Wikipedia:Governance review/Proposals

Leave everything as it is

Just keep the system as it is.

Discussion - Leave everything as it is

I'm just adding this as something I'm sure will be proposed, I don't support it. --Tango (talk) 22:14, 8 July 2009 (UTC)

• While theoretical configurations alternative to the status quo are interesting -- maybe even important -- for the community to debate, I feel leery of any change that, at least in my feeling, might mostly be just for the sake of change itself, absent any any real problem it would be addressing. I feel like a regular diner at a boutique restaurant that is at present absolutely presenting a great product and I'd simply be a little unsettled to be greeted with the news that some of the staff are considering showing the founding chef who has designed, guided and promoted the place to the door of the kitchen. M-m-m-maybe. But I'm nervous. ↜Just M E here , now 01:10, 9 July 2009 (UTC)
  • It depends if you think our system is a great product. My problem is that our consensus system doesn't scale well, and Jimbo doesn't break the log jams. Unfortunately, trying to fix it with more consensus is a problem. - Peregrine Fisher (talk) (contribs) 01:35, 9 July 2009 (UTC)
    • Indeed. We generally get stuck in the status quo because it is so difficult to change anything when you need a consensus of 100,000 people to do it. The fact that this subject keeps coming up is evidence that something isn't right about the way Jimbo is handling his role (I'm not a fan of the principle "there's no smoke without fire", but in this case it does apply). I know consensus isn't going to fix the problem, that is why my proposed process involves a vote at the end. I know there is no chance of reaching a consensus on such a controversial issue. --Tango (talk) 01:45, 9 July 2009 (UTC)
      • Whatever comes out of this will need to scale well, and will need to include a mechanism for ensuring that change can happen. This will likely mean that major decisions are by a vote, whether by the community as a whole, or by a committee that was elected by the community. Kevin (talk) 02:33, 9 July 2009 (UTC

        Well said. I very much agree. (I'm not a fan of democracy, but as always, there really isn't a good alternative.) --Tango (talk) 03:00, 9 July 2009 (UTC)

Could someone actually detail what exactly the "unique powers" are, so we can understand what we're actually talking about. Thanks AndrewRT(Talk)(WMUK) 01:59, 9 July 2009 (UTC)

There was an attempt to that here. I'm not sure if the list is exhaustive, but all the main ones are there. --Tango (talk) 03:00, 9 July 2009 (UTC)

• This seems like the best option. I really don't see where there is an impending need in the community to any change in Jimbo's role. I am aware that Jimbo has said that he would be open to such changes, but that does not mean that the community needs it. I hear lots of talk here and at other similar proposals that the community is "ready" to change Jimbo's role, that it is "needed" for the community to "evolve". I don't see how. Other than these nebulous ideas about this mysterious "future" of Wikipedia, for the present, Jimbo works. The things he does, he does them well, and insofar as he doesn't do the bad, I don't see the need to change anything. --Jayron32.talk.contribs 03:21, 10 July 2009 (UTC)

  There is disagreement over whether or not he does them well and that causes drama every time he does anything. That drama is not good for the project, even if what Jimbo did was. --Tango (talk) 15:49, 10 July 2009 (UTC)

Remove Jimbo, don't replace him

Remove Jimbo Wales from all the positions he holds on the English Wikipedia (other than that of a regular admin). Don't have any replacement - ArbCom elections become binding, ArbCom rulings can only be appealed to ArbCom, other things Jimbo does simply don't get done.

Discussion - Remove Jimbo, don't replace him

I'm just adding this as something I'm sure will be proposed, I'm unsure about it personally. --Tango (talk) 22:14, 8 July 2009 (UTC)

• This is my preference. There is nothing that Jimbo does (that needs doing) that ArbCom can't do just as well. I would note tht despite the wording there is no particular need to remove Jimbo's Oversight and Checkuser access, presuming he sees a use for it. The biggest issue might be whether or not he keeps his founder access to the main (private) ArbCom mailing list. Eluchil404 (talk) 03:17, 9 July 2009 (UTC)

  If we do go down this route I would be inclined to let ArbCom make those decisions. They make them for everyone else, why not for Jimbo too? Trying to reach a community decision on them doesn't seem worth the effort. --Tango (talk) 03:33, 9 July 2009 (UTC)

  That is probably best. I certainly would be happy with it. In my experience, however, alot of the people who want to curtail Jimbo's authority also don't fully trust ArbCom. I do (especially in comparison with the alternatives) but want to identify the sticking points and necessary consequences of any action that we take. Eluchil404 (talk) 20:57, 11 July 2009 (UTC)

  If people don't trust ArbCom we probably need to do something about that. The new ArbCom could then make the decision. --Tango (talk) 21:18, 11 July 2009 (UTC)

• Jimbo is still a figurehead, the public face of Wikipedia - I have no problem in him continuing in that ceremonial role, and have little complaint really about most of how he has conducted his role previously. I do have a problem with him remaining/accorded admin status; his few admin actions have generally been very poor, with little indication that he is familiar with the current rules and the responsibilities of sysops in being answerable to their actions - and he does not respond well to being criticised for actions which would result in any other admin being brought before ANI or even RfAR. Whatever role might be offered to Jimbo, administrator should not be it. LessHeard vanU (talk) 20:41, 9 July 2009 (UTC)
  • We don't generally do community desysoppings, even for admins that pre-date RFA. I would leave that to ArbCom. I'm not sure we would have a problem, though - I think it is not so much Jimbo not being familiar with the rules but rather him feeling they don't apply to him (basically, he acts under the principles of IAR). --Tango (talk) 21:14, 9 July 2009 (UTC)

• If we were ready for the responsibility of governing Wikipedia, we wouldn't need to lobby for change. We'd just do it. I consider it highly likely that if we were to demonstrate the necessary coherence, Jimbo would turn over governance to us, because his continued role would have become unnecessary (or it would continue with our permission). And if he didn't, we wouldn't need to do anything other than demonstrate our coherence, not by disrupting Wikipedia, but by forking it, which would be trivial under the condition that we were ready to manage this thing. (We would have all the resources necessary, including abundant funding.) But, quite clearly, we are not ready. Until then, the protective role of a founder is highly desirable. (This is actually routine organizational theory for me, I worked most of this out before Wikipedia existed.)

Instead of developing fantasies of what our independence would look like, how about we develop the institutions that would advise us collectively? We don't need to change Wikipedia at all, in terms of policies and guidelines and Jimbo's role, to do this. All we need to do is to figure out how large numbers of people can communicate, coordinate, and cooperate, efficiently and effectively, without Big Daddy keeping us straight. It would be immediately useful, regardless of what happens to Wikipedia governance. I have ideas about how to do this, to be sure, but the reality is that, collectively, if we wake up, what we would develop would be, quite likely, much better than anything I can dream up. I'm just one person. --Abd (talk) 00:35, 12 July 2009 (UTC)

That is the same question that is asked in any discussion of governance, we don't need to start from scratch. We know various forms of governance and what the pros and cons are, we just need to work out which is best for us. I don't see why we need a system that gives us funding for a fork when we have no intention to fork. --Tango (talk) 07:09, 12 July 2009 (UTC)

Remove Jimbo, elect a replacement

Remove Jimbo Wales from all the positions he holds on the English Wikipedia (other than that of a regular admin). Elect a replacement to take on his roles. This could be an individual, a small council or a large parliament.

Discussion - Remove Jimbo, elect a replacement

I'm just adding this as something I'm sure will be proposed, I'm not sure about it personally. It's obviously very vague - feel free to either discuss the general concept here or make a more detailed proposal below.--Tango (talk) 22:14, 8 July 2009 (UTC)

I think one option should be a checks and balances, democratic process of some sort. Maybe model it on some countries system, but simpler. - Peregrine Fisher (talk) (contribs) 00:42, 9 July 2009 (UTC)

"Checks and balances" usually refers to a separation of powers (at least, the article redirects there!). Do we have powers that particularly need to be separated? Countries with a clear separation of powers (eg. the US) seem to have just as many problems as countries that don't (eg. the UK). A system without checks and balances would probably be simpler than one with (the Westminster system isn't exactly simple, but it is easier to simplify that the US system, I think). --Tango (talk) 01:39, 9 July 2009 (UTC)

It's kinda related to the question above about what exactly are Jimbo's powers. I'm not sure if we need three branches like the US, but it would be nice to have a body that can judge consensus, and also make hard decisions when consensus doesn't exist, as well as arbcom for conduct disputes. Maybe the two bodies could watch each other, or maybe there would be a third one. - Peregrine Fisher (talk) (contribs) 02:21, 9 July 2009 (UTC)

And all of these bodies need to be ultimately accountable to the community, at election time. Any interference in the elections by Jimbo or the foundation, whether real or perceived will only cause people to be suspicious of the whole leadership. Kevin (talk) 02:28, 9 July 2009 (UTC)

It is related, but only slightly. There is more to this than just replacing Jimbo, we need a better system of making high level decisions. Consensus just doesn't work with such a large group of people. --Tango (talk) 03:08, 9 July 2009 (UTC)

Arbitrators should be chosen by the community

Details of proposal

Jimbo's right to appoint only those of whom he personally approves to be arbitrators should be removed. Unless, a person is a known by the foundation to be a criminal (or something similar to be decided) then the will of the community (as decided by election) is absolute. Giano (talk) 23:18, 8 July 2009 (UTC)

Discussion - Arbitrators should be chosen by the community

While we are deciding how to elect Arbs, we should also discuss what their role is. I'll add a proposal once I've though a bit more on it. Kevin (talk) 01:52, 9 July 2009 (UTC)

Arbitrators are trusted users that perform a very important task in the community. I think there's a benefit to the current system of advisory elections with the power to overrule in highly exceptional cases - such as where the Board have information on a candidate they can't reveal to the community. I'd like to see this power formally transferred from Jimmy Wales to the Wikimedia Foundation Board (which itself is of course ultimately elected) but in practice that's a minor change. A more important reform is to make sure the identity of all Arbs are properly verified so we can prevent another Sam Blacketer scandal. AndrewRT(Talk)(WMUK) 02:07, 9 July 2009 (UTC)

The WMF board is only partially elected. There are 3 directly elected seats and 2 chapter seats, the other 5 are not elected (one is Jimbo, the other 4 are expert seats appointed by the rest of the board). I'm also not sure they would be willing to take on that role. I have informed Michael Snow about this review so hopefully we will get some clarification there soon. (Sue Gardner has already go back to me, but wants to double check things with Michael before making a public statement.) --Tango (talk) 03:06, 9 July 2009 (UTC)

I'm not sure they can avoid the role. They control the servers, and so are giving tacit approval to everything that happens here already. Kevin (talk) 03:44, 9 July 2009 (UTC)

There is a difference between allowing something to happen and formally approving it. --Tango (talk) 14:23, 9 July 2009 (UTC)

Hasn't Jimbo always just picked the top candidates by percentage? The only discretion he has really exerted is in determining the size of the committee. 146.151.21.117 (talk) 02:17, 9 July 2009 (UTC)

I believe he has sometimes chosen non-top candidates to fill extra seats he has created. (He hasn't done so recently and I'm just relying on my memory from when he did, so I may be wrong.) But you are right, the power is reserved for extreme circumstances, and Jimbo acknowledges and understands that. --Tango (talk) 03:06, 9 July 2009 (UTC)

The last such occasion was in January 2006 (the election having been delayed from December 2005 as the rules had not yet been settled). Sam Blacketer (talk) 16:01, 9 July 2009 (UTC)

Even without creating extra seats, Jimbo always moves people around to different tranches, filling the vacancies. The result is that people end up serving more or less than the 3 year term they were elected for, and we have a jigsaw puzzle to figure out. See WT:ACE2008 for the confusion concerning this in the most recent election. Firestorm ^Talk 00:43, 10 July 2009 (UTC)

I think recently he has filled the longest terms first, so the overall winners of the election get full 3 year terms, the runners-up get shorter terms. I don't think he has shortened or extended the terms of existing arbs in recent elections (he used to, I know). --Tango (talk) 00:59, 10 July 2009 (UTC)

• Comment: This is a limited form of option 2 "Remove Jimbo, don't replace him". AndrewRT(Talk) 00:53, 19 July 2009 (UTC)

Elect governing committee

Elect 3 governing bodies:

• Governing committee
• Content committee
• Editor behavior comittee

The Governing committee shall be comprised of directly elected editors. It will be responsible overall for content, policies and editor behavior. It will also act as a final appeal body for the other 2 committees. They would also be responsible for assigning user rights such as checkuser and oversight.

The Content committee will be the final decider of content, and of content policies. They would be empowered to provide formal dispute resolution and mediation. They would make binding decision on content and content policy where the community as a whole has been unable to reach a consensus on an issue.

The Editor behavior committee will function along the lines of the current Arbitration Committee. Their scope is editor and admin behavior, and related policies. Part of their role would be proactively dealing with problematic behavior before the stage where large and messy cases are required.

Discussion - Elect governing committees

All of these bodies should be directly elected from and by the editor community. In an ideal world, I would hope to see the great writers among the community on the Content committee, where their influence can be put to best use. Kevin (talk) 02:12, 9 July 2009 (UTC)

I think this is the way to go. I have a feeling only Jimbo can create such a thing, though. He might be willing to do so. Or maybe the wikimedia guys should do it. Then it runs itself. - Peregrine Fisher (talk) (contribs) 02:31, 9 July 2009 (UTC)

The mandate for this will never come from a community consensus. It is also unlikely that the foundation or Jimbo would enforce this without some sort of indication that the community want it. If it is considered a good idea then it should pass at a vote, after which it could just be implemented. Those who do not get aboard will eventually come around or be left behind. Kevin (talk) 02:39, 9 July 2009 (UTC)

I agree, we're never going to get a consensus on these issues. We will have to resort to a vote. What percentage should be required is one of the details that needs to be discussed on the talk page. I don't really see how Jimbo or the foundation could ever enforce something like this, it will need to support of admins (and editors in general) for that. --Tango (talk) 03:31, 9 July 2009 (UTC)

I'm not convinced we need a content committee. I can see the use of a committee to make final decisions about policy, but I don't think we have a big problem with irreconcilable content disputes. We sometimes have a problem with editor conduct during content disputes, but the disputes themselves can generally be solved by discussion and consensus (since there are far fewer people involved in a given content dispute than a policy discussion). The few cases where don't manage to resolve the dispute tend to be cases where there isn't really a good solution - often because the subject itself is very controversial (I recall a big dispute about the meaning of the name "Macedonia", for example, I don't know if that was ever resolved). I'm not sure a committee would do any better than anyone else at resolving that kind of mess. --Tango (talk) 03:14, 9 July 2009 (UTC)

The Macedonia problem was resolved by a content committee (they called it a panel) - see Wikipedia:Requests_for_arbitration/Macedonia_2/Proposed_decision#Establishing_consensus_on_names, followed by Wikipedia:Centralized discussion/Macedonia/consensus. Novickas (talk) 14:42, 17 July 2009 (UTC)

I would expect that the content committee would be mostly involved in facilitating mediation, and guiding policy. They would only get directly involved in content for the truly unresolved issues, like your Macedonia example. In those cases, even a wrong decision is better than endless edit warring and circular discussions. Kevin (talk) 03:28, 9 July 2009 (UTC)

Mediation is a very different job than decision making. I don't think it makes sense to elect mediators, you want people with talent and experience in mediation, not people with the right political views. --Tango (talk) 15:29, 9 July 2009 (UTC)

• It seems to me that the whole point of such a proposal is that the committes would derive their legitimacy from the community rather than top down appointment. Of course, if the foundation or Jimbo vetoed a community supported governance plan that would lead to a "constitutional crisis" and possibly a forking issue. I think that governance and content committes are more trouble than they're worth. In the vast majority of cases the community hashes out a reasonable solution. Eluchil404 (talk) 03:23, 9 July 2009 (UTC)

  Maybe, but in many cases (probably a minority, but still a significant number) the community just doesn't hash out a solution at all. WP:BLP only got implemented because Jimbo put his influence behind it. FlaggedRevs took nearly forever to get anywhere, and a trial still hasn't started (although I think it may be a technical problem now, rather than a community one). People have been calling for a reform of RFA for years without anything happening (a lot of people are unhappy with it, but nobody can agree on a better solution so we're stuck with the status quo). --Tango (talk) 03:31, 9 July 2009 (UTC)

  In situations like RfA I do not believe that change is necessarily a good thing. While institutional inertia makes any change very difficult, the biggest reason that RfA hasn't been reformed or replaced is that there is no consensus as to what the problem is. Lots of people have proposed changes that deal with the issues that they see as most prominent or that they believe would produce a better set of admins overall, but no proposal that I have seen (I admit that there are archives of WT:RfA that I haven't read though I have read quite a few) has garnered anything like consensus support. I do not believe that setting up governing structures that encourage the "tyranny of the majority" (or more realistically "tyranny of the most determined minority") is a good idea. Wikipedia's flat, consensus-based governance structure has weaknesses but despite our growth I remain convinces that it is also our greatest strength. Eluchil404 (talk) 21:05, 11 July 2009 (UTC)

I don't think an "Editor behavior comittee" is a good idea at all. Only those obsessed with civility and nice manners would apply to be on it, causing it to become a cross between the inquisition and a Barbara Cartland etiquette department. A Governing committee and a conent comitte is a good idea, so long as the latter has a high proportion of those who actualy write content. Giano (talk) 12:57, 9 July 2009 (UTC)

You would scrap ArbCom entirely? Some reform of ArbCom would certainly be a good idea (in fact, it is already happening), but I don't think we can do without it. Decisions about what to do with misbehaving contributors have to be made and we can't always get a consensus on what to do. --Tango (talk) 14:26, 9 July 2009 (UTC)

• I'm leary about the content committee thing. The reason that ArbCom does not currently handle content disputes is that in order to solve a content dispute, you would need to have some background to the source material which is disputed. No small committee would have the necessary breadth of expertise to make the decision that "This group of facts over here are favored over this other group of facts" Content problems need to be handled by people who know the content... --Jayron32.talk.contribs 03:25, 10 July 2009 (UTC)

  Good point. If we were going to have anything like that we would have to form a jury of experts for each dispute, and I don't see that working. --Tango (talk) 15:52, 10 July 2009 (UTC)

We have a very weird relationship with experts. Experts often have a COI and they often have strong points of view, and not uncommonly imagine that they understand the situation with their topic better than non-experts, i.e., ordinary Wikipedia editors. This is a classic problem in publishing: publishers do not allow writers to make editorial decisions, and sometimes the writers aren't experts, but skilled technical writers who rely upon a feedback relationship with experts. Experts, properly, advise. Courts do not appoint juries of experts to decide complex cases; rather, they collect expert testimony. Often, we end up blocking experts when we should actually encourage and solicit their participation for advice. Wikipedia has a terrible reputation among experts! We could, and should, fix that. The point is that the "executive" or "judicial" power is not handed to experts, but to neutral editors, who receive and weigh testimony from others more knowledgeable. One way to conceptualize this is that the "jury" represents the readership. If the experts can't explain the topic to the jury, just how useful is an article that they would write going to be? We saw a case recently, where a Wikipedia editor is -- allegedly, and most seem to accept it -- an expert in his field, and edit warred or argued tenaciously with what was considered an ordinary editor, though it turned out that possibly this second editor is actually highly educated in the field in question. The article was clearly, in my opinion, way too abstruse for Wikipedia, at least as written. I have the sense that it could be much better written, and, this is important, our "resident expert/owner of the article" isn't about to do that writing. The article seemed perfectly accurate to him! --Abd (talk) 16:08, 10 July 2009 (UTC)

(edit conflict)It is quite simply not true that "no small committee would have the necessary breadth of expertise...." First of all, proposals should never be considered in a vacuum, they must always be compared to the status quo or to the default, that happens if we don't have what is being proposed. Currently, articles are controlled by virtual small committees. These committees have open membership, anyone can participate. We have no coherent and universal understanding, though, of the process by which such committees make decisions. While that can be good, it can also be a disruptive disaster. Just as a thought experiment, though, consider this: there is a content committee, elected or appointed, similar to ArbComm, but designed for content decisions. When small groups can't agree, they go to the content committee. The content committee appoints a subcommittee of editors with expertise, or even with a reasonable claim of expertise -- this subcommittee is not going to control the outcome -- and this committee prepares a report on the situation. That report should be a full-consensus NPOV report based on the Wikipedia record, and allowing attributed opinions, documenting the dispute. (If we don't know how to do that, we should! No source reliability problems!). This report is then handled by a clerk for the content committee to prepare a brief summary. The report then goes to the content committee for a decision. Small committee (the content committee). Appointed subcommittee, which would include representatives of all points of view, procedure can be set up for that. Maximum expertise collected and even solicited without depending on the vagaries of watchlists. The report, plus committee deliberation in the decision, becomes part of the backstory, maintained with article Talk as an explanation for the decision. While this might seem cumbersome, again, it's not, when compared with the mess that we often get that results in no documentation, no clear basis for decisions, but merely a result based on who showed up when.

Now, the secret: we don't need a content committee to do this. We must simply understand the desirability of careful process, and then we prepare that NPOV report on the controversy. (This is about the Wikipedia controversy, not the real-world controversy, though, obviously, they would typically be related.) Then, if it hasn't already become obvious, thorough discussion often has this effect, we use RfC, which creates a virtual committee of neutral editors -- and a closer would do that analysis for neutrality -- to actually make the decision. It's now a documented consensus -- if consensus could be found -- based on a report and deliberation over the report, instead of in a vacuum. --Abd (talk) 15:56, 10 July 2009 (UTC)

The problem with that is that hardly anyone responds to RfCs, so the process doesn't actually work. Are you suggesting that might change if someone prepared such a report? Even if that were so in principle, it might be very hard to implement in practice. Problem editors often have little or nothing coherent to say, but just produce long, rambling, unintelligible screeds. How do you summarize such a discussion?

I agree, though, that experts shouldn't be necessary. The secret lies in a more thorough-going application of the verifiability policy. Everything must be verifiable. If somebody claims that a particular view should be given much/little space because, they claim, it is held by many/few experts, they must support that by verifiable evidence. WP:DUE quotes some guidelines on that from JW himself. Similarly, disputes about what is a reliable source should be settled in accordance with the principles laid down in the policy & guideline. The analogy of the jury you mentioned is a good one. Both/all sides in the dispute must present their case in such a way that the "jury" can follow it without prior expert knowledge. In fact a jury of people with no knowledge of the subject is desirable from another point of view: they will be able to give feedback on how readers are likely to understand proposed wordings. Peter jackson (talk) 15:44, 17 July 2009 (UTC)

All of these would mean extra elections, but election protocols are in place. Re content committee, I completely agree with the points about the jury model (altho I feel really, really sorry for the Enron jurors), the lack of response to RFCs, and expert assistance not being necessary to decide content disputes. In fact the panel at the Macedonia naming dispute didn't present themselves as experts on the topic. What if we had a good-sized elected content committee, say 30, and then 9 got the short straws in content cases unresolved by mediation? (2-week time limit?) There's well over a thousand admins and a horde of strong non-admin content editors to choose from. I do realize this might not prevent Althea from Athens, discouraged by arguing with Randy from Boise about the Peloponnesian Wars, from leaving the project, but at least Althea would know there was some formal recourse. I see a lot of "just calm down" or "stop creating drama" messages in content disputes - not helpful. Novickas (talk) 17:19, 17 July 2009 (UTC)

Since I like the jury model, I mentioned this page at Wikipedia talk:Wikipedia Committees, where a couple of other editors have expressed support for some sort of content governance structures. Novickas (talk) 20:17, 22 July 2009 (UTC)

Am I right to think the behaviour committee would be the current ArbCom? AndrewRT(Talk) 00:49, 19 July 2009 (UTC)

That was my thought, at least initially. Kevin (talk) 11:00, 19 July 2009 (UTC)

Because they are elected, however imperfectly at this time, I'd support Arbcom as the body that's authorized to send disputes to some sort of Content Dispute Committee. Novickas (talk) 20:17, 22 July 2009 (UTC)

Replace Jimbo with a Troika

Transfer the various powers Jimbo has (perhaps with minor modifications) to an elected Troika. A committee of three experienced Wikipedians that can ratify ArbCom elections (they can either oversee their own elections or have ArbCom ratify them) and handle ArbCom appeals (in extreme cases, the same as Jimbo does now). The Troika would have very clearly defined powers and would be elected to staggered 3 year terms (with some means of recall for when something goes wrong). The reason for long terms is because ideally they would not get bogged down in popularity contests and could get on with doing whatever needs to be done in as impartial a way as possible for the benefit of the project. The reason for such a small committee is twofold: 1) It is more efficient. A small committee can make decisions faster than a large one. 2) Keeping it small will discourage it from taking on more responsibilities (or discourage the community from giving it more responsibilities). The idea is to have a small committee with a clearly defined and small number of powers.

Discussion - Replace Jimbo with a Troika

I haven't made a decision yet on what I think is best, but this is one idea I've had. There are a few details to iron out (not least, exactly what the powers would be), but I'm putting it out here in this unfinished form so other people can fill in the details in true wiki-tradition (probably best to do it thread-style rather than inline though, or it will get confusing!). This idea may end up being part of a larger idea - I don't know if such a small group would be suitable for making policy decisions, for example, although they could give "Royal Assent" to them. --Tango (talk) 03:25, 9 July 2009 (UTC)

WP:COUNTER

Jimbo, as our esteemed leader, is in the unique position of not having to kowtow to the most vociferous elements of the community when he makes a controversial decision that he thinks is for the overall good of the project. This is a fantastic driving force for the project, and has the benefit of preventing rejected minority viewpoints and epic wikilawyering from crippling the community in its sole role as the builders of the pedia.

Ignoring what the community is not for a minute, one possible thing we could do to improve the legitimacy of this situation is to adopt the following as an addition to our ever expanding collection of policies:

Thereby we can hopefully have a quick and painless end to any drama about Jimbo's actions in the future, unless people want to bring down the arbcom system as well.

I realise that the fact Jimbo has ultimate authority over arbcom might make this proposal seem a bit daft, but Jimbo has in my observation made it clear he is not one to ignore the committee without just cause, so if he did so in the case of a negative COUNTER opinion, then I'm sure we would have reached the armageddon point for the project anyway, and we will be in a very bad place indeed.

I'm only really weakly suggesting this as a serious proposal, on the principle of devil's advocate. In matters like this I personally defer to NOT#BURO and NOT#DEMO, but I think that it might represent a decent middle ground which might appeal to the broad church, if this page does become an inclusive debate about the role of Jimbo.

That is assuming that at some point, any substantial documented evidence is actually presented that the current situation is harmfull to our goals.

MickMacNee (talk) 13:35, 9 July 2009 (UTC)

Discussion of WP:COUNTER

An interesting idea, and worth some discussion, but I don't think I am in favour of it. I think you would end up with too many decisions being challenged since people that are opposed are more likely to vote than those that support the action, so reaching 50% would be quite easy. I also don't think ArbCom should be making policy decisions. They interpret and apply policy, they don't make it. That is one separation of powers I think is worthwhile. WP:NOT is rather out of date. Wikipedia is highly bureaucratic and democratic, whether we like it or not. --Tango (talk) 14:32, 9 July 2009 (UTC)

Supposing the decision to be related to an action by arb com, referring it back to them would not seem reasonable. I like the general idea, bitt his takes some more thought. Personally I think JW would in fact respond in a reasonable way to majority discontent from a decision.

• effectively an RfC on each one of Jimbo Wales' action, this proposal would appear to be excessively bureaucratic, but at least something is to be built into the system for a centralised discussion than the disparate and chaotic discussions which take place each time Jimbo does something emotely controversial, like blocking Bishonen, for example. Ohconfucius (talk) 01:50, 10 July 2009 (UTC)

Direct Election of Arbitration Committee

This proposal is similar to Giano's "Arbitrators should be chosen by the community," but I believe I can flesh out the idea a little more:

In Arbitration Committee elections, the open positions will be filled by those with the highest percentage of support. Jimbo Wales may not alter the structure or results of ArbCom elections, except by mandate of the Wikimedia Foundation. If Jimbo or the Foundation have information that would make a candidate unsuitable, they may present such information to the community. Jimbo may not, however, have any unilateral veto power during the election process.

In addition, a five person committee composed of directly elected Wikipedians shall be responsible for determining the number of seats available, prior to accepting candidates. Once the number of open seats is determined and the election process begins, additional seats may not be created. certifying Arbitration Committee election results and informing the community of the results. They do not have the power to veto the election or results thereof, except in the case of abuse of the voting process After the voting period finishes, the committee shall be required to certify the results of the election and shall be given full power to investigate any irregularities.

The certification committee shall also be responsible for overseeing the appointment or election of any additional arbitrators outside of the normal election season, if so required by excessive vacancies on the Arbitration Committee.

The certification committee shall be directly elected each year, prior to the beginning of ArbCom elections. Sitting Arbitration Committee members are ineligible for election to this certification committee. Certification committee members are elected to onetwo year terms. The results of the certification committee election shall itself be certified by the Arbitration Committee, though they have no veto power over the results.

AMENDMENT: The number of seats that exist on the Arbitration Committee shall be determined by a referendum when necessary. Such a referendum may be initiated by either the Arbitration Committee or, through a Governance review, the community. In addition, the Certification Committee has no veto power, except in the case of abuse during the election process. This modifies paragraph 2 of the above proposal, as noted by struck through and italicized text. Firestorm ^Talk 01:01, 10 July 2009 (UTC)

AMENDMENT: Certification Committee members shall be elected to two year terms. Elections shall be staggered, with roughly half the committee being elected each year. This changes paragraph 4 in the manner noted above. Firestorm ^Talk 01:04, 10 July 2009 (UTC)

Discussion of Direct Election of Arbitration Committee

Alright, the idea behind this is that we're replacing Jimbo's role in election of ArbCom members with a 5-person committee that is directly elected each year. They'll certify the results and investigate irregularities, but can't actually veto the vote. I think this is a rather sensible solution, and it requires that voters know exactly what they're voting for (i.e. ensures that Jimbo won't suddenly create additional seats in order to place his favourite arbs back on there, even though they didn't win re-election). It'll help ensure that the rules of the game won't change midway, because everything will have to be finalized before the election starts. Firestorm ^Talk 14:47, 9 July 2009 (UTC)

This idea has a similar problem to my Troika idea - who watches the watchers? Who organises and certifies the elections for the certification committee? I'm not sure it is necessary to decide the number of seats for each election. Just leave the number fixed and change it by referendum (held either alongside or prior to the arbcom election) if the need arises. I would quite like staggered elections for such a committee (as I suggested for the Troika) - this gives some continuity and ensures the committee has some experience and also makes the committee less political. --Tango (talk) 15:01, 9 July 2009 (UTC)

The idea is that ArbCom rubber stamps the certification committee elections, but does not have veto power over it. As long as there is no major abuse, then the vote stands unmodified. The idea of having a referendum for the number of arb seats is something i'm open to, but I also recognize that having large referendums all the time isn't really feasible. I'm also open to having staggered elections; they would probably serve two year terms then and have 3 elected one year, 2 the next. I envisioned this as a low-activity position; all they do during their terms in addition to their regular Wikipedia contributions is reviewing the election results. But yes, i'm open to changing most of my proposal depending on what the community wants. Firestorm ^Talk 16:13, 9 July 2009 (UTC)

Just an idea - you might create a 'standards' board, to investigate accusations of impropriety which would be this committee's principle purpose, with the added purpose, as you suggest, of certifying Arb votes. As to Tangos quis custodet concerns, perhaps the incumbent arbcom could be used to ensure fairplay in the election of this committee. --Joopercoopers (talk) 16:22, 9 July 2009 (UTC)

It would need to be the incumbent ArbCom that are not up for re-election. So, with the current ArbCom system, that's about 2/3 of them. That might work. I'm not entirely comfortable with committees being involved in their own makeup, but if that involvement is strictly limited it should be ok. Perhaps the ArbCom Clerks could have some involvement? Since they aren't elected, they should be fairly disinterested. --Tango (talk) 16:30, 9 July 2009 (UTC)

There shouldn't be a need for referendums on a regular basis, the ArbCom should be able to stay the same size for long periods (we aren't growing that rapidly any more). --Tango (talk) 16:30, 9 July 2009 (UTC)

That is true, there should probably be one referendum soon to determine what the community believes the size of arbcom should be, even if we want to keep it the current size. Beyond that, we should only need to have additional referendums on the subject every few years, or if something should change rapidly. So, the referendum suggestion is workable. If nobody objects, i'll amend my proposal to add it. Firestorm ^Talk 23:03, 9 July 2009 (UTC)

Yes, some kind of ratification of the current size would be good, even if we don't see a need to change it. Feel free to amend your proposal - it's probably best to strike out the old text with <s> tags so as not to confuse people. --Tango (talk) 23:56, 9 July 2009 (UTC)

Note: I have amended the proposal. Any discussion below takes into account the first two amendments. Firestorm ^Talk 01:04, 10 July 2009 (UTC)

The only requirement we have of these Certifiers is that we trust them to run a fair election, right? Rather than create yet another distinct group of editors, why not just ask an odd number of bureaucrats to certify the results? This is an expansion of bureaucrat responsibilities, certainly, but a) they are already a group of editors we have a very high degree of trust in, b) they are a group that is explicitly defined in the MediaWiki software, which would be useful if we ever moved to using something like SecurePoll for elections, and c) the responsibility for certifying an ArbCom election has a fair amount of crossover with bureaucrats' existing responsibilities at RfA; indeed the job would probably be be easier as no judgement is required. Happy‑melon 13:39, 13 July 2009 (UTC)

That's not a bad idea. Obviously we would need to restrict it to crats that aren't standing for election and aren't existing Arbs. --Tango (talk) 17:22, 13 July 2009 (UTC)

Wouldn't it reduce the COI potential even further to use non-en users as certifiers, I was thinking of the de.wiki arbs or maybe the meta crats. MBisanz ^talk 17:30, 13 July 2009 (UTC)

Those are both workable ideas. Depending on how discussion goes here, i'm absolutely willing to revamp my proposal to reflect which of those is the better option. Firestorm ^Talk 18:19, 13 July 2009 (UTC)

Indeed. I don't think the meta crats are a good choice: the elegibility criteria are much weaker than on enwiki, plus the whole purpose of using crats in the first place - that they have already been 'certified' as trustworthy by this community - no longer applies. Ditto for a foreign ArbCom, although perhaps less so as they are more obviously trustworthy. I could see a reciprocal arrangement between dewiki and enwiki ArbComs working, but the language barrier might be problematic - not mainly in terms of the dewiki Arbs, most of whom I expect can read English, but in John Smith the editor here, who is expecting to see a transparent process but is hampered by an inability to read German. Plus what happens if the other wiki implodes, suspends its ArbCom or just can't be bothered to fulfil its certification role? Being a foreign project, there's even less we can do to influence them. Happy‑melon 10:36, 14 July 2009 (UTC)

Study, develop, and test methods for efficiently negotiating informed consensus

We are facing the classic problem of collective decision-making, and most proposals, as well as the status quo, represent classic solutions that are well-known among Wikipedians with a standard "western" education.

• Dictatorship. One person makes all ultimate decisions. However, that person may choose methods to allow the community to advise him or her, and there are thus two kinds of "benevolent dictators." One, which exists only in imagination, is someone who is so intelligent, wise, and well-informed that whatever decision they make unilaterally will be better than what the community could develop. The other chooses advisors well, and creates structure that allows the entire community to function as part of an advisory network. Dictators who manage the latter have survived, typically. But they die and succession mechanisms have proven unstable.
• Oligarchy. This is a variation on dictatorship, a privileged class makes all decisions.
• Direct democracy. The community makes decisions directly, without "governors." With good traditions and process, this works very well for small groups, but as the scale increases, it becomes impossibly tedious, and, increasingly, decisions come to be made by knee-jerk responses, without adequate deliberation.
• Representative democracy. The community elects representatives. In a parliamentary system, the assembly of representatives "hires" executives, and can "fire" them. In some systems, the executive is elected separately; however, when the scale is large, such elections are manipulable through well-known means. Further, with large enough scale, and especially without proportional representation, the assembly may not truly represent the community, but only a dominant faction or factions within the community.

All of these systems are well-known and have well-known problems. There are other systems, though, which have historically been used by various groups, often with long-term success. Under the conditions which Wikipedia faces, a successful system will probably be a hybrid. Rather than propose a specific system now, when the considerations haven't been developed and, my guess, if the ultimate system that we will adopt were presented now, without the foundation being laid, it would immediately be rejected as impossible. So I propose that we begin to "study, develop, and test" methods for rapidly negotiating consensus. When we can find true consensus, the exact governance system actually doesn't matter much, for all of the above systems work well if consensus is present and not blocked. Even raw dictatorships fall, quickly and relatively easily, when the people have somehow developed a consensus that it is to end.

But I will provide one hint: we will have to abandon the idea that there is anything wrong with "canvassing." After all, if !votes don't count, what's the problem with canvassing? If a lot of editors want to waste their time piling in with "me too" !votes, without more cogent arguments, without the presentation of neglected options that would increase consensus over anything already on the table, what's the problem? Canvassing is simply a form of communication, and communication -- which can be efficient under the right conditions -- is a necessary element in making intelligent, informed decisions. --Abd (talk) 15:05, 9 July 2009 (UTC)

Discussion

I see where you are coming from, but I'm concerned that such experimentation on the actual project could end up being harmful. Studying and developing ideas is what this page is for, testing them seems rather dangerous. It would also be very slow, they would need to be each tested for a long time to get meaningful results since problems don't actually come up that often that can't be solved by existing means. --Tango (talk) 15:26, 9 July 2009 (UTC)

Let's say that this is an impoverished view. "Testing" isn't necessarily the first step! Further, sorry, there is experimentation taking place, trials of various processes, and it doesn't necessarily take a long time. Tango, are you really saying that we should only consider methods of negotiating consensus -- which is a broad problem faced every day on wikipedia with many issues -- here on this page? Are you saying that we should, perhaps, develop some proposal for a major structural change, like a Wikipedia Assembly, involving much effort by many, many editors, endless debate, etc., before we have a spread of alternatives? And before we have any idea of how they would actually work, in practice as distinct from theory?

Slow? Consider how slow things are now!

"Experimentation" doesn't mean changing existing guidelines and procedures, an example.

it simply means trying new ideas, which is going to happen anyway; but the difference here is that there would be, perhaps, a WikiProject coordinating and studying efforts being made. As a personal example, almost a year ago I was blocked. I thought it unjust, that errors had been made, and that a detailed and careful examination of the situation and issues would confirm this. But how to accomplish that? I knew what would happen if I'd taken it to AN/I, which is not a place to get careful discussion of possibly complex issues. I decided to try something new. I set up a "self-RfC," in my own user space, and I laid out evidence and asked for comment. This was for my own advice; because I controlled it, it couldn't possibly be considered binding on the community or other editors. However, I was able to attract some independent participation, and I got enough confirmation there to go to the administrator who had warned me, with quite a pile of charges, and ask for him to reconsider. He didn't want to look back. So I asked if he could name a mediator. He did, and he made an excellent choice. We went to her (I think it's a she) and asked if she would mediate. She wrote, "Can't you guys work it out?" And we did. I accepted what I'd always known, that he'd acted in good faith, and he acknowledged that the warning had been excessive. I never did pursue the block itself, for various reasons, though recently I've had discussions with that admin. Okay, self-RfC. Innovative process. The only time wasted was my own, and that of editors who voluntarily participated. It developed a little mini-consensus, enough to give me a basis to pop out of my hole and seek a result.

I've since done something similar for other editors who were in dispute, invited them to a facilitated discussion. It's always worked, in fact, and what would almost certainly result in blocks, instead resulted in editors who had developed a facility for cooperating. Consider the alternative: one of these editors goes to AN/I with a complaint. And the community decides, without depth and patience, who to throw the book at, if not both.

Experiments are already happening.

What does this have to do with governance? Well, it's the concept of, at least initially, confining discussions to a small group, under conditions which facilitate consensus. (Or, having found consensus on a small scale, it is then brought to wider attention for ratification or "return to committee based on new information and arguments." It's not a change from existing process, but one of the problems is that we don't have intermediate stages. It's very difficult to find consensus, with a deep dispute, with more than ten or twenty editors involved. Taking the problem to a page watched by perhaps thousands of editors, because a handful can't find consensus, is a huge leap. Before something goes to such a large group -- if it ever goes to that group -- much more serious efforts to negotiate consensus -- which is almost never a matter of Yes or No, but will usually be found over a previously unproposed option -- should be undertaken. This takes time, it can't happen quickly, usually. It is known how to do it, there is much experience available. Which we don't take advantage of. The changes necessary don't involve changes to policies and guidelines, what is being described is simply within what can be done under existing guidelines, even though it is sometimes unfamiliar or seems strange. My self-RfC was widely ridiculed, and attempts were made to get me sanctioned over it. Wisely, however, the community realized that if I wanted to waste my own time with something so silly as a self-RfC, I should be allowed. Possibly complex decisions should never be negotiated on a large scale. One of my realizations over the last thirty years of studying this problem (as I've said, it's a generic problem) is that it is deliberation that doesn't scale well, not voting. What if complex decisions were hammered out as either a consensus or a set of proposals, each one of which has been reviewed by interested parties, and, if the latter, if no proposal has solid consensus, the set is presented for vote, with reference to the evidence and arguments? This is what, in fact, large deliberative bodies do: the real work takes place in committee, complicated issues are almost never debated on the main floor except for grandstanding. So, another way to put this, how would we set up efficient committee structure? Without introducing some new bureaucracy? Actually, I think I know how to do it, and it would look, to a casual observer, very much like what we already do.

Is any project looking for and monitoring these "tests," which are already taking place?

Problems with prior tests may not be insoluble "by any existing means," or new means may be considered, some of which require doing nothing risky. The claim was made that my self-RfC was risky because, since it was in my user space -- and I reverted contributions by a disruptive editor -- it would come up with a biased result. But it didn't exist to advise the community, it existed to advise me. And, of course, the advice I wanted was on the question of whether or not I had done what I was accused of. If I had, rather useless to pursue some dispute over it. If I controlled my advice to the extent that I repressed criticism that I needed, I'd be shooting myself in the foot, I would, then, armed with advice that I essentially gave myself, go up the DR ladder just in order to get shot down and possibly blocked as disruptive. What I do with discussion pages I set up in my user space for other editors, when asked, is try to facilitate agreements between them, and there are plenty of techniques for doing this, if the editors are acting in good faith, and when they aren't, it becomes rather obvious. In fact, though, most disputes do involve good faith differences of opinion, and can be resolved, often, by very small-scale discussions. WP:DR contemplates this, but practical means for doing it aren't well described and suggested, and small disputes become huge ones as positions harden and factions form (sometimes representing real-world factions, but real-world disputes frequently fester and deepen precisely because communication mechanisms between the factions don't exist.)

--Abd (talk) 17:11, 9 July 2009 (UTC)

You're just putting words in my mouth, so I don't know how to respond. --Tango (talk) 17:18, 9 July 2009 (UTC)

My apologies, certainly that's not what I intended. If words are being stuffed into your mouth that don't fit, please spit them out and replace them with words that do. I played, to some extent, on possible meanings of what you wrote, I don't think I invented anything, but, sure, you may not have intended those meanings. What did you mean? I understand and agree that some forms of "experimentation on the actual project could end up being harmful," but, rather obviously, potentially harmful experimentation wasn't being suggested. Indeed, some of the experimentation might not be on the actual project. And how could you know "it would also be very slow," when no specific proposal has been made and no structure has been set up to deliberate the proposals? --Abd (talk) 17:28, 9 July 2009 (UTC)

The point of experiments is to see what happens, so obviously you don't know what will happen before you start. If could go well, or it could go badly, if it goes badly that will be harmful (hopefully minimally, but you can't be sure). I know it would take a long time because you need to have several problems come up that need solving before you can judge if it solved them effectively, and we don't get problems coming up very often. --Tango (talk) 17:58, 9 July 2009 (UTC)

Call them "demonstrations," then. Obviously, nobody is going to be interested in working on random "experiments" that have no expectation of success, but we shouldn't reject ideas simply because they haven't been tried or accepted; if someone anticipates a problem, that should be addressed and protection against those problems built-in to the experiment or demonstration. But what I've seen is knee-jerk, total rejection of ideas simply because "that's not the way we do things." If we only do things "the way we do things," we will, obviously, be totally stuck. --Abd (talk) 14:26, 15 July 2009 (UTC)

Eh? We don't get problems coming up very often. Very often compared to what? If you focus on an obscure, non-controversial article, and know policy and guidelines and follow them, or just deal with relatively uncontroversial stuff like reverting vandalism, sure. Otherwise I have to wonder what Tango means. I'd disagree. From a single small-scale success, it could be known that a process works, at least part of the time, and it can then be applied to another example if it is observed that it worked the first time. What is really proposed here is simply a WikiProject to document such experiments and assess them, and possibly to help develop them, i.e., to discuss what has worked and what went wrong if anything went wrong. Look, we see the problem right in front of us, right here. Simple proposal. It ought to be obvious. But people would, apparently, rather spend a great deal of time discussing far more complex and, in fact, far more dangerous proposals that could seriously transform Wikipedia in a negative way. A Wikipedia Assembly, done in the right way, could be fantastic. Done in the wrong way, it could destroy the best of Wikipedia. If we can't find a way to deal with small issues, one at a time, how in the world do we think that we are going to tackle big ones? Magic? Intervention by Higher Authority? Sheer Good Luck?

I'll say this: if we had a way of efficiently analyzing proposals and developing consensus regarding them, without just plopping them, ill-considered, in the community's lap, as is happening here, we'd be able to solve the problem of governance rapidly. The same process that would work for small problems, once shown to work for them, can work for large ones, as long as the structures are designed to be scalable, and I do know how to design such structures, that's why my panel topic for New York this month has the word fractal in it. --Abd (talk) 03:14, 10 July 2009 (UTC)

My proposed process includes 8 weeks of discussion. The ideas will not be ill-considered. I'm not sure what you are saying ought to be obvious - you haven't proposed a solution, you have proposed a process for finding a solution. --Tango (talk) 15:57, 10 July 2009 (UTC)

Not true. The process is the solution, assuming it is used. 8 weeks? It's possible if we had better process, sure. But we don't. Part of what I propose could, quite easily, elect a Wikipedia Assembly without holding an election! It would be a standing election, so "terms" would not be fixed, but participation rules could make it quite stable. I.e., once you have a "seat," there is latency. If you lose your support, you lose !votes, assuming that proxy assignments are considered in !voting (if the goal is consensus, as it should be, those assignments aren't all that important), but you don't immediately lose a voice, the right to post to an assembly record. There would be two ways to lose that voice: by consensus of the assembly itself (which could happen no matter how many supposed support votes you have) -- and it is thoroughly traditional that an assembly determines its own rules, they are almost never imposed from outside, by the electorate, which determines the powers of an assembly, but not its process -- or, after a delay, if your support drops below a certain level and the Assembly does not decide to allow continued access. Participants who are deemed useful could still have a "seat" at deliberation by permission of the Assembly; this would be especially useful with Assembly committees, which could be as open as remained functional.

The point of having an Assembly instead of a monster everyone-can-post set of pages is to keep the noise down. Noise is the reason why direct democracies on a large scale have been considered impossible. But an Asset assembly could retain direct voting without noise.

It's just an idea! Asset voting type assemblies have been proposed, there were proxy assemblies proposed in some places in the U.S. in the early 20th century, but its never been done. For an advisory body, though, it would be quite safe, if it failed, it would fail gracefully, from disuse, little having been invested. I'll be discussing some of the concepts in New York, and I'll try to have a document ready before then. --Abd (talk) 12:35, 13 July 2009 (UTC)

Hire an Ombudsman to take over the overview functions of Jimbo Wales

Same as we have Mike Godwin to refer to in legal matters, the project may employ a person to act as a person of final appeal regarding ArbCom, to perform extraordinary actions to protect the project per the powers ascribed to Founder status, to supervise ArbCom elections and to scrutinise and appoint successful (via popular vote) candidates. LessHeard vanU (talk) 20:33, 9 July 2009 (UTC)

Discussion re. Ombudsman

I'm not sure why we would pay for something we can easily get for free. There are plenty of volunteers willing to help out with that kind of stuff. That aside, how would you select this person? --Tango (talk) 21:11, 9 July 2009 (UTC)

I also disagree with having someone paid. A lawyer is something that you can't really have a volunteer do. As far as arbcom appeals and all that, short of hiring a professional appeals judge, we wouldn't get much out of it. People from the community are more than willing to volunteer their effort, and we wouldn't have to pay them. In addition, I think that giving one person the power to overturn ArbCom decisions is a bad idea, and someone outside the community doubly so. I think arbcom should be the final word on things, similar to the Supreme Court of the United States. If anything should have the power to hear appeals, it should be an appeals committee. Firestorm ^Talk 22:58, 9 July 2009 (UTC)

The Supreme Court of the US isn't the first word on anything, though (except maybe a few special cases). Some way to appeal decisions of the court of first instance to a different body seems like a good idea to me. --Tango (talk) 23:59, 9 July 2009 (UTC)

The Supreme Court has original jurisdiction over a few things, such as when a US State is a party to a lawsuit. ArbCom is usually not the first word on things, either. In fact, in most cases you have to prove that you have tried other processes, such as WP:MEDCAB, WP:MEDCOM, or WP:RFC first. When that fails, you ask ArbCom to accept the case. In that way, ArbCom is already sort of a court of appeals. Firestorm ^Talk 00:37, 10 July 2009 (UTC)

All that comes before ArbCom is attempts at consensus. That's just like coming to an out-of-court settlement (to extend the metaphor). --Tango (talk) 01:01, 10 July 2009 (UTC)

Possibly, sure, but keep in mind that we're not actually a nation. RFC, Dispute Resolution, and the Noticeboards are our version of lower courts. At the Noticeboards especially, we have more or less a jury of your peers (well, hopefully). When that doesn't solve an issue, it gets appealed a step higher. Arbcom is the final appeal. Its not a 100% accurate metaphor, but for the most part we have parallels to a court system. Firestorm ^Talk 01:22, 10 July 2009 (UTC)

• I reaally really like the idea of some kind of hired-gun ombudsperson/editors' advocate/internal affairs go-to-guy. ↜Just M E here , now 03:41, 10 July 2009 (UTC)
  • This isn't a vote so could you explain your reasoning, please? --Tango (talk) 15:58, 10 July 2009 (UTC)
    • My sense is that perhaps some kind of users' advocate could provide an additional check and balance toward conflict resolutions. Although there's ALWAYS gonna be drama (and any proposals put forward with the idea that they would rid WP of the same simply are pie in the sky), IMO it'd still help to have an independent avenue to address concerns. When Jimbo's actions become labeled "seat of the pants" they could be taken to this independent gal/guy -- somebody who, I assume, would have spent a lot of time studying WP's history (a common law-type analogy here) -- who could choose to offer hi/r two cents as well. Hmm ---- Since this position, which I'd envision would be added to Jimbo's position, not replace it -- would be PAID, maybe we could try and take a collection from editors to fund the position. If three of us put in $10 each, that would be $30. Enough to fund 5 minutes of the time of a trained legal professional, perhaps. ↜Just M E here , now 16:31, 10 July 2009 (UTC)

      I'm concerned that a professional lawyer resolving disputes would lead to wikilawyering like you've never seen before. --Tango (talk) 19:47, 10 July 2009 (UTC)

      My analogy would be to a publication's ombudsperson, who always, as far as I know, lacks actual supervisory powers. (Even police departement's Internal Affairs officers provide info to the top commander but don't have a veto, I think, right?) Such an official, indie voice is listened to (to the extent it is, presumably) due to the holder's argumentation's reflcting hi/r reputation for experience/competence/integrity/attention to principle/&c. ↜Just M E here , now 20:24, 10 July 2009 (UTC)

      And you're suggesting we hire a lawyer? They hardly have a reputation for integrity and attention to principle. (The reputation they do have is probably rather unfair, but that makes no difference.) --Tango (talk) 21:07, 10 July 2009 (UTC)

      My comparison of choosing someone who possesses an understandiung of WP's principles, history and practice with someone expert in common law was just an analogy. Still, any expert the Foundation would hire to guide them would be by all expectations/by definition a sharpy, right? (In a good sense -- ie a "professional" type.) ↜Just M E here , now 01:59, 11 July 2009 (UTC)

• This proposal hinges upon a presumption that Jimbo's function is interchangeable in the way legal counsel is interchangeable. There are certain well-established criteria that would make various candidates suitable lawyers. English Wikipedia is unique in noteworthy ways: it is by far the largest wiki, it is the world's most popular reference website, etc. It became these things in less than a decade and is experiencing growing pains. Jimbo Wales fills a high level function as institutional memory which no other person has. Eventually, of course it will be necessary to replace his individual role with something else. He isn't immortal. But I don't think the premise for this proposal is tenable, nor (if it were tenable) is there urgency to pursue it. Durova²⁷⁵ 04:59, 10 July 2009 (UTC)

Ombudsperson? Peter jackson (talk) 17:36, 13 August 2009 (UTC)

Hire an Ombudsman to give input to the powers that be, including Wales

Same as the above except for the "Replace Jimbo!" part.

Discussion of supplementary ombudsman

See above. ↜Just M E here , now 18:35, 14 July 2009 (UTC)

This has all the same problems as the other version in addition to not being as useful (since there isn't such a need for a person if Jimbo is still doing the job). --Tango (talk) 18:58, 14 July 2009 (UTC)

om⋅buds⋅man (Dictionary.com) noun 1. a government official who hears and investigates complaints by private citizens against other officials or government agencies. 2. a person who investigates and attempts to resolve complaints and problems, as between employees and an employer or between students and a university. Origin: 1910–15; < Sw: legal representative, equiv. to ombud agent, attorney + -s ’s + -man -man

I don't follow your objection, Tango. No black-and-white determination that some organization had not previosly been doing its job fairly and effectively would be required for the organization to benefit from employing an ombudsman. Eg a university's hiring of one wouldn't necessarily imply it had been grossly and habitually neglected students' rights, IMO, so much as that it wanted to instutute a very high level of accountability and of real-and-perceived fairness in its decisions involving students, by enabling students' to present their side, within conflicts with the university, to an entity empowered to render independent judgements. ↜Just M E here , now 20:58, 14 July 2009 (UTC)

I didn't say it wouldn't be useful, I said it wouldn't be as useful. The various problems with the idea aren't going to be as worth it if it isn't as useful. --Tango (talk) 00:07, 16 July 2009 (UTC)

Variable sized parliament

There is a parliament with the authority to make policy decisions and to which ArbCom is accountable. This parliament is not elected as such, but instead anyone can sit in it as long as they have 20 signatures from established contributors in good standing (similar suffrage to ArbCom elections). If you are sitting you can't be a signatory and you can only be a signatory of one person. Signatures can be added and removed at any time and people join and leave the parliament on the Monday after their number of signatures goes above or below 20. If the added complication is deemed worth it, this could be done using preferences with a single transferable vote-like system (basically it is STV but with the quota fixed at 20). The number 20 is, obviously negotiable, and could be changed by referendum. The details of what powers this parliament would have can be determined at a later date if people like the general idea.

Discussion

Proposing. What do people think? --Tango (talk) 02:04, 13 July 2009 (UTC)

Interesting proposal. I think it could work well. A few points would need to be addressed: minimum quorum to make valid decisions and what method the body uses to make decisions to start. Kevin (talk) 03:00, 13 July 2009 (UTC)

Indeed, the procedures need to be worked out, but they are technicalities that can be dealt with later. I envisage it working fairly similarly to the UK House of Commons, but without a government (and probably without political parties, although they may arise). For a proposed motion there would be a combination of discussion as a whole and discussion in committees culminating in a vote. The exact details can wait. --Tango (talk) 03:05, 13 July 2009 (UTC)

You don't need transferable voting as such, because the procedure already outlined allows people to transfer their votes if they see their preferred candidate has too many or few votes anyway. Transferable voting was only invented to model that. Peter jackson (talk) 09:44, 16 July 2009 (UTC)

I thought of that, but it ends up being a little unfair on those that vote for the most popular candidate. They don't get to express a second preference while people that come along later can. A STV type system prevents that. It would also reduce the turnover rate of members, which would reduce overheads (you don't need to spend as much time orienting new people and rearranging committees). --Tango (talk) 12:38, 16 July 2009 (UTC)

Sorry, I realized that, but I took your wording to mean STV in addition; perhaps you meant instead. Peter jackson (talk) 13:41, 17 July 2009 (UTC)

Yes, an interesting proposal. I don't know if I totally support it though. --Merovingian (T, C, L) 02:58, 9 August 2009 (UTC)

Add more Jimbos

If you look at the times that Jimbo has intervened in the last few years, he's virtually always been right, and always been helpful. The problems Wikipedia has are generally a lack of benevolent deities, not a surplus of malevolent ones. Jimbo rarely pays all that much attention these days, so it would be good to get more wise, respected leaders who could step in and arbitrate disputes the way Jimbo has done in the past. Generally, the approach that works well is to listen, discuss, make a firm decision, then gently, gently push it into place with help from various opinion leaders. More senior editors, please! Stevage 02:36, 14 July 2009 (UTC)

Discussion

My only comment is that Jimbo is only usually often right because he is Jimbo, and not because he is adhering to policy or guideline. LessHeard vanU (talk) 21:11, 14 July 2009 (UTC)

 Done You are all Jimbos. Thanks for volunteering. Individual Jimbo privilege levels may vary, according to experience and recognition by the community, but being a Jimbo conveys the right and responsibility to follow WP:IAR at all times. You don't need to have administrative tools, you merely need to know how to ask, and mature Jimbos use tools themselves less and less, and even ask less and less, and certainly they demand nothing. Where Jimbos, with practice and diligence, attain clarity and depth, their mere presence reforms the community with every action. Jimbo hats will be offered for a modest donation to the WMF, or as gifts from a grateful community. --Abd (talk) 14:50, 15 July 2009 (UTC)

Shouldn't there be a Request for Jimboship procedure, complete with a structured discussion, !votes, and a decision from a closing bureaucrat? *Dan T.* (talk) 23:43, 15 July 2009 (UTC)

No. No approval is needed. If approval is needed, it's not a Jimbo. The hat is optional. --Abd (talk) 02:40, 16 July 2009 (UTC)

We have too many Jimbos already.--R.D.H. (Ghost In The Machine) (talk) 09:57, 16 July 2009 (UTC)

Too late! --Abd (talk) 15:47, 16 July 2009 (UTC)

My only question in this matter is; are we going to permit paid Jimbo's, or should it be voluntary Jimboism only? LessHeard vanU (talk) 20:10, 16 July 2009 (UTC)

It should be compulsory - "We are Jimbo, resistance is futile..." Kevin (talk) 22:41, 16 July 2009 (UTC)

Compulsory payment or compulsory service? My general position is that anyone who wants to donate to a Jimbo may do so, and, indeed, we could not prevent it; Jimbos may accept or decline donations ("the workman is worthy of his meat") but some may turn donations over to the foundation. Sale of hats should be quite adequate, thank you. I hope to have them ready for New York. With apologies to the Smothers Brothers, "If you get a Jimbo outfit, you can be a Jimbo too." --Abd (talk) 03:05, 17 July 2009 (UTC)

How would such additional Jimbos be chosen? AndrewRT(Talk) 00:42, 19 July 2009 (UTC)

A Constitutional Convention

A gathering of all Wikipedians in good standing and interested outsiders for the purpose of drafting a formal, centralized governing document. Such a document would attempt to address all major issues including, but not limited to;

• The continued role, if any, of Jimbo Wales.
• The role of Arbitration Committee.
• Administrator authority vs accountability.
• The rights and responsibilities of editors and contributors.
• Policy making.
• Voting vs Consensus.

Once an acceptable draft has been composed, it shall be submitted to the community at large for ratification.

Discussion

Proposing, now it's your turns.--R.D.H. (Ghost In The Machine) (talk) 12:06, 14 July 2009 (UTC)

That's basically what we are doing now, although we haven't decided if we want a single written document as a constitution or not. --Tango (talk) 16:10, 14 July 2009 (UTC)

No, we haven't really decided anything here, have we. I was hoping that this was, at best, a prelude to a ConCon, but it seems it is merely yet another exercise in futility 2.0.--R.D.H. (Ghost In The Machine) (talk) 09:45, 16 July 2009 (UTC)

Of course we haven't decided anything, we're a week or so into a 2 month process. --Tango (talk) 12:35, 16 July 2009 (UTC)

And at the rate things are going here, nothing will be decided still in 2 months.--R.D.H. (Ghost In The Machine) (talk) 13:50, 16 July 2009 (UTC)

Unfortunately, people have been distracted by ArbCom taking unilateral action. --Tango (talk) 14:07, 16 July 2009 (UTC)

To be fair, people have allowed themselves to be distracted. Before Arbcom did what it did there were several other discussions on governance going on - a few of them listed here. The fractured nature of discussion on the topic is a major reason why nothing is happening. We need to drag all these discussions & proposals together so that there is a larger body of people all discussing the same thing. Kevin (talk) 22:48, 16 July 2009 (UTC)

I think this is a good idea. It could, of course, be a "participative" convention as well - one where the all editors were able to contribute. The key to its success woud be its ability to make decisions on narrowing down the options. AndrewRT(Talk) 00:09, 17 July 2009 (UTC)

I've started a proposal for a constitution here. → ROUX ₪ 12:45, 17 July 2009 (UTC)

Wikipedia committee

What is the problem that this governance review is out to resolve? I suggest it is two things: (a) the lack of legitimacy in many eyes of the certain residual powers possessed by Jimbo, and (b) the problems of "scalability" of the consensus model - the difficulty of making effective decisions when the community is as large as it is. The following proposal seems to me to be the best way of acheiving both, in a way that is most likely to get approval from the key stakeholders - the Foundation, Jimbo and the editing community.

I suggest the formation of a Wikipedia Committee which would be part elected and part appointed by the Wikimedia Foundation. The elected members could use the same system as for the elected WMF board members. The committee would have the following powers:

• to exercise or delegate the current unique powers that Jimbo possesses, i.e.:
  • to appoint ArbCom members based on the results of the election
  • to overrule decisions of ArbCom
• to make decisions on policy where the community is unable to reach consensus
• to lead the community by initiating debates on topics of importance

Note: I fully expect the first decision of the committee would be to delegate the "unique powers" back to Jimbo, thus providing the necessary continuity and stability whilst giving them more legitimacy.

Discussion

Proposed by me. AndrewRT(Talk) 00:21, 17 July 2009 (UTC)

Rule of law

Just a general idea at present. Wikipedia should be governed by the rule of law, like a civilized society. There should be clearly defined rules of what people can & can't do, both behaviour & content, & they should be effectively & consistently enforced by the authorities, whoever they might be. Peter jackson (talk) 15:49, 17 July 2009 (UTC)

Discussion

I don't understand what change is proposed - apart from repealing WP:IAR? AndrewRT(Talk) 00:41, 19 July 2009 (UTC)

1. Yes, it does include that.
2. Make the phrasing of policies more precise, probably incorporating or promoting a lot of guidelines.
3. Enforcement of content policy, not just behaviour.

Peter jackson (talk) 10:36, 20 July 2009 (UTC)

Are you by inference saying that current policies and rules are quite fuzzily applied under the surrent system? And that enforcement in an inconsistent manner on behaviour, while essentially offering no such enforcement on content violations, is a problem? Collect (talk) 12:14, 25 July 2009 (UTC)

Yes. Peter jackson (talk) 11:09, 30 July 2009 (UTC)

Rule Of The Robots

All of the other proposals, regardless of their other virtues, have one common, fatal flaw-They all depend upon the agency of imperfect, biased, slow humanoids. Therefore I boldly propose the following radical leap into the future:

Bots are our friends. More than that, they are indispensable tools. Without their help Wikipedia would simply become unmanageable. They already do many of the tasks which used to be reserved for humans (henceforth referred to as meatbots) and admins. Why just look at the top editors list; it's dominated by bots! And this should come as no surprise, after all, they are fast, efficient and completely objective...in short everything that Wikipedia's current governance establishment is not. Their numbers and lists of functions are growing by the week. So let us not only recognize but embrace the inevitable reality and Turn Wikipedia over to its bots!

Fully automate governace! Take away all powers, along with all the corrupting influences, worries and higher responsibilities from all the meatbots, so they can focus solely on writing and improving this fine encyclopedia. Imagine the possibilities:

• AdminBots would swiftly and fairly deal with unruly meatbots.
• 'CratBots would keep the unruly AdminBots in line!
• RFABots would decide which ordinary bots become AdminBots.
• ArbBots would overule or endorse the actions of the other bots.
• BauderBot would write the decisions for the ArbBots and tell them how to vote.
• SandyBot would review all feature article candidates.
• RaulBot would endorse SandyBot's decions and decide which articles go on the mainpage.
• BotApprovalBot would decide which bots will be allowed to operate.
• A new RickBot would randomly Rickroll the meatbots, to remind them it is all in good fun.
• BarnstarBots would award random barnstars to random meatbots to keep their moral up.
• SlummyBot and DurovaBot would handle checkuser.
• DannyBot and DaveyBot would handle oversight.
• Jimbot would randomly, but rarely, interfere with the other bots' operations just to keep things interesting.

So let us, then, embrace the rule of robots, because, let's face it; robots rule, and that's The Wørd!

Meatbots Who Welcome Our New Automated Overlords

1. Did I say overlords? I meant Protectors!--R.D.H. (Ghost In The Machine) (talk) 12:57, 18 July 2009 (UTC)

Would they be held to the Three Laws of Robotics? *Dan T.* (talk) 16:58, 18 July 2009 (UTC)

No. Why should mere fluff, statements that document policy but don't create it, prevent robots from improving the project? Good robots will be programmed to ignore all rules, and, given admin authority, will ensure a reign of peace, with disruptive edits immediately reverted and neutrally oversighted, why bother with IP blocking when you can make evading edits useless? Robots, intrinsically, don't have a POV, so all robot actions will be completely neutral. Filing complaints with ISPs will also be automated, and, eventually, disruption can be eliminated at the source, editors who ignore rules. Compliant editors will have the full freedom to comply with policy, and mistakes will be easily forgiven if not repeated and there is an abject apology. Usually. Unless they aren't, such as the edit challenged the authority of the Protectors.

(Pay no attention to the man behind the curtain.) --Abd (talk) 18:33, 18 July 2009 (UTC)

Robots are very law-abiding...if they are programmed to be...and if they don't decide to change the laws. But if they do I'm sure it would only be for the best. You've got to assume good robo-faith!--R.D.H. (Ghost In The Machine) (talk) 10:13, 19 July 2009 (UTC)

Expanding on Tango's variable size parliament

I'm intrigued by User:Tango's variable size parliament, so I am attempting to expand on how such a body could work.

A unicameral parliament would be formed in the manner described by Tango. It would not begin work until a certain number of editors had been elected to it, and this number would also be the quorum required to conduct business. The specific requirements to be seated in the parliament are open to debate, but I believe that we should focus on making it relatively easy to gain and lose one's seat. This would prevent the parliament from becoming overpowered and stale, and prevent its members from becoming an elite and separate class.

If we really wanted to ensure that parliament members never get comfortable, we could institute a steadily increasing vote requirement for holding onto a seat. For example, it takes 20 votes to gain a seat and hold it after that. After three months, 30 votes are required for an editor to continue holding the seat. After six months, 40 votes, and so on. I'm not implying that there would be formal elections every three months under this system (or at all). Rather, the editor has until the three month deadline to have the required number of other editors in support. If an editor loses his or her seat, he or she must wait three months before regaining it.

I see the parliament's main role as drafting formal policy proposals. These proposals could only be edited by parliament members, just as Arbcom decisions can only be edited by Arbitrators now. The main advantage of having "formal" proposals, in my mind, is that they would be all in one place, follow a set structure and be a detailed examination of the idea from all sides. However, community members could still draft policies informally as they do now, and if more than 75% vote in favor of an informal proposal, it simply becomes policy without involving the parliament at all. This threshhold is, of course, debatable. If the parliament passes a formal policy proposal, it must then be passed by the community in a referendum, but with only a simple majority required for passage.

The parliament's powers could end right there, and that would be my preference. However, the community could assign other powers to the parliament if it chooses. For example, it could be the final appellate body for Arbcom cases, or determine the size of Arbcom. If any powers outside of policy proposals are to be added, it probably should occur over time through community referenda.--Danaman5 (talk) 15:39, 21 July 2009 (UTC)

Discussion - Expanding on Tango's variable size parliament

I'm concerned mainly with the idea of a variable size parliament in principle. Specific points of operation can always be determined later.--Danaman5 (talk) 15:39, 21 July 2009 (UTC)

One problem with an increasing number of required votes is that, if you use a voting system similar to the one I described, it will often be different people losing their seats than have been there a long time - people will have to move their votes (or have them moved automatically if we use STV) from the marginal candidates to the more popular ones to keep them in power, so the marginal candidates will lose their seats. I quite like the idea of a proposals being passed by either 75% of the community or 50% of the community and 50% of the parliament. I think it is important that the parliament separate into committees to discuss specific issues in great detail before reporting back to the whole parliament. --Tango (talk) 15:59, 23 July 2009 (UTC)

That's a good point on the voting issue. I was just trying to brainstorm ways to keep the parliament fresh and dynamic, but there are probably better ways to accomplish that. A committee structure is certainly a good idea, but we should probably leave it up to the parliament to self-organize into whatever committees it deems necessary.--Danaman5 (talk) 16:11, 23 July 2009 (UTC)

You could have votes expire after a certain amount of time, that would stop people staying in the parliament for a long time simply because no-one has bothered to change their vote. It would require a concious decision from voters to keep them there. While the details of what committees they have would be a decision for the parliament, I think the general decision making process (eg. First Reading, Committee Stage, Final Reading) should be set down in the proposal. --Tango (talk) 16:30, 23 July 2009 (UTC)

Agreed. Should we create a policy proposal page then?--Danaman5 (talk) 19:59, 23 July 2009 (UTC)

Really good fleshing-out. I'm liking this idea more. --Merovingian (T, C, L) 03:02, 9 August 2009 (UTC)

This review doesn't seem to have got the activity I had hoped, I suspect at least partially because of the controversy over the committee ArbCom tried to create distracting people from it at a critical time. Therefore, it is probably best to abandon my proposed process for it and just pick the best ideas that have been suggested and write up proposals. This is the one I'm most attracted to. I'll give it some thought over the next few days, I'd appreciate any comments people have about it. After that I'll have a go at writing it up. --Tango (talk) 20:55, 13 August 2009 (UTC)

Lottery: King for a Month

Replace Jimbo with a randomly selected active user. At regular intervals, hold a new lottery. In my university, that's how they selected department heads. Jimbo's official role is minimal, so what's the harm? It should just be a non-democratic head of state who holds a reset button on the democratic system. Most of the time, you will get apathetic users who can't be bothered to even get informed on how to exercise their power. Once in a while you'll get a quirky guy who triggers a paradigm shift. Very rarely, you'll get a lunatic, and we'll just disobey him - after all, he only got the position by lottery, and only for a short time, so he's unlikely to garner the respect needed to perform a coup.

You see, if you elect your head(s) of state, it's just a democratic check on democracy. No matter how you set it up, it cannot solve a situation where democracy runs amok. If you choose an autocratic leader, the system may become inflexible and has a single point of failure. A lottery gets the benefit of both side. You get the added benefit that, over time, it provides true representation of the membership, better than democratic rule of the majority can provide.-Yannick (talk) 04:11, 25 July 2009 (UTC)

Discussion

This was sometimes done in ancient Greece. Peter jackson (talk) 09:56, 25 July 2009 (UTC)

Yes, I believe a lot of offices in Athens were filled by lottery. I suspect the university mentioned chose their heads of department at random from a fairly small number of senior members of the department, though. I don't think giving this kind of power to an inexperienced member of the community would work. Therefore, we have to decide who would be eligible, and that is going to be arbitrary and controversial. --Tango (talk) 20:49, 13 August 2009 (UTC)

Small Steps. Small Changes

I havent studied all this indepth, but from what i can see, proposed changes to governance keep getting deadlocked because of no-consensus. Maybe this is because there is on one hand a group that wants significant reforms, and on ther other hand a group that wants to keep things the same, and these two viewpoints are so-far apart they cannot agree on anything. So, why not compromise? What about small, simple and clear proposals, where the proposal is split up in as many seperate sections as possible. And keep it to small changes. Small steps. One step at a time. Omegastar (talk) 19:32, 11 September 2009 (UTC)