Wikibooks talk:General voting rules/Proposal

I started this policy as a response to some of the criticisms that were mentioned on the talk page of the other proposal. --Whiteknight (talk) (projects) 17:11, 12 July 2006 (UTC)


 * I think this page is going in a more positive direction and certainly is a huge improvement over the previous voting policy. In this case, however, it should be moved (IMHO) to a guideline status instead of necessarily an enforced policy, as the only thing that really can be used in a solid metric is the suggestion that discussions should trail off and be left alone for about a week before the decision is finally reached.  From my own experience on Wikibooks, it usually takes about six weeks to get every most semi-regular Wikibooks users to be able to come and look at a particular issue and voice their opinion on it.  If you make decisions in less time than that, some people will be left out of the process.  If you wait longer, it seems as though nothing is happening at all.  Of course it has taken more than a year to get Wikiversity going and a decision made about that.  --Rob Horning 18:11, 27 July 2006 (UTC)


 * Well, the wikiversity decision is not an internal wikibooks matter, but has been (if i understand correctly) delayed by the WMF. We could change this to say that votes should last at least 6 weeks. I dont think, however, that this should be just a guideline. By making this an enforced policy, we are saying that community concensus is the official method for making decisions here. If we just make this a guideline, people can say "We have a majority vote on this issue, and concensus is just a guideline, so we will declare ourselves the winners". The beauty of this policy, in my mind, is that it set's down the correct method of operations without creating any beauracracy, additional complications, or confusion. It's vague enough to be interpretable, but strong enough to protect minority opinions from a majority vote. --Whiteknight (talk) (projects) 12:14, 28 July 2006 (UTC)

Tackling the nature of consensus as a separate issue
The process of voting is a different issue from the problem of majority votes etc. I favour the way the current proposal avoids this problem for the moment. Once the process is agreed then brave souls can venture upon the perilous ground of discussing majority votes etc. RobinH 09:19, 28 July 2006 (UTC)


 * Luckily, majority votes only really need to occur on BOTM and COTM, which come with their own voting rules. Everything else that i can think of really deserves concensus, so there is no reason to include information about majority votes here. At least, that's my opinion. --Whiteknight (talk) (projects) 12:08, 28 July 2006 (UTC)

Voting Time
I realize that Wikibooks is a much quieter place than some other wikis, but to have a vote sit for two weeks after the last individual vote is cast seems like a really long time to me. A month for voting I like, but to have it sit without votes or for archiving seems way too long. I figure the last vote before a decision can be maybe a week, and the archive decision be maybe three days. I don't see a need to keep a vote out when someone can just access the archive immediately. -Matt 12:54, 28 July 2006 (UTC)


 * That does make good sense, although as a practical matter, because we are such a small community, things don't always get moved or archived in a timely manner. I'll reduce the time restriction here, because there is no particular reason to wait a week to get it archived. --Whiteknight (talk) (projects) 13:07, 28 July 2006 (UTC)

Rewrite
I have rewritten the proposal - though it is still heavily based on Whiteknights' and RobinH's draft. The changes are here and the version before my changes is here.

Largely I have sought to make it more specific - in particular by identifying explicitly where we currently use voting on Wikibooks. I have also imported the voting rules from BOTM and COTM as it seems reasonable to have some short qualifying period before someone is eligible to vote. I would not have thought new users would find that particularly problematic. There is a carve out to allow new users to comment on modules and pages they themselves have actually edited (so that if their book is listed on VFD they can still outline why they believe it meets our inclusion criteria, even if they have not as yet met a 20 edit minimum).

There are a few other minor changes. As ever, comments from others would be helpful. One point that might be worth discussing in particular is - should every new policy/major change in the practical effect of an existing policy always be affirmed by vote, or is it enough to note that there are no strong dissenters? Jguk 07:38, 3 August 2006 (UTC)


 * This is all well and good until a vote is needed to resolve an issue where people take fixed positions. The lack of any voting policy means that this issue cannot be resolved. RobinH 23:14, 3 August 2006 (UTC)


 * I like this rewrite, although I worry that perhaps it gets too specific in some places. It was a nice touch to list the various areas where a vote can occur, to help ground this policy in actual occurances. I don't necessarily agree that a user must have a minimum of 20 contributions to cast a vote (except where already specified), but that isn't a significant-enough sticking point for me to abandon this policy. I would like to note, that I've spent some time on wikipedia, looking at their policys in this area, and I found that wikipedia does allow for a supermajority decision when concensus cannot occur. They do not, however, give a specific percentage to define a supermajority.
 * I do think that we should reinclude the fact that votes should be properly advertised, because it is impossible to declare "community concensus" when most of the community wasn't invited. Except for "Circumstance A" votes (which are always happening, and don't require special notification), all pending votes should be advertised on staff-lounge or community portal, or the mainpage, or somewhere else appropriate. I think there are a few specific points that we should emphasize, and that a general voting rules policy cannot do without:
 * All users (or most users, barring an edit requirement) are allowed to vote in any discussion
 * Votes are decided by concensus, with few noted exceptions (COTM, BOTM, WJBOQ)
 * Concensus is poorly defined by this policy on purpose, and is not defined by a percentage, or a "majority rules" (unless we allow for indesputable super-majorities).
 * So long as we emphasize these points, I can't imagine that the specific text of the policy are as important. --Whiteknight (talk) (projects) 00:10, 4 August 2006 (UTC)


 * Also, I just had the thought that perhaps we could include the suggestion, not a requirement, that if a decision is needed and a concensus is not obtainable within an acceptable timeframe, outside arbitration can be asked for (again, assuming all voters agree to this). perhaps this is just a pipe-dream, however. --Whiteknight (talk) (projects) 00:13, 4 August 2006 (UTC)


 * Votes seem to fall into three categories:


 * 1. Market research ie: book of the month etc.


 * 2. Elections


 * 3. Decision making


 * The market research types of votes do not really need a policy. They are trivial and have been managed in Wikibooks for years without a formal policy. Fortunately, on the decision making front, there have been no really major disputes to date. I think there should just be a note in this voting policy that users undertaking market research can set their own rules. RobinH 10:02, 4 August 2006 (UTC)


 * I agree that we don't need to talk much about the "market research" stuff, because those places are already well established, and have a predefined set of voting rules to follow. I also want to point out that I made a few minor changes to the text of this page, to move the section on "concensus" towards the top of the page, and to put more emphasis on that point. As far as i am concerned, the topic of concensus is the most important topic that we can discuss in this policy. --Whiteknight (talk) (projects) 14:38, 4 August 2006 (UTC)

I have removed "except in the case where a minority is determined to block a policy and maintains the discussion for months to prevent a conclusion" and "If a vote has been advertised three times over a period of six months and then blocked by the same individual or minority on each occasion then, on the fourth cycle, a vote may be taken and implemented." See to see the changes in context.

I removed the first clause because the point is already covered more aptly by referring to "substantive ongoing discussions". Discussions that are needlessly strung out are not "substantive". I removed the sentence as I do not think any one person should have a right to "block". Yet I do think that 49% of Wikibookians being opposed to a policy proposal should be enough to block, even if 51% of Wikibookians have voted for it four times.

The aim is to seek consensus. Bona fide concerns may be raised as part of those discussions. All of those should be addressed (maybe by amending the text, maybe by the proposer explaining why he does not consider the other person's concerns to be concerns). You should be able to either agree a text by this method, or failing that agree that all concerns have been properly discussed, albeit not with conclusions to everyone's liking. Only then should there be a poll (if one is thought necessary by anyone).

I have been considering adding a bit to suggest, where there are ongoing discussions, that it is often helpful for a dissenter to be invited to say how he would change the text to meet his concerns. It is a useful way to help break the impasse. It forces the dissenter to specify which bits of the proposal he is willing to accept, and which bits he thinks should be changed. This focuses the debate, and you may well find that the proposer is willing to accept those changes to get the policy passed unanimously, Jguk 07:01, 7 August 2006 (UTC)


 * The current policy allows a single user to prevent a vote. A time limit on spoiling the voting process is required. I suggested six months and have reinstated the sentence that described the maximum delay. Although you removed the second clause you did not seem to address the problem of blocking the voting process above, you raised the problem of majority voting but the second clause is about blocking the voting process itself. As the policy stands a single user can prevent any decisions that use voting, in any part of Wikibooks, forever. RobinH 08:39, 7 August 2006 (UTC)


 * I don't think so. The current policy mentions that all discussions should reach a general concensus, which is not the same as 100% agreement. small minorities with stubborn disagreements can almost be glossed over entirely, if they completely refuse to compromise or reach concensus. Concensus and compromise, by this policy, will become the official way of doing things, and people who don't strive to reach a compromise are essentially removing themselves from the discussion entirely. The same would apply, i think, to a large majority, that stubbornly refuses to compromise with the minority: As soon as you abandon the ideals of concensus and compromise, you abandon your right to be counted. This may sound a little hard-line, and I dont think that text like this should make it into the policy proposal. However, we do need to remember that "consensus" does not mean "everybody always agrees 100% and is perfectly happy". People who won't compromise, and won't work towards a satisfactory conclusion are in violation of this policy. --Whiteknight (talk) (projects) 13:10, 7 August 2006 (UTC)


 * You say that "small minorities with stubborn disagreements can almost be glossed over entirely". I think we need to specify how this glossing over might occur. Similarly for the majority that refuses to compromise. This was why I put in a 6 month period for reaching agreement when discussions are deadlocked in this fashion. RobinH 13:17, 7 August 2006 (UTC)

Robin, with apologies for the formal style, I set out below why I don't see your proposed additions as suitable:


 * 1. I do not agree that any user or group of users has a "block". I believe every user acting in good faith should have his concerns answered. That answer may be an amendment to the text - it may be comments as to why those concerns are not shared. In short, it does not have to be an answer he agrees with. Ideally some compromise would be possible, but if not, each users' concerns should still be answered.


 * 2. If a user just says "oppose", it is reasonable to ask that user to explain why, answer those points and give that user an opportunity to make a further response. After that, you can ask that user to make changes to the proposal to make it acceptable for him, or to acknowledge that regardless of any chances, he will not support it. If the former, these can be discussed properly. If the latter, it is fair to say that you intend to go to a poll, that user's comments will be available to those voting - if other users agree with him, the proposal will fail, otherwise it will pass.


 * 3. If substantive discussions appear to be going slowly, and this has not happened yet, invite the other user to amend the text so that he would support it. In all likelihood you'll find some points you can accept and other points you find good but can improve on further. At the very least, the points of disagreement will be highlighted. If it is clear there will be no unanimity over a small amount of policy, then it is reasonable to suggest letting the community settle the point.


 * 4. If discussions become off-topic or repetitive, it is unlikely that they are substantive anymore. It is reasonable to ask your interlocutor what points of outstanding disagreement have not yet been discussed (as opposed to discussed but not in a way which he likes). Address those issues an then say you will go to a poll. If others agree with your interlocutor, the proposal will fail, otherwise it will pass.


 * 5. In summary, there are no "blocks" - an idea that your wording would introduce to Wikibooks. There are just discussions for as long as they are productive. If they cease to be productive it is because (1) you have reached agreement; or (2) you will not reach agreement. In the latter, either the community as a whole is divided, in which case the proposal must fail; or alternatively, there is (are) just one (or two) dissenter(s), in which case the supermajority can push through the policy in the knowledge that the dissenter(s) have had fair opportunity to argue their case, Jguk 14:45, 7 August 2006 (UTC)


 * I'm in pretty whole-hearted agreement with this. Nowhere does this policy state that every decision must acheive 100% agreement. the idea of a supermajority then is implicit in this, but by not providing a specific percentage that defines "supermajority", and by not strictly defining how many voters represent a "consensus" we are leaving this policy open to interpretation. Wikibooks has operated, for better or for worse, without any voting policy for over 3 years now. A new policy, in my opinion, shouldnt introduce any new rigid guidelines or specifications that we haven't been using up till this point. By formally specifiying concensus as the way things work here, we are simply removing the ability for a 51% majority to declare itself "the winner", and we are requireing that opposing viewpoints be considered and discussed before any decision can be made. Creating a gigantic, strict, voting apparatus would be counter-productive here. --Whiteknight (talk) (projects) 15:05, 7 August 2006 (UTC)

Changes to RobinH's last version
Following my points above, I have made some changes to the last version by User:RobinH. Those changes can be viewed here:. Specifically, i made the following changes: I hope these changes aren't too drastic, and they can be rolled back if they are not satisfactory. --Whiteknight (talk) (projects) 15:21, 7 August 2006 (UTC)
 * 1) I removed references to a majority decision being required. Instead, I point out that stubborn users who refuse to find compromise are in violation of policy, and are voting in bad faith.
 * 2) I removed the "three times" rule, because it seems to me that a discussion can not be declared "over", the discussion and compromise stages are simply dragging on longer. A discussion should be permitted to take any length of time, so long as forward progress is being made, and all users are offering compromise.
 * 3) I state that concensus cannot be declared until all good-faith dissenting opinions are addressed, and a suitable compromise can be made
 * 4) I state that the 20 vote minimum is mostly a guideline, and that exceptions can be made if the discussion is in good faith.


 * I am happy with ensuring respect for small groups and trying to obtain a consensus compromise but as it stands the policy invites users not to compromise - why compromise when you can simply block all further progress? A user who blocks a vote on a project may believe they are doing this in good faith, they may believe that it is the world that is wrong and refusing to compromise.


 * Do you have any proposals for overcoming this problem? In primitive social organisations the King was used to break such deadlocks, as societies became more advanced a group of Lords would exercise this power, however, in Wikibooks the user who is involved in blocking a project is as likely to be an admin as an ordinary user. In yet more advanced societies the constitution or law resolves such problems. RobinH 09:13, 8 August 2006 (UTC)


 * I have to be honest with you, i dont think that the current text of this proposal allows for a "block". A single user who won't compromise, regardless of how right he thinks he is, is still not compromising. Nowhere does it say that a decision must make everybody happy. Nowhere does it say that a decision requires 100% support. All this policy says is that an attempt must be made to satisfy all dissenters. If they refuse to be satisfied, that does not indicate that no attempt was made. I do not think this is an issue, and I do not think that this policy supports such a situation, nor allows a single user to block a vote indefinately. --Whiteknight (talk) (projects) 13:07, 8 August 2006 (UTC)


 * My problem is that I cannot see a method of preventing a single user from blocking a vote or the outcome of a vote anywhere in the policy. I believe this will encourage any user who spots this flaw to "block" any discussion that they oppose, either indefinitely or until their view is accepted (accepted simply to allow the discussion to progress). Can you explain how, if an admin user decides to simply block a discussion, the block might be overcome? At present a gang of admins could simply threaten the other user but this seems very uncivilised and unconstitutional.  RobinH 10:02, 9 August 2006 (UTC)


 * The current text states that "concensus" is not 100% agreement. This means that a vote could pass if it were 6-to-1, or something like that. All that is required is that an effort is made to compromise. Whether that effort succeeds or not is of no consequence. A lone user cannot, therefore, block a vote, assuming enough of the community is united in passing the vote. Admins do not have any extra authority in these matters. All an admin has is the extra permissions to block users and delete edits. This does not give them the authority to block a vote, or single-handedly alter policy. --Whiteknight (talk) (projects) 14:25, 9 August 2006 (UTC)


 * I an sorry to stick at this but there seems to be nothing to stop a user with an axe to grind just saying that they do not feel a consensus has been reached. The only resort of the other users would be to hold a vote on whether the dissenting user was correct. The dissenting user can then maintain that there is no consensus on taking the second vote. The policy as it stands does not deal with this impasse and positively invites users to exploit it! A lone user can block progress but a small, active group would be even more effective.


 * The solution to this problem is to specify that, after a certain amount of discussion (say six months and three attempts at voting) a majority decision is accepted.


 * I introduced the point about admins because a gang of admins could indeed ban a user who refuses to compromise and blocks a vote endlessly. This would be wrong but it is the only remedy available to stop a persistent user from blocking a vote under the policy as it now stands. It would be better to have a voting policy that gives a clear indication of when a majority vote is applicable. RobinH 15:45, 9 August 2006 (UTC)


 * I definately understand your reservation, I just don't see how such a problem would even present itself. The opinions of the individuals don't matter as much as the general opinions of the community. If a single user, faced with community oposition states "consensus has not been reached, because I still dissent", that is not the correct use of the word "consensus" --Whiteknight (talk) (projects) 19:01, 9 August 2006 (UTC)


 * Like Whiteknight, I strongly doubt such a circumstance would arise. If it did, I imagine members of the community would first tell that user that they believed what he was doing was wrong, and then if that did not stop it, block for a short period of time. Come the end of the day, any user decides to stop editing constructively and just ignore community decisions, they will be chucked off the site. This is an unwritten rule already. If necessary, Jimbo or a steward would come in and de-admin. Commonsense would prevail. As I note though, I really can't see us ever being in a position where we'd have to go that far, Jguk 19:42, 9 August 2006 (UTC)


 * There you go: "come the end of the day, any user decides to stop editing constructively and just ignore community decisions, they will be chucked off the site". This is precisely what bothers me. Who will chuck them off the site and on what authority? Clear rules avoid this uncivilised behaviour. What both you and Whiteknight are saying is that there is a point at which dissent becomes destructive. When framing policy we should identify this point and avoid the summary justice that will otherwise occur. RobinH 23:04, 9 August 2006 (UTC)

I don't agree that a dissenting user could be "chucked off the site". It is conceivable that an issue could arise that will polarize a minority against the majority, and a "stalemate" can ensue. Now, assuming that all the users are casting opinions civily, and making at least a nominal effort to reach concensus, the issue can be discussed ad infinitem until it is resolved satisfactorially. Now if a user is not acting in good faith, and is intentionally being disruptive, that's a different story entirely. Wikipedia runs off a very simple consensus model, with only the addition of available 3rd party mediation, if necessary. Wikipedia's policy can be found at wikipedia:Wikipedia:Consensus. Here is an excerpt that I think is demonstrative:
 * Precise numbers for "supermajority" are hard to establish, and Wikipedia is not a majoritarian democracy, so simple vote-counting should never be the key part of the interpretation of a debate. (That does not mean that voting should be entirely discredited; it can be quite useful.) However, when supermajority voting is used, it should be seen as a process of 'testing' for consensus, rather than reaching consensus. The stated outcome is the best judgment of the facilitator, often an admin. If there is strong disagreement with the outcome from the Wikipedia community, it is clear that consensus has not been reached. Nevertheless, some mediators of often-used Wikipedia-space processes have placed importance on the proportion of concurring editors reaching a particular level. This issue is controversial, and there is no consensus about having numerical guidelines. That said, the numbers mentioned as being sufficient to reach supermajority vary from about 60% to over 80% depending upon the decision, with the more critical processes tending to have higher thresholds.

Another wikipedia policy, wikipedia:Wikipedia:Straw polls has this to say about voting processes while attempting to reach a consensus:
 * Even if a large number of people vote for one option but some don't, this doesn't mean that that's the "outcome". It means some people are disagreeing, and that has to be addressed.

Another thing that is said in the consensus policy that (i think) specifically addresses User:RobinH's objections is as follows:
 * It is difficult to specify exactly what constitutes a reasonable or rational position. Nearly every editor believes that his (or her) position is reasonable; good editors acknowledge that positions opposed to their own may also be reasonable. But Wikipedia's consensus practice does not justify stubborn insistence on an eccentric position combined with refusal to consider other viewpoints in good faith.

Now, I'm not saying that wikipedia should be our cookie-cutter for policy, but I do want to point out that wikipedia has been doing business (very successful business, might I add) with only a simple, vague voting policy that broadly specifies that discussions are settled by common consensus. Wikipedia does include additional mechanisms for reaching consensus, such as "Requests for comment" "Surveys" "neutral mediators", etc. but those are just alternate methods for continuing a very large discussion, and wikibooks doesnt need to worry about discussions of the same size as wikipedia has. I hope some of this helps. --Whiteknight (talk) (projects) 01:13, 10 August 2006 (UTC)

And as an addendum, while I am thinking about it, I can't see a situation where this would matter. BOTM and COTM are decided already by majority vote. RFA decisions are consensus-based, but are ultimately decided by the bureaucrats who need to perform the "sysopification" (acting in accordance to community consensus, if it can be shown to exist without doubt). Alterations, additions, or omissions of wikibooks policy are never so urgent that a prolonged discussion cannot take place. Situations where there is a dispute over module content are rare here (although they could get more common, as our numbers grow), although temporary page protection can prevent edit wars, and a page can remain protected until a decision can be reached. --Whiteknight (talk) (projects) 01:41, 10 August 2006 (UTC)


 * Policy is an area where this sort of blocking could occur. It could also happen in discussions on editorial content. Fortunately Wikibooks is reasonably civilised at the moment with many users committed to the project as well as their own contributions. If we succeed in our dreams for Wikibooks then we will be at far greater risk from problems such as blocking. I would guess that in about 4 years Wikibooks will be a truly major site with external organisations focussing on how evil we are to offer books on evolution, women's rights etc..  If we wait until a major violation of good faith occurs we may be unable to change policies to deal with problems.


 * You and Jguk both acknowledge that the current wording allows blocking and both of you agree that this would be a bad thing. You also both agree that without any wording to cover blocking the available responses are unsatisfactory. Why not add some wording to deal with blocking? If the risk of blocking is low over the foreseeable future then the wording will not be needed, if the risk is high then we will have done everyone a service. Wording to prevent blocking is a win-win position so why not allow it? RobinH 09:47, 10 August 2006 (UTC)


 * Wikipedia has an arbitration committee of a number of admins who decide these things. However, their processes have a number of downsides: the very existence of a high-profile dispute procedure attracts people to it. I get the impression that if there was no such committee, many disputes would be resolved well before it got to that stage. To give an example, somewhat perversely their ArbCom recently ruled that it is acceptable to keep user subpages analysing another user's behaviour. Yet the only possible reason to do this is that the user preparing those subpages has no intention to discuss or compromise on the dispute, but instead to bring a dispute to ArbCom in the aim of banning another user. I hope we never have policies here that promote the escalation of disputes.


 * However, their ArbCom does offer some benefits - it certainly defines the pinnacle to which disputes can be raised. Looking at the suggestion of an ad hoc committee and comparing it to ArbCom - I would note that if we were to have reasonable numbers of people on that committee, and to select it in such a way that doesn't give the impression that we are trying to determine the result of any action in advcance, would mean that every active admin on Wikibooks would be on such a committee.


 * The meat of the question is, however, how do we deal with a disruptive user (as any user who bluntly refuses to accept a community decision will very swiftly become)? To my mind it is irrelevant as to whether the user becomes disruptive following a vote or for any other reason not covered by this policy. And current policy in practice is that, if disruptive users do not return to editing non-disruptively, they leave the site until such a time as there is confidence that they will not edit disruptively in the future.


 * I therefore see no need to make reference to trying to block a decision from being made or users refusing point blank and repeatedly to accept community consensus in this policy.


 * Robin, if I may put my dispute resolution approach into practice, may I note that I see it as unlikely that you will agree with Whiteknight and myself on the wording in this proposed policy in this regard. This leaves open a number of options: 1) you could agree to step aside now (I'll be upfront and say, particularly bearing in mind Whiteknight's comments, that I will not agree to step aside now); 2) Whiteknight, you and me could leave this discussion a while to allow others to comment - and indeed, ask others to comment - and then reconsider our position in the light of any comments; 3) we could look towards proceeding to a vote - either offering only a version without your suggested wording, or alternatively offering a choice between our wording, your wording or no policy. I'd be grateful if you'd indicate which option you'd prefer - or alternatively, if you have another suggestion, let me know what it is. Kind regards, Jguk 11:44, 10 August 2006 (UTC)


 * You will notice that I have not been changing the text of the proposal but have been discussing in good faith here, on the talk page. Whiteknight realises that there is a problem and has proposed mediation/arbitration as a possible way out. You have also agreed that there is a problem and have suggested that if a gang of admins can decide that a user is blocking a decision unnecessarily they should ban the user. I prefer Whiteknight's approach because it is not entirely summary justice.


 * That said, I am not really in favour of either judicial approach. All that is needed is an extra rule in the proposal that allows up to three blocks over a period of 6 months and then accepts that a majority decision must be taken. The policy should note that this is only a course of last resort and a single vote is preferred. Such a policy would avoid arbitrary interventions by admins. RobinH 13:30, 10 August 2006 (UTC)


 * First off, I want to be very clear right now, that despite what I may have said inadvertantly, I do not believe that the current text of this policy allows for a blocked vote. I suggested mediation/facilitation/arbitration (or whatever it should be) as an attempt to compromise on this situation, but I realize that suggestion is only going to create another policy proposal that we all aren't going to agree on.
 * The addition of a rule that says "up to three blocks over a period of 6 months and then accepts that a majority decision must be taken" requires that we specifically define what a "block" is. What is a block, exactly? Is it simply a matter for which we don't have 100% agreement? Consensus doesnt require 100% agreement, so people disagreeing with a policy doesn't constitute a "block". Is a block a large percentage of dissenters? If 40% of the people are dissenting, we don't have consensus, even if we do have the numbers to pass a majority vote. We cannot be making policy, or deleting pages, or electing admins on a 60% majority. I therefore don't consider a large percentage of dissenters to constitute a "block" either: if that many people disagree, the motion has decidedly failed completely.
 * If one person disagrees with the community at large, it's tough beans, but the community wins. Now, there has been text added to this policy that states that a genuine effort must be made to appease the dissenter(s). If an effort is made, but the dissenters cannot be appeased (due to stubbornness, disruptiveness, eccentricity, whatever), again it's tough beans: The community still wins. Notice that I am saying "the community" here, and not "the majority". At no point to we resort to a majority vote, but instead we accept that "if all reasonable people want the vote to go a certain way, and go about it reasonably and respectfully, then they should get what they want.
 * With that said, I don't think that me, Jguk, and RobinH talking amongst ourselves is representative of the "community" here at wikibooks (although good points have been made by all). I'm going to go down the list of active admins, and send each one a message that there is an important vote going on here, one that could (with the addition of arbitration or mediation, or whatever) affect the admins directly. I will make the note as short, and as unbiased as I can, and I will send it to User:Jguk and User:RobinH first, to ensure that it is unbiased. Hopefully we can get enough people in here to decide this thing one way or another. --Whiteknight (talk) (projects) 17:58, 10 August 2006 (UTC)


 * A "block" is where one user or a small group declare that consensus has not been reached and that they are acting in good faith by prolonging discussion and overturning votes indefinitely. The current wording of the policy permits such filibustering. In fact it encourages it!


 * You say "If one person disagrees with the community at large, it's tough beans, but the community wins". Please can you put this in the wording of the policy if this is what you believe. What about 2 dissenters?


 * You say "The community still wins" but dissenters are part of the community. What fraction of the community is the "community at large" (ie: non-dissenters) and how do you measure this? You have already said that where n people vote it is n-1 people, but what if there are only three voters?


 * These questions are central to this policy. If we do not answer them we will end up with each contentious vote actually being decided by the most influential or active admin. Perhaps that is what you and Jguk want but if so please state it clearly. RobinH 08:37, 11 August 2006 (UTC)


 * Robin, we've been over this several times. In your example, the small group cannot simply declare consensus has not been reached. That's not how it works. They have to set out their objections to the policy or decision. The community addresses their criticisms and hopefully comes up with a better proposal. If the community decides the small group's criticisms aren't significant enough to change the proposal, the decision can be enacted over their objections. So long as the criticisms have been heard and addressed (either as "okay, we'll fix the proposal" or "eh, the criticism is minor/wrong/etc"), the community can enact the decision. There is no point at which the small group can then declare the process invalid. If they do attempt to declare the process invalid (just by saying so), the community should easily be able to distinguish between their good faith objections and their bad faith monkeywrenching.
 * In normal consensus decision making, a block is a last-ditch attempt to stop a proposal. When it occurs, the person blocking is putting their reputation and working relationships on the line. In blocking, they are saying that they believe that if the proposal is passed it would be detrimental to the organization and that they are willing to leave the organization if the proposal is enacted. This is very serious as it indicates a breakdown in the working relationships in the organization; the concerns raised by the blocker should have been addressed more fully and completely earlier in the process. In any case, the organization can still enact the proposal over the blocker's objections. This person would usually then leave the organization as there has been a fundamental disagreement over the purpose and direction of the organization. This is okay.
 * At wikibooks, someone could declare the same thing as a blocker, but because we don't have as close of working relationships, because we have a large anonymous group, and because we have open membership with no requirement of buy-in on certain ground rules, this kind of statement carries much less weight and users are more likely to pass a proposal despite a statement of this sort. In effect, there is not a real way to block on wikibooks, at least not in terms of formal consensus procedures. If decisions are made with which users fundamentally disagree, can work to change those decisions, or they can leave. Kellen T 10:04, 11 August 2006 (UTC)


 * You are using the word "community" in place of the word "majority". If 4 out of 10 voters had a particular view the other six would by called the "community". Suppose we stick to terms like majority so that we both know what we are talking about and dont exile a user from the "community" just because they hold a minority viewpoint on a particular issue.


 * What you are saying is that if a minority resists the "community" (ie: majority) decision then they are laying their reputation on the line and might be asked to leave Wikibooks.


 * Why not avoid people putting their reputation on the line by saying that a majority decision rules - that is, after all, what you have said above (when all the vicious bits about exiling dissenters are removed).


 * My objection to this proposal, as it stands, is purely in the way that it talks about consensus. Any outsider understands consensus to mean general agreement between all parties and this proposal suggests that a vote cannot be decided unless there is a consensus. The current wording encourages blocking and does not say how a vote might provide a decision. We have now had four alternatives for decision taking:


 * 1. Whiteknight: a special admin intervenes
 * 2. Jguk: an ad hoc group of admins intervene
 * 3. Kellen: the dissenter falls on their sword
 * 4. RobinH: if consensus is not obtained after 6 months the majority decision is allowed.


 * Which do you prefer? RobinH 18:55, 11 August 2006 (UTC)
 * It's not about exiling dissenters. The options are open to the dissenter; they can choose what they feel the appropriate response is. If they feel strongly enough about a specific position on an issue that the community disagrees with, they probably shouldn't be contributing here. They aren't required to leave, but they are required to abide by the community consensus if they want to contribute. Existing examples where a contributor should "leave" (i.e. not contribute): If a user disagrees that content should be NPOV. If a user doesn't want to work cooperatively on a text with others ("sole author" syndrome). Kellen T 21:52, 11 August 2006 (UTC)


 * You dont need to answer this question, I have just read your own policy proposal and you prefer 1. On ordinary issues using a majority vote. 2. On high impact issues voting once, then, if there are objections, voting a second time and going with the majority. This seems fine to me - why didnt you say this was what you thought?! RobinH 19:21, 11 August 2006 (UTC)


 * "Community" == "vast majority". Not "simple majority" which is where we get into disagreements/misunderstandings. I'm pretty sure I explained a couple times that dissenters can be overridden by the community position, but that this should only be done after several rounds of feedback and criticism have been addressed in the proposal. Kellen T 21:52, 11 August 2006 (UTC)

Interwiki Voters
Another topic that I just remember is what I call "Interwiki voters". This is when people from other projects come in, and cast a vote. For example: "I'm User:AwesomeDude03 from en.wikipedia, and I think we should keep this page because I like it LOL". Do we count votes like this? What if they have less then 20 contribs here, but a significant number on their home project? I don't think we should count votes like this from people who aren't wikibooks users. --Whiteknight (talk) (projects) 17:52, 4 August 2006 (UTC)
 * In my opinion we should never, ever allow these kinds of votes. I almost want to discriminate users who may have the "20 edit count" but write all over their talk page that they're just from Wikipedia. The Muggles' Guide was VfDed days after its conception from some Wikipedia "person" that never even edited here before. Fighting back things like that is especially important to me because it's stopping another way Wikipedia pushes itself onto other projects. A great user here, Kernigh, seems to have left Wikibooks due to things like that. Until other Wiki projects gain respect for all others, I think we need a lot of discretion towards the weight of an inter-wiki voter's vote. -Matt 14:18, 5 August 2006 (UTC)
 * I highly agree with this. I don't think that people, no matter how many wiki-edits they have, should be able to vote here without having even the slightest idea of wikibooks policy. Alot of garbage gets brought here from wikipedia because they dont know what we are all about, and then wikipedians come here to defend the action, without ever referencing our policy, or considering that wikibooks policy is different from wikipedia policy. --Whiteknight (talk) (projects) 17:43, 5 August 2006 (UTC)

I appreciate that it was me who suggested expanding the minimum 20 edit rule beyond the BOTM page (with an exception to allow editors of modules to argue their case on votes connected with them). However, I'd also hope that it is the spirit, rather than the precise wording, of the proposed rule that is followed. If a user with fewer than 20 edits makes a comment pertinent to the subject at hand, it should be left standing so that we can see if the point persuades Wikibookians. If a user with fewer than 20 edits makes a nomination on VfD, that nomination should be left standing if, in a more experienced Wikibookian's opinion, it is worthy of discussion - although it should be removed if (probably quite innocently) it is misguided.

To my mind the 20 edits minimum is to allow the removal of clearly unsuitable or off-topic comments/proposals and to give us a little bit more protection against sockpuppets (as there would be no need to prove a suspect account with fewer than 20 edits to be a sockpuppet to ignore it). After all, apart from BOTM, COTM and Wikijunior BOTQ, no other votes are decided purely numerically anyway: the occasional not-strictly-qualifying commenter is not necessarily a problem. On the other hand, if a user has more than 20 edits and is making inappropriate proposals and off-topic comments, we should not be scared of saying that, Jguk 06:44, 7 August 2006 (UTC)


 * Sounds like a good policy - as long as it makes clear that we welcome comments and suggestions, but also makes clear that we're not obliged to count votes by people who haven't done a certain amount of editing.


 * A more general comment - I'm not a policy buff so don't expect too much comment from me, but I think it all looks pretty good. Good work, people. --Singkong2005 07:44, 7 August 2006 (UTC)


 * I removed the previous exception about the 20-vote-minumum rule, and replaced it with the following text:
 * The 20 vote minimum should be counted as a general guideline, and not as a strict rule. Users with few edits cannot be expected to fully understand the workings of wikibooks policy, and are occasionally assumed to be voting in bad faith.
 * Perhaps this addition is too wishy-washy on my part, but it allows for exceptions to be made, and decisions based on policy interpretation, not the following of strict numeric rules. Users with fewer then 20 edits, assuming they have something meaningful and relevant to say, should be allowed to discuss issues. I don't think, however, that a user with fewer then 20 votes should be able to start a vote (and by extension, nominate a VfD, nominate an admin or request adminship, etc). I have not added text to that effect to the policy, although i think it is worth some consideration.

Facilitators/Mediators
As i mentioned above, wikipedia has a certain allowance for neutral mediators/facilitators. I wonder if we should add in a clause that a neutral, mutually respected third party can optionally be called in to determine if there is general concensus or not. This third party, essentially, would only read the arguments, and determine if there is concensus one way or the other, and not decide the vote single-handedly. For instance, if there was a particular discussion where 27 people are for action A, and 3 people are in favor or action B, a third party would come in and say "There is consensus for option A". If there were 18 people in favor of A, and only 12 people for B, the mediator would say "There is no clear consensus". This is just an idea, however, I haven't added any such text to the policy proposal. --Whiteknight (talk) (projects) 01:19, 10 August 2006 (UTC)


 * I proposed an ad hoc committee of administrators for this type of work (see Ad hoc administration committee). A small committee would be a fairer option than one judge. Unfortunately this was rejected. RobinH 10:16, 10 August 2006 (UTC)

Whiteknight, that is not mediation or facilitation, it is arbitration. In mediation, the mediator takes no view, but encourages the parties to come to a middle view. In facilitation, the facilitator helps the disputants argue through the case. Arbitration is where someone (or some people) from outside make a decision about a dispute.

On the substance of your suggestion - if parties can agree on who the outside arbiter would be and agreed to be bound by that arbitration, then I have no difficulty with it. Better still would be to agree a short list of potential arbiters (five, say) before any such dispute arises, and then to ask the next arbiter in line to make the decision (with a provision to move on to the next arbiter if no response is received within 168 hours). The arbiter's job would be limited to saying whether there is consensus and, if so, what that consensus opinion is. If we could get stewards to agree to the role (which should not be too time consuming) so much the better, Jguk 11:53, 10 August 2006 (UTC)


 * Ad hoc administration committee was almost exactly this proposal. Why did you oppose it so forcefully? Shall we resurrect it? RobinH 13:38, 10 August 2006 (UTC)


 * No, I still don't like that proposal for a number of reasons: (1) it adds an additional level of bureaucracy, (2) it requires the use of a majority vote (which I am strongly opposed to), and (3) that policy deals only with incidents of policy infringements. Disagreeing over what is the best course of action here on wikibooks hardly qualifies as a "policy infringment". I admit that i'm a little frustrated that we can't even agree on a simple voting policy, and I refuse to add any more complexity until we can sort out the basics. We should not have to form a 3-person ad hoc committee, or go calling for a steward every time there is a 10-1 vote, and we just can't sort it out our selves. On second thought, I take back my suggestion that perhaps an optional mediator or a facilitator is a good idea: I think now that it is not. --Whiteknight (talk) (projects) 17:38, 10 August 2006 (UTC)

I've got problems!
Hi, sorry for my delay in commenting on this newly modified proposal. I'm glad to see that people are trying to work out the issues involved in our decision making processes, but I don't think this proposal really fits the bill. I specifically think the "voting rules" are confused by including BOTM/COTM with things like administrator/bureaucrat/sysop rights, deletions and policy statements.

I've written User:Kellen/Decision making, which sums up how I think this policy should be stated. I think it's clearer, simpler, and easier to apply. Comments welcome here. Kellen T 03:58, 11 August 2006 (UTC)


 * I like Kelly's version much better. Specific cases of vote-counting and decision making (i.e., book of the month, RFAs, etc.) should be separate policies in any case. This discussion seems (to me) to be about top-level voting for policy decisions, such as this policy. -- SB_Johnny  | talk 12:48, 11 August 2006 (UTC)


 * I like the concept of this proposal very much as well. I agree that this policy should probably not discuss BOTM, COTM, or WJBOQ, especially because those processes already have predefined voting rules that this policy--regardless of the langauge we use--will not alter. I do like the step-by step consensus method, but I think that this policy could use a little bit more prose: more discussion on what consensus is, more discussion on the requirements of voter eligability, more discussion about how a vote is called and how it is finally decided. --Whiteknight (talk) (projects) 13:58, 11 August 2006 (UTC)

(Edit Conflict)


 * Well, the basics are on meta, though not entirely helpful in describing the mechanics of building and declaring consensus. Consensus doesn't mean unanimous approval... it just means all parties either argee or concede to the majority. For example:
 * Personally I don't like the language about voter elegibility, as I see it as contrary to the principle of consensus as a way of running wikibooks ("anyone can edit" is of course also an example of this principle). Having a minimum edit-count for BOTM, etc. doesn't bother me at all (because that's really just an expression of the community's POV), but it rubs me wrong to require this of wikibookians who are willing to spend the effort in helping to make policy.
 * However, I do understand the motivation for this requirement, and I am willing to concede the point, because I don't see me (relatively mild) discomfort about a violation of principle to hold as much weight as the level of comfort the community might gain by enacting this policy.
 * IMO, consensus is reached when the minority is willing to concede the point. Of course there might be a small minority unwilling to concede on any point, but there comes a time when the "sense of the group" can be felt, and consensus can be declared (if that language sounds Quakerly, that's probably because I'm a Quaker... we've had a few centuries of practice). -- SB_Johnny | talk 14:22, 11 August 2006 (UTC)


 * I would be happy with Kellen's version if the User responses were modified as follows:

User responses
Users who do not agree with the community decison should accept the decision and, if they feel strongly, have one of the following options open to them:
 * Propose alternative policies.
 * Propose alterations to existing policies.
 * Accept the community's decision.


 * (I do find all this talk of "community" instead of "majority" strangely reminiscent of Marxism in my youth and hence rather sinister.) RobinH 19:04, 11 August 2006 (UTC)
 * I agree that there is a certain amount of communist underpinnings to this, but I don't think it's a bad thing. I would say that all wikibookians are generally equal to one another, although I will admit that we do tend to run as a "respectocracy", where older, more respected members do seem to have more clout. But, there is nothing we are going to do to change that. --Whiteknight (talk) (projects) 20:10, 11 August 2006 (UTC)

Leaving the project is a perfectly valid response to fundamental disagreements about how the project should operate. It may seem harsh, but it's sometimes appropriate, and it's what happens anyway. Kellen T 21:54, 11 August 2006 (UTC)
 * This is true but it is a general choice for a user and not specifically recommended as a response to losing a vote. RobinH 23:14, 11 August 2006 (UTC)
 * I agree. while that is of course an option, mentioning it on the policy page seems a bit like saying "well, if you don't like it, then go play by yourself, neener-neener." Best to focus the actual rule on the positive aspects. -- SB_Johnny | talk 23:23, 11 August 2006 (UTC)

Multiple Versions
I am going to go through the history, and pick out some of the previous versions of this policy, and put them on separate subpages, so that they can be viewed side-by-side, and edited independantly. We can discuss each individual version, and reject the ones that aren't good right off the bat, and start to single in on the versions that the community likes the best. --Whiteknight (talk) (projects) 14:11, 11 August 2006 (UTC)


 * Version 1 My original version of this policy.
 * Version 2 Jguk's version
 * Version 3 RobinH's revision of Jguk's version
 * Version 4 Kellen's version
 * Version 5 My "prose rewrite" of Kellen's proposal, that makes just about all the same points, but uses fewer bullets and lists.

I think, that we should keep all discussion on this page for now, until we can start weeding out some of these versions that we don't like. --Whiteknight (talk) (projects) 14:22, 11 August 2006 (UTC)


 * I am happy with Kellen's version as of the date and time: RobinH 19:24, 11 August 2006 (UTC)

Straw poll
Seems a bit counterproductive to be working on 5 different versions at the same time... Can we maybe narrow it down a bit and then debate and modify as needed? Nonbinding, of course, just a sense of prevailing opinion at the moment. -- SB_Johnny | talk 22:39, 11 August 2006 (UTC)

Poll: which do you prefer as a basis for further work? (Please add positive endorsements only, rather than anti-endorsements):


 * Version 1
 * Version 2
 * Version 3
 * Version 4
 * Version 5
 * 1) -  SB_Johnny  | talk 22:50, 11 August 2006 (UTC). Seems a good place to start.
 * 2) Please, let's just get something official. For something as simple as voting, we really are putting together a lot of writing about it. This version seems to be a great compromise of almost all the ideas presented here. It's time to have something in place without too much voting about voting and reaching a concensus about how to reach a concensus. -Matt 23:45, 11 August 2006 (UTC)
 * 3) People like it wordy, so let's do it. Kellen T 00:42, 12 August 2006 (UTC)
 * 4) Version 5 looks fine to me and answers the misgivings that I had earlier.RobinH 10:31, 12 August 2006 (UTC)
 * 5) I agree. Version 5 is the best version so far. Now, there are certainly some finer points and additions that could be changed later, but we need to have a basic voting policy in place before we can do anything else. --Whiteknight (talk) (projects) 00:35, 14 August 2006 (UTC)
 * 1) Please, let's just get something official. For something as simple as voting, we really are putting together a lot of writing about it. This version seems to be a great compromise of almost all the ideas presented here. It's time to have something in place without too much voting about voting and reaching a concensus about how to reach a concensus. -Matt 23:45, 11 August 2006 (UTC)
 * 2) People like it wordy, so let's do it. Kellen T 00:42, 12 August 2006 (UTC)
 * 3) Version 5 looks fine to me and answers the misgivings that I had earlier.RobinH 10:31, 12 August 2006 (UTC)
 * 4) I agree. Version 5 is the best version so far. Now, there are certainly some finer points and additions that could be changed later, but we need to have a basic voting policy in place before we can do anything else. --Whiteknight (talk) (projects) 00:35, 14 August 2006 (UTC)

Minimum edit counts for voting
I was going to go ahead and be bold to change the section, but I guess I'm trying to see if I can make some reasonable wording on this first. Here is the gist of what I wanted to add to the The right to vote paragraph:


 * The minimum number of edits required to vote according to vote may rise over time due to abuse. If large numbers of individuals are flooding voting pages after meeting this minimum standard, administrators reserve the right to raise the minimum voting standard to gague the opinion of more regular Wikibooks users.  Consistant and widespread abuse of this standard may result in other policy changes and making higher minimum edit counts permanent.

Mind you, when I changed the BotM and CotM minimum voting requirements to become 20 edits, it was done in part because it was being abused before when all you needed was simply a registered user account. It has worked so far as it has forced some people to actually get involved with Wikibooks before helping to decide the current Book of the Month, which in some cases was made strictly as a form of advertisement for that book and not an expression of the overall quality of that book. It was suggested to make the minimum edit number to over 100 edits, but I argued that we should stick with 20 just to see if it might be effective.

The flip side of this is that I also want to be accomodating to brand new users, including some very effective individuals from academia who want to use Wikibooks for textbook development. This should be encouraged, and legitimate users who want to get involved should be welcomed. At the same time we need to protect this project from those who would do harm. Generally, social vandals (especially sock puppeteers) don't have the patience to create a dozen accounts and perform hundreds of edits on them of a constructive nature first. --Rob Horning 18:25, 11 August 2006 (UTC)


 * I don't agree necessarily that there should be any limit whatsoever, but I will certainly conceed the point that perhaps it is a necessity. I certainly don't think the limit should ever be higher then 20 contributions. For comparison, Meta requires that people must have at least 100 contributions to become an admin, and I don't think that the only people who should be able to vote are people who are "well known and trusted" enough to become admins (even if only on another project). I am inclined to say that we make the minimum number of votes zero, but use this clause to reserve the right to increase this limit due to abuse. --Whiteknight (talk) (projects) 19:36, 11 August 2006 (UTC)

Call for concensus
Based on the quick, informal straw poll above, I would like to officially call a vote to accept General voting rules/Proposal/Version 5 to be the current working version of this policy. If accepted, I would like to replace the current text of General voting rules with the version 5 text. We can also mark the other 4 versions, and the current text of General voting rules/Proposal with rejected.

I would also like to discuss the possibility of moving this policy to enforced, with the understanding that additional points and changes can be discussed and added under the framework of this policy. If we choose not to enforce this policy, we will still be better off because we will have a single version to work with. --Whiteknight (talk) (projects) 12:30, 14 August 2006 (UTC)


 * Sounds good, though there is still some tweaking going on regarding "Low-priority" decisions. Seems we do have consensus to work on that version rather than the others though. SB_Johnny  | talk 12:36, 14 August 2006 (UTC)


 * I'm going to move the version 5 text to the main page at General voting rules, and I'm going to move the other versions to rejected for now. We can reopen any of them up if we need to. --Whiteknight (talk) (projects) 17:02, 14 August 2006 (UTC)

Are we wieghing in somewhere now? If so, where? If not, why? -- SB_Johnny | talk 14:25, 16 August 2006 (UTC)

Attention: The text of version 5 has been moved to General voting rules, and the current discussion (including the current vote for consensus) is happening at Wikibooks talk:General voting rules. --Whiteknight (talk) (projects) 16:55, 16 August 2006 (UTC)