of the AB., we have the following proposed wordings: Proposal A: “The Advisory Board consists of 9 to 11 participants elected by the Advisory Committee. The Team appoints the Chair(s) and a Team Contact for the Advisory Board as described in [5.1]. The appointment of the Chair(s) MUST be consistent with 3.1.1 Conflict of Interest Policy. The Advisory Board MAY invite other people to participate in Advisory Board meetings at its discretion, and usually invites the CEO.” Proposal B: “The Advisory Board consists of its 9 to 11 participants elected by the Advisory Committee and its Chair(s). The Team appoints the Chair(s) and a Team Contact for the Advisory Board as described in [5.1]. The appointment of the Chair(s) MUST be consistent with 3.1.1 Conflict of Interest Policy. The Advisory Board MAY invite other people to participate in Advisory Board meetings at its discretion, and usually invites the CEO.” |
|
I don't think the fix is urgent, as the current wording does let the team pick, and the team can pick whoever, and as we can indeed invite the CEO should they not be chair. But I do think making the CEO a member out of right, rather than someone we might want to invite if the chair was someone else, is the right thing to do. An AB that wouldn't invite the CEO to its meetings would be pretty broken, so I see no reason to allow that. @fantasai's proposed edit in #223 (comment) does fix that, while not changing the practice that the Team picks the chair, so I'd support that. If we want to wait so see if we can say something more specific as a result of the experiments mentioned in #223 (comment), that's ok with me too, but @fantasai 's suggestion still looks correct, and does not in contradict that. |
|
@fantasai your proposed wording above captures most of it, but right now the practice is that the CEO is a member (something that effectively changed when Jeff took up chairing - before that the longstanding practice was that the CEO was actually just an observer, but attended pretty much every meeting). |
|
@fantasai I think Proposal B needs tweaking to allow for the Chair to be either a member of the AB or an invited non-member. Right now it looks like the Chair is always an invited non-member. |
|
@nigelmegitt said
That's my preference - I think the chair should not be an AB member for the same reason I think it should not be the CEO. Again, we haven't made any decision here, and the AB is in a holding pattern while it experiments. |
dwsinger
removed
the
Agenda+
|
(I cannot remember why I added the label) |
frivoal
removed
the
DoC
|
Note: The AB provided feedback and wording is being drafted in #514 |
|
effectively addressed in p2021 |
dwsinger
added
Closed: Accepted
https://www.w3.org/2018/Process-20180201/#ABParticipation
It seems to me that having the CEO chair the AB meeting isn't the best organization. It would make more sense for the AB to appoint someone else as the chair, to optimize for best chairing rather than best CEOing, and let the CEO be free to participate as a regular opinionated participant without having to balance the group.
My suggested wording:
This means technically that the AB can choose a chair who is not an elected member of the AB, but I think that's a feature not a bug. :) It might sometimes be useful.
The text was updated successfully, but these errors were encountered: