Wikipedia talk:Straw polls

Old talk

I disagree with this proposal completely. I don't think the introduction of formal democracy into Wikipedia will help it at all. DJ Clayworth 15:40, 27 Oct 2003 (UTC)

Yeah I could see this point. I myself am not all that in favor of the policy, but I thought I should put it forth for discussion. Dori 15:42, Oct 27, 2003 (UTC)

On the gripping hand it might be nice to have a set of rules we never invoke. A sort of doomsday option: "Oh no, he's going to invoke democracy, we had best disperse!" -- Cimon Avaro on a pogostick 16:19, Oct 27, 2003 (UTC)

The trouble with Doomsday Weapons is that once built, they tend to get used. DJ Clayworth 18:23, 27 Oct 2003 (UTC)

I'd like to see this adopted as a guideline, deleting any mention of its being binding. Voting is not the same thing as democracy, you can have either without the other. Wikipedia already has a raw but effective democracy to it, and the informal and formal votes we've had are part of it. I think this as a guideline could help a lot, so people organising new votes don't need to reinvent the wheel. And, like any other quideline we have, it should evolve and improve in Wikifashion. Andrewa 18:19, 27 Oct 2003 (UTC)

That's a very good idea. It doesn't mean every dispute has to be voted on. But when a vote is held, it should follow some rules, which tend to be missing in the informal votes we sometimes have. I would propose to add two rules:

1) To prevent foul play, voters should have to be logged-in users with a certain minimum number of edits in their history.

2) If there are more than two options to a question, it has to be decided in advance how to proceed if no option wins an absolute majority (i.e. should a relative majority be sufficient or should there be a runoff between the two most-voted options). --Wik 17:41, Oct 27, 2003 (UTC)

Considering the history of voting experiments as a replacement for consensus-building in this project, it is to be hoped that it never becomes binding. Straw polls to gauge interest are, of course, something else. (Brion VIBBER)

Well, since having this be a policy may do more harm than good, how about making it a guideline. Is anyone against that? If not, I'll reword it to match a guideline. Dori 15:22, Nov 12, 2003 (UTC)

Why not have this policy out there, and then if a member wants to run a vote according to this policy, they can just say: This vote is being run according to the policy at Wikipedia:Voting policy. -- Mattworld 18:13, 15 Nov 2003 (UTC)

Most people seem to feel that the policy would be too restrictive. The notion of a guideline is still valuable I think and what you are suggesting can still be done with a guideline. Dori 00:22, Nov 16, 2003 (UTC)

Changed to guidline content

OK, I removed most words that made it look like a policy page. If someone objects they may discuss it here and/or revert the changes. I will move the page to Wikipedia:Voting guideline if there are no objections and link the page from various other pages so it gets more attention. I think it needs more work, but I think it would be a good idea to at least standardize the voting process a bit. Dori 00:31, Nov 16, 2003 (UTC)

Moved

(from Wikipedia:Voting policy draft to Wikipedia:Voting guideline)

1. Does this mean all Votes for Deletion need to be announced on Wikipedia:Announcements and on Wikipedia:List of ongoing votes?
2. What kinds of votes do you have in mind, specifically?

Kingturtle 06:00, 17 Nov 2003 (UTC)

I just wanted a way to standardize the voting process. I think votes should follow at least some structure and publicized more. This is just my attempt. Edit at will Dori 06:05, Nov 17, 2003 (UTC)

---

Dori, I liked it better before you made the big edit to change it from 'policy' to 'guideline'. The way it's phrased right now, in the middle of a discussion, somebody can suddenly format a vote block and vote, then nobody else can even edit the option they prefer to make it sound reasonable. Also, I strongly recommend that we at the very least have a list of acceptable voting methods. The three methods that have been accepted around here in the past are average voting, approval voting and condorcet methods. DanKeshet

Moved

(from Wikipedia:Voting guideline to Wikipedia:Polling guidelines)

Removed:

1. The completed polls may be used to set policy and/or resolve disputes

This needs to be rephrased. Polling is just one way to reach consensus. --mav 00:52, 7 Jan 2004 (UTC)

See also wikipedia talk:Vote, though there's little of value there. What's the last item about - people frustrating the vote - what sort of thing are you thinking of? Martin 21:45, 12 Jan 2004 (UTC)

guideline 12

People who cast a vote in a poll must provide a motivation for their vote in the form of a comment. Simply stating for or against is insufficient. Since an unmotivated vote is useless, it does not count in the poll, and may be deleted at will by anybody. (Polls are there to create consensus, they're not straight and binding votes. This rule prevents such abuse.)

I've noticed people turning off their brains and just voting blindly during polls. Okay, so maybe they thought about it first, but we'll never know that if they don't explain their thoughts, will we? :-)

Hopefully this will fix a lot of trouble.

Let's see if people are even paying paying attention at all (I have this suspicion some people aren't). If this guideline is uncontested for 7 days, I'll assume it's acceptable, and start applying it. That's fair enough warning, I guess. Kim Bruning 19:31, 14 Jul 2004 (UTC)

gotcha ;-) how many hours was that? dunno if the if nobody notices it must be ok logic holds, although I sympathise with your goals Erich 13:22, 15 Jul 2004 (UTC)

I like that logic. I tell my kids something similar "you only get in trouble when you get caught, so don't bother me" ;) Sam [Spade] 13:30, 15 Jul 2004 (UTC)

This is absurdity, especially since it is derived from the trolling poll. There, most poll votes without comments are those voting "Yes (I agree with a very elaborate and well-spelled-out policy)". Voting yes with ~~~~ is equivalent to saying "yes, I agree with this policy. it is well formulated because of X, Y, and Z." If everyone had to compose their own little blurb to that effect just to vote we would have an extremely bloated page. Those who have taken exception to one or more points, regardless of whether their vote is yes or no, have left comments. A whole fleet of vacuous, essentially identical comments by other voters only masks these concerns unless one wishes to read every single comment for its content.

All in all, this proposed policy seems like that terrible mandate from an elementary school teacher that you answer a whole sheet of yes-or-no questions with complete sentences and no pronouns. —Ed Cormany 15:23, 15 Jul 2004 (UTC)

Well, basically the idea of polling is to discover where a consensus lies, not to determine a yes or no gut reaction kind of answer. Polls are not supposed to answer yes or no questions. If that's what you're using them for, perhaps you shouldn't be holding the poll. Regardless, if people don't start putting down why they're voting, we cant find out what they're thinking, or how the polls are working.

* If people start putting in short blurbs like you think they will, then we can move to ban polls as being Not The Wiki Way.

* If people actually start to think before voting, then polls are useful, and perhaps they can be used more.

There's only one way to find out! :-) Kim Bruning 17:06, 15 Jul 2004 (UTC)

sorry for "cutting and running" by the way Kim, it was late at night Brisbane time when I spotted your suggestion and I was a bit too tired to be very articulate. (can I also say i find 'not the wiki way' to be pretty unhelpful... if you mean 'not the way we've always done things around here' or 'not what I think the Wikipedia is a at its best' than that is what you should say... I think at least) Erich 04:17, 16 Jul 2004 (UTC)

---

Here's the proposed rule change again: 12. People who cast a vote in a poll must provide a motivation for their vote in the form of a comment. Simply stating for or against is insufficient. Since an unmotivated vote is useless, it does not count in the poll, and may be deleted at will by anybody. (Polls are there to create consensus, they're not straight and binding votes. This rule prevents such abuse.)

My logic for doing it this way is that no one ever seems to actually read the polling guidelines. I'd like to add a sanction for failure to read them. People edit rules all the time, so that's no big deal. By changing the guideline and adding a mild sanction to poke people with in exactly this manner, it might keep people from falling asleep at the wheel. ;-)

I'll just rv it back now, and see when the next person bites (if ever), I hope you don't mind! Else you just know it'll never get discussed. We have 6 days left. If by the end of that time the consensus is to drop, I'll abide by the consensus.

For obvious reasons, there shouldn't be a poll in this instance. In the mean time, is there any reason you actually happen to disagree with the rule?

Kim Bruning 13:40, 15 Jul 2004 (UTC)

ohh now I get it... revert wars... now that is the wiki way!! (just joking... kinda) ;-) Erich 04:17, 16 Jul 2004 (UTC)

I must strongly disagree with this proposed change in policy. People should be able to simply agree or disagree with what is being put forth by a poll. I don't know of any formal legislative body that requires its members to provide a justification every time they vote. This policy would only serve to cripple the voting process. It would take longer for votes to accumulate, it would make people less inclined to vote, and it would encourage mob rule, as people would be able to attack minority votes and claim their justifications were insufficient.

I also oppose the manner in which this policy change is being presented. According to the Announcements page, if unopposed (who defines opposition? what constitutes opposition?) the policy will simply apply. Where's the discussion? Where's the consensus? New policies should require community approval to go into effect, they shouldn't simply need to lack opposition. The need for approval encourages dialogue about the policy and can lead to refinements and understanding. The reasons for proposing this policy are equally specious. The policy is intended to be a sanction against not reading polling guidelines? Couldn't we find better ways to encourage people to read them (that don't involve discouraging them from voting)? I feel strongly that this proposal must be brought to a vote (the old-fashioned way) instead of just falling into effect.

Acegikmo1 17:15, 15 Jul 2004 (UTC), hoping to not sound too harsh

yeah what Acegikom1 said! including the bit about being harsh. Erich 04:17, 16 Jul 2004 (UTC)

The justification for the proposed rule is that people are "turning off their brains and just voting blindly". It's a shame if that is happening, but requiring people to write out their motivation won't solve the problem. A brief justification of a vote is wholly insufficient to weed out insincerity and knee-jerk reactions from thoughtful participation. The simplest example is complete consensus. Suppose there is a poll on the topic Vandalised pages should be reverted to a previous version, and it happens that everyone completely agrees. Then voting under the new guideline will look something like this:

• • User 1: Yes, vandalism makes the content of an article worse, and reverting vandalism will make it better again.
  • User 2: Yes, reverting vandalism protects the content of an article from being made worse by vandals.
  • User 3: Yes, article content will be made worse by vandals unless we revert it to a previous good version.
  • User 4: Yes, but I like the way User 1 justified it better than the way User 3 justified it.
  • User 5: Yes, and I think all the justifications given above are equally good.
  • User 6: Yes, vandalism makes the content of an article worse, and reverting vandalism will make it better again.

Under the guidelines without the new number 12, we would have had six yes votes, and we wouldn't have known whether they were thoughtful and sincere. Under the quidelines with the new number 12, we have six yes votes, and we still don't know whether they are thoughtful and sincere. Maybe User 5 genuinely embraced what everyone else said. Maybe User 6 happened to think about it on his own in exactly the same terms User 1 did, and couldn't come up with a more articulate way to say it. Maybe User 4 hasn't reflected on the issues at all, and is drawing a distinction just to appear to meet guideline number 12. There is no way to know.

In short I strongly oppose the new guideline. Since it won't solve anything, implementing it would be akin to disrupting Wikipedia to make a point. By all means, let us promote the development of consensus by participating in thoughtful debate, but let us not silence the voices of people who we suspect are not sufficiently thoughtful.

Peace, --Fritzlein 18:00, 15 Jul 2004 (UTC)

If those situations do occur a lot as you predict, then we can remove polls from the dispute resolution process entirely, since we'd have shown conclusively that polls Do Not Work

On the other hand, perhaps polls will start a lot of interesting discussion this way, that's what I'm hoping for in any case.

(see my comment above for details) Kim Bruning 18:15, 15 Jul 2004 (UTC)

Your arguments seem based on the premise that the only valid function of a poll is to help create consensus. You consider the unanimous poll I gave above as an example of how polls Do Not Work. I respectfully disagree. Polls have another very useful function, namely establishing whether or not consensus already exists. If a vote goes overwhelmingly one way, we don't need a poll (or any other means) to establish consensus, because it already exists. If a vote is evenly divided, the poll makes it obvious that there isn't an existing consensus, and further discussion is necessary if consensus is to be achieved. Either way, the poll has served a purpose, and my example is a way in which polls Work Perfectly Well.

We should by no means forget the goal of producing consensus, as opposed to merely testing whether it exists. To the extent that polls are an element in producing consensus, they are doing double duty. But even if polls do nothing but measure consensus without advancing it, they serve a useful function.

Peace, --Fritzlein 18:48, 15 Jul 2004 (UTC)

You're absolutely right! The problem is that sometimes polls are misused. I've seen people try to use polls in the recent past to force POV of a vocal minority group into articles. In the end things worked out ok I think, mainly because people were paying attention.

To make things easier, we need some way to check whether a poll has been conducted correctly. Even if it's a bit more work to read through, fine, but at least in the new worst-case we'd have *some* meta-information, no matter how vague, as opposed to none at all.

Kim Bruning 19:09, 15 Jul 2004 (UTC)

I don't think the proposed guideline is helpful, and I have removed it. In many polls or surveys on Wikipedia, the reasons for someone's opinion/vote have already been stated by others, or are so basic that people think it odd to express them. Requiring a comment is not useful when all someone feels like adding is "I agree with X", or "me too". It's so much easier for them to do with just ~~~~. Part of the point is to find out the opinions of people who might not join the discussion because they don't have anything original to say.

To deal with the problems of polls, I have instead renamed this page Survey guidelines and changed the text accordingly. I proposed this idea at Wikipedia talk:Current surveys and nobody objected there, so I've gone ahead with it. I think this will help address the problem of blind voting and the misperception some people have that polls have a binding effect. --Michael Snow 23:18, 15 Jul 2004 (UTC)

fair enough Michael, It raises the point of how policy does get agreed though. Does it come down to consensus on the policy pages in the light of surveys? Erich 04:17, 16 Jul 2004 (UTC)

See Wikipedia:Policies and guidelines#How are policies decided?. That may not answer your question fully, of course. I would say it does come down to consensus, either via discussion or based on existing practices that nobody objects to. Surveys can be a useful tool to figure out consensus, certainly. More substantive discussion also helps, and is more useful than surveys for thoroughly airing the issues.

This is a broader subject than just the topic of surveys, but you're getting at the issue of whether our policy-formation system itself is adequate. You may be interested in reading an extended discussion I had with UninvitedCompany on my talk page about this. --Michael Snow 05:33, 16 Jul 2004 (UTC)

This sounds like it might have a chance of working. Thanks Michael. :) Let's stick with it for a bit and see what happens. Kim Bruning 07:50, 16 Jul 2004 (UTC)

Voting via templates

(Portions of this conversation stamped before 14:40, 2005 Apr 14 (UTC) were moved here from Wikipedia:Village pump (policy))

Recently, a couple pages (Template:Nn and User:Dr Zen/keepschools) have come up for deletion which were being used as VfD votes by some users. The voters were adding these as votes using double-braces, then adding their signature.

I am unaware that there is any current policy preventing this practice, but I think we can almost all see how disruptive and inappropriate it is. In effect, a voter is voicing a comment such that anyone can make a single edit and change all the related vote pages. I'd like to propose this following:

"All votes and any associated comments must be presented in simple text only. Active use of a template or any other form of transclusion is disallowed wherever votes are being recorded. Such votes will not be considered when evaluating the results."

Where is the best place to add this guideline? What sort of objections to it are there? -- Netoholic @ 15:32, 2005 Apr 13 (UTC)

I have no objections to it in the slightest.

As to the best place, perhaps the Deletion policy page would be the best place for it? If it does become policy a mention at Wikipedia:Guide to Votes for deletion would probably be apropriate and worthwile. Thryduulf 16:29, 13 Apr 2005 (UTC)

• I wouldn't mind such a policy, but be sure to determine if there's a concensus first. Every article is a case on it's own and should be assessed as such. Mgm|^(talk) 18:02, Apr 13, 2005 (UTC)

Restate, please, to avoid any confusion about whether the policy attempts to ban using a subst call to a template, which could never be proven and which is not subject to the could-be-silently-changed reasoning. --Jerzy (t) 19:00, 2005 Apr 13 (UTC)

• I doubt anyone cares or can complain about using subst: - the point is that the text of a vote should not be modified except by editing the vote page. I added the word "Active" above. -- Netoholic @ 19:14, 2005 Apr 13 (UTC)

I'm not sure that additional instruction creep is needed. See also Wikipedia:Deletion policy/Personal attacks, which dealt with the same sort of issue. This is probably less likely to be used against users shouting at each other in a school vfd, where I'm not convinced that anyone is reading what other people are saying anymore, than it is against users who transclude their signatures. —Korath (Talk) 19:50, Apr 13, 2005 (UTC)

I think the personal attacks thing is very different. The issue of "vote templates" creeps in occasionally, and should be documented as a bad practice, with a realistic idea of what to expect if you try it (i.e. your vote doesn't count).

As far as signature transclusion, that is already a poor practice on its own, but outside the intended scope of my suggestion. -- Netoholic @ 20:08, 2005 Apr 13 (UTC)

• I concur with this idea. Good catch! — Stevie is the man! ^{Talk | Work} 21:05, 13 Apr 2005 (UTC)

• It's a sound idea. VfD is large, slow, and busy enough as it is—it doesn't need to deal with transcluded comments and sigs as well. --TenOfAllTrades | Talk 00:20, 14 Apr 2005 (UTC)

• All that templates do is automate the existing ideological voting on VfD. I think that is a much bigger issue than the use of templates. The reason ideological voting occurs is that deletionists and others (formerly GRider), regular bring up for votes articles about whose inclusion/deletion is an ideological issue. If people would stop engaging in socratic VfD nominations then VfD would be a much nicer place.
  As an alternate solution. All templates used in a VfD vote should be substituted in silently. This would solve the issue of the modification of a template being used to vandalize votes. Otherwise I that template insertions should be ignored and not processed.
  That said, I dont think template voting would be anywhere near as controversial but that it has been used by some inclusionist voters, to subtly parody mindless deletionism. Think Dada or Pop art. Klonimus 02:14, 14 Apr 2005 (UTC)
  • Speaking of factionalism (see below), we'd all be better off if you wouldn't cast this as a partisan debate. The problem is not that "deletionists nominate articles whose existence is an ideological issue"--that essentially absolves all keep voters of any responsibility. The problem is that we have no consensus on school articles, and we need one. In addition, users attempting to subtly parody delete voters ala dadaism veers dangerously close to WP:POINT. Please--this is a procedural debate, not an ideological one. I think the only real solution is to come to a workable compromise regarding schools. Meelar (talk) 20:50, Apr 14, 2005 (UTC)
• "all templates used in VfD votes should be substituted in silently" while not a bad idea, requires a software change and the implementors have more important things on their hands. The suggestion that this is controversial because inclusionists use it, however, is an example of mindless factionalism, not to mention paranoia.
  Anyway. I agree with the proposal, but we should also do some consensual poll to make it official. I also oppose subst-ing the template, or copy/paste voting in general, but I can't think of a feasible way to prevent people from doing that. Radiant_* 11:54, Apr 14, 2005 (UTC)
  • We only need polls when consensus is unclear. Noone has yet come through and defended this practice or to state that it is even acceptable. Korath mentioned a desire to avoid the instruction creep, but I'm unclear whether he objected to the practice or just thinks this is obvious and needn't be documented. -- Netoholic @ 14:40, 2005 Apr 14 (UTC)
    • The latter - it's obvious and doesn't need new rules. If nothing else, it's clearly a breach of civility. —Korath (Talk) 15:32, Apr 14, 2005 (UTC)

• The issue of changing a large number of votes is valid, while no one's demostrated an advantage to this. Boilerplate text can be included with {{subst: if desired. Otherwise, template votes shouldn't count. Demi ^T/_C 17:36, 2005 Apr 14 (UTC)

• I don't believe this is worth its contribution to instruction creep. I believe this is a temporary problem brought about by the current disagreement on inclusion guidelines. While it is probably unrealistic to expect complete consensus on inclusion guidelines, I would hope that the present matter of schools will be settled. The Uninvited Co., Inc. 22:28, 14 Apr 2005 (UTC)

Make it semi-policy

• I don't think this is instruction creep, it's a very clear "Don't do this" case similar to the Tally Box Pox of last month or thereabouts. Since this primarily affects VfD, the easiest way to do this would be adding a Policy Consensus topic stating the proposal that "votes containing non-substed templates are invalid" or something similar. Then wait a week, and if it gets a lot of 'agree' votes and few to none 'oppose' votes, it's consensual enough for a semi-policy. Which would mean that if/when anyone sees a templated vote, they should subst: it and inform the original voter to not do that. Radiant_* 15:24, Apr 20, 2005 (UTC)

Companion page

I'm compiling suggestions on how to choose a voting method at Wikipedia:How to hold a consensus vote. (The name can change to match the terminology of other Wikipedia discussions.)

The page starts with the strong suggestion that you don't vote at all and have an informal discussion instead, and then suggests some voting methods, and how to present them to minimize confusion. The emphasis is to keep the voting as simple as you can while reaching a consensus.

I made this page because I've seen several surveys in a row where the procedure became more contentious than the actual issue, and so consensus became impossible when it might have been possible before. I'd like to prevent that by making some suggestions on what works and what doesn't.

RSpeer 01:00, May 9, 2005 (UTC)

Fix the loopholes

As the ever-persisent disputed over Gdansk/Vote prove, our current policy is in need of a serious redesigning to fox the loopwholes. Please comment on Template_talk:Gdansk-Vote-Notice#Constructive_proposal and help us draft a new policy proposal. --Piotr Konieczny aka Prokonsul Piotrus ^Talk 11:09, 12 Jun 2005 (UTC)

Proposal now official, now at: Wikipedia:Village_pump_(policy)#Fixing_giant_loopholes_in_Wikipedia:Survey_guidelines. --Piotr Konieczny aka Prokonsul Piotrus ^Talk 19:36, 16 Jun 2005 (UTC)

Fixing giant loopholes

Note that the discussion that lead to this propsal is User:Piotrus/Sandbox#Fixing_giant_loopholes_in_Wikipedia:Survey_guidelines.

Wikipedia:Survey guidelines, a guideline for all manners of surveys and votes on Wikipedia, is deeply flawed in its existing rules and fails to address several important issues. A prime example is that in Gdansk/Vote it allows both sides of the dispute to claim they are immunie from 3RR rule, as well as disputing the very vote results. As there are proposals for new votes similar to the Gdansk/Vote, I feel we must fix the policy ASAP - otherwise, those votes will be nothing but a giant time loss for everybody involved, including poor participants of RfC, RfA and admins enforcing 3RR rule, who - judging on Gdansk/Vote results - will soon be asked to chose sides in various interpretations of the vote. I hate instructions creep, but I am afraid this is necessary in this case. I think the following changes have to be implemented to survey 'how to' creation and enforcement:

1. This is supposedly an official guideline, but it states first: These guidelines provide a framework that may be followed when creating a new survey. These are not binding in any way. It would be funny if it wasn't sad. What's the point of official unbidning guidlines, especially when they are used to change/counteract official policies (like 3RR)? We need to make some of them obligatory for carrying all votes on Wikipedia, or at least those that affect official policies. I definetly think that each survey must follow points 2, 4, 5, 6, 7 and 8.
2. Points 1 and 3 are 'good wishes', and should be moved to the introduction section of the guideline
3. Point 9 is too general: Where there is a sign of activities intended to frustrate the intent of the survey, those who can opine may be restricted. A lack of restrictions is usually best, so this may be invoked after the polling has started. We need to specify what are the 'acitivies intended to frustrate the intent of the survey' and how 'those who can opine may be restricted'. In current form those rules allow for, for example, for a restriction of vote to 'Yes' only.
4. Point 10 is even worse: If the majority of opinion is in one direction, but a significant minority of people oppose it, work to find a solution that can be accepted by as many people as possible. It can easly be used to dispute almost any vote after the voting has ended. It should either be deleted, or we need to define significant majority *gasp* with clear qunatiative percents - and make it clear to the voters (by adding the relevant requirement to step 2 of 'creating a new survey section') that they know that the vote needs more then normal majority (50%+1) to pass. Otherwise any user can claim that anything is a 'significant majority' and the vote is unbinding. I'd suggest that a Wikipedia vote must require at least a 75% majority to be binding, otherwise it can be only a guideline. This should be cleary stated at the beginning of the survey, so that everybody participating would know how much votes are needed for the motion to pass (or not).
5. Further, we need to add to that section an obligatory information on who can vote, preferably make this a rule in survey creation instead of leaving it to arbitrary decision of not always impartial vote creators. I.e. make a rule that only users who have at least x major edits and registered y time periods before voting can vote. I'd suggest 100 edits and a week. Note that such a rule applies to current meta:Election_candidates_2005.
6. If a vote would influence and official policy (like the 3RR) then information about the vote should be added to the relevant policy discussion/talk page. Similarly, if a vote would influence mainspace articles, note on the vote should be added to them as well (as was done in Gdansk/Vote example).
7. We need to add an information what is the minimal numbers of voters that makes the vote valid (for example, a vote with 1-2 users is not very useful). While for the obscure topics that generate little interest even 1-2 votes may be sufficient, I suggest at least 30 voters for a binding vote in cases the vote would go against an official policy (like the suspention of 3RR rules against losers of the vote reverts, as is the case with Gdansk/Vote).
8. We need to decide whether and if so, under what rules can a vote/survey be repeated. I think a vote can be repeated until a binding desision is reached. If a vote has passed a decision, a new vote should not bring the issue again until at least a year have passed since the last one or there is an Arb Com agreement to start a new vote. And if a vote is repeated, all past votes on the issue should be copied and included in a new vote unless the vote owner changes decision. An exception can be made for instances where it is quickly discovered that the wording has been too vague and thus the primary vote failed to solve the issue and that some sort of repeated vote is necessary to make it clearer what is intended. I'd suggest that any user can request a revote on such grounds (clarification needed), informing all past voters and asking them for agreement, and if 75% agrees that a revote on those grounds is needed, it can be carried out. However, if they don't, then a revote on those grounds cannot be repeated for a year.
9. We need to make clear if the current changes apply (or can apply) to old votes.
10. We need to make a rule about vote enforcement. What's the use of a vote if those who don't like the result can ignore it? If a vote is binding, that all edits contraty to the vote decision should be reverted and not count toward 3RR rule.

Well, that's all I can think of now. A final note: I don't want to formalize all votes and surveys, only those that like Gdansk/Vote infringe/break other official policies like the 3RR in case of Gdansk/Vote. Fuzzy and inbinding votes are good - but not in important matters. So, what do you think? --Piotr Konieczny aka Prokonsul Piotrus ^Talk 19:31, 16 Jun 2005 (UTC)

You misunderstand the point of surveys. A survey or poll is not a binding vote. Kim Bruning 19:53, 16 Jun 2005 (UTC)

Tell that to the people engaged in Gdansk/Vote and claiming it gives them immunity to 3RR. Apparently quite a lot of people (enough to force desisions on 3RR blocks or lack of them) considers this survey binding enough to be above the official policy of 3RR. --Piotr Konieczny aka Prokonsul Piotrus ^Talk 20:06, 16 Jun 2005 (UTC)

That's *ahehehehe* wrong. Which people were it? *innocent look* *rubbing of hands behind back* Kim Bruning 11:13, 17 Jun 2005 (UTC)

Some of the above Moved from Wikipedia:Village pump (policy)

Since there have been no meaningful (content) objections so far, I will be changing the guideline in a few days. I will post a notice on VP(p) again, and RfC and AN this time to attract more attention. --Piotr Konieczny aka Prokonsul Piotrus ^Talk 13:39, 19 October 2005 (UTC)

You've had one objection from Kim Bruning, and here is another from me. Wikipedia is not a democracy. A survey cannot overturn policy. Secretlondon 14:14, 19 October 2005 (UTC)

Who create policies? We do. Surveys are just a formalized voting, and Wikipedia is build around consensus. If majority of users want something, it has to be changed - Wiki has always worked that way. I don't see a difference whether the majority choses to speak on a talk page of through a more formal survey vote. A note about a survey will be added to the policy page, so all interested parties will be informed. Perhaps, if you think it may be abused, we can add a requirement that such votes need to be notified on AN, RfC, VP(p) and should require a given number of participnats/majority support? --Piotr Konieczny aka Prokonsul Piotrus ^Talk 14:38, 19 October 2005 (UTC)

Anyone who knowingly violates the 3RR can be banned, without exceptions. Surveys do not provide an exemption for that rule. That is the only thing that must be clarified on the page. Superm401 | Talk 14:24, 19 October 2005 (UTC)

Replied above. I doubt 3RR would ever be changed, it is too useful. But nothing should be above the will of the majority - if for example at some distant future, 90% of Wiki users want to change 3RR to 4RR, shouldn't they be allowed to do so? --Piotr Konieczny aka Prokonsul Piotrus ^Talk 14:38, 19 October 2005 (UTC)

I am not sure of my views on this -- I would like to see a specific draft of alternate text as part of such a proposal. But in any cas one cannot IMO simply put a proposal out and if it received no positive comments at all, implement it on the ground that there have been "no objections". I object to implementing this unless a significant number of editors specifically indicate support, wheter in a poll or in discussion. DES ^(talk) 15:59, 19 October 2005 (UTC)

The above proposal is a result of agreement and collaborative work by several users. We would like to hear more opinions on this now. --Piotr Konieczny aka Prokonsul Piotrus ^Talk 16:48, 19 October 2005 (UTC)

Fine. My opnion is that the above text is not a proposal, but a skeleton from which an actual proposal might be built, and that it is not anywhere newar a sufficnent basis on which to be "changing the guideline". I want to see a detailed text of what the revised guideline would look like, and to see multiple people specifically supporting that text. I don't object to developign such a text, adn to then seeking support for it. But commeths above suggested that you thought you were ready to change the workign guideline. i think you haven't come near that stage yet. DES ^(talk) 18:13, 19 October 2005 (UTC)

I think it is a good start. Policy making is very tough business, and I am grateful for Piotrus for getting things going. The comments above have my support (since I was also involved in the first disussion round). I am not sure if they can be implemented all at once, but a one-by one approach should be possible. I think especially the points "what is a majority", "who can vote (edit count)" and "how many votes needed for a valid vote". And, most of all, how to enforce the vote outcome if users refuse to follow the majority vote. The only way currently possible in the Wikipedia:Blocking policy would be for disrupting wikipedia. A 3RR Excempt was also voted on in the Gdansk vote, but continues to cause problems. Yet a vote enforcement is needed, otherewise we'll open the doors to anarchy. -- Chris 73 Talk 09:41, 20 October 2005 (UTC)

Time needed to reach consensus

Consensus must be reached about the nature of the survey before it starts. Allow about a week for this process

Is it good to specify a timeline in this way? I would rather suggest something like "Consensus must be reached about the nature of the survey before it starts. Let this process last until it is finished, which can often can be expected to last at least a week." / Habj 11:04, 27 July 2005 (UTC)

Polling/survey templates

I've created AfD-style poll closure templates at polltop and pollbottom. Like the AfD-style closure templates, these are meant to be used with {{subst:}}. There's a demonstration of what these templates do on each talk page (the demos are identical, as both templates are necessary to get the effect). Hopefully I haven't recreated work already done. :P If someone wants to, it might be worth adding a link to these to the survey guidelines... but only after suitable discussion, and of course, a survey. :P Comments welcome and appreciated. --Locke Cole ^{(talk) (e-mail)} 09:16, 21 November 2005 (UTC)

What on earth?

Have a look through the revision history of this one. This is textbook m:instruction creep and I've edited it accordingly. Wikipedia is not a game of Nomic. Or shouldn't be - David Gerard 20:08, 24 November 2005 (UTC)

I've merged it to Wikipedia:Straw polls as was suggested in the header - David Gerard 20:32, 24 November 2005 (UTC)

Move to delete 99% of all Lists and Categories of Jews

Please read the sixteen point introduction at Wikipedia:Centralized discussion/Lists by religion-ethnicity and profession#Move to delete 99% of all Lists and Categories of Jews: Sixteen reasons why this should become a fixed Wikipedia policy and related discussions at Wikipedia talk:Centralized discussion/Lists by religion-ethnicity and profession#Proposed amendment: remove all Jewish-related lists. Thank you. IZAK 11:36, 4 January 2006 (UTC)

What happens to old polls?

There needs to be a guideline as to what to do with old polls. When can/should they be archived? Should a summary of the poll topic and its results be given a permanent place at the top of the article's talk page? Any thoughts? -- noosphere 00:30, 2 April 2006 (UTC)

It's just an issue of appearence..whats the need of another rule?--Urthogie 11:49, 2 April 2006 (UTC)

The need is for clarity so as to minimize pointless arguments. These issues should be hashed out once and a guideline created, so stupid arguments and edit wars don't start up over these issues in the future. -- noosphere 01:37, 1 May 2006 (UTC)

Contradiction

This guideline has a contradiction: in the lead, it describes straw poll as not having "opening and closing times as votes do", while in the "Creating a survey section" step 4 is "A deadline for the survey should be considered so as to resolve the issue in a timely manner". Explicitly stating some sort of loose timeframe seems like good advice to me; what do others think? --jwanders^Talk 08:49, 26 July 2006 (UTC)

Guideline status discussion

DavidHOzAu has removed the following text recently added by "Radiant!"

This page explains how to make a good straw poll. See Wikipedia:Voting is evil for an explanation of why we usually don't use straw polls.''

The page certainly does give guidance on making straw polls, and it's a truism that we make little use of straw polls as a decision-making mechanism. I'm not really sure what DavidHOzAu's objection can be, so I've taken the step of restoring the text. --Tony Sidaway 10:57, 22 September 2006 (UTC)

Stating that a guideline (of any sort) is usually not used is like saying that this is not a guideline. Furthermore:

“ This page is considered a guideline on Wikipedia. It is generally accepted among editors and is considered a standard that all users should follow. However, it is not set in stone and should be treated with common sense and the occasional exception. When editing this page, please ensure that your revision reflects consensus. When in doubt, discuss first on the talk page. ”

There is only an occasional exception, and guidelines are a standard to follow: Stating that we usually don't use straw polls is a load of you-know-what.

If Radiant!, who added the text, wanted to dispute this guideline there is a reccommended way to do that. His edit was totally non-standard. I'm reverting his edit until he discusses this intelligently on this talk page and convinces me that straw polls are the exception rather than the rule in Wikipedia's decision processes.

Adminship is not an excuse to ignore consensus. --DavidHOzAu 23:48, 24 September 2006 (UTC)

I don't understand what your grievance is here. Could you try to restate it and maybe I'll follow. Is the crux of the matter that you think we actually do make decisions by straw poll? --Tony Sidaway 00:06, 25 September 2006 (UTC)

• I'm sorry, perhaps I didn't make myself clear, despite all my highlighting. Let's look at my earlier post again:

• Stating that a guideline is usually not used is like saying that this is not a guideline.
• There is only an occasional exception
• guidelines are a standard
• Stating that we usually don't use straw polls is a load of you-know-what.

'nuff said.

• In regards to your question: It doesn't matter what I think: the presence of {{guideline}} should indicate to you that the the Wikipedia community thinks that straw polls have an important place on Wikipedia. If anyone wants to dispute this guideline, please do so the right way. Thank you. --DavidHOzAu 00:23, 25 September 2006 (UTC)

DavidHOzAu, Please note that we don't actually make decisions by straw poll, and we haven't done for years, because Voting is evil. [ælfəks] 00:25, 25 September 2006 (UTC)

Funny, I could've sworn we were using approval voting at Wikipedia:Village pump (proposals)/Sidebar redesign/Final draft vote.

Anyway, what should've been added is something to make it clear that discussion interprets surveys, instead of sounding like surveys are discouraged period. Feel free to add it back in its unmodified form, but IMHO I'd edit it a bit first. --DavidHOzAu 01:02, 25 September 2006 (UTC)

The important difference is between "approval voting" to pick the most popular design for a standard, and "majority voting" on any random motion. The former works, the latter does not. >Radiant< 16:01, 25 September 2006 (UTC)

We use straw polls in many places on wikipedia. They help gauge consensus. This is indisputable. In no way does this mean that these straw polls are binding - they're not binding. We *do* in fact make decisions based on polls, but hopfully not on polls alone. I agree with DavidHOzAu's edit to remove the text - because it is in contradiction to the nature of this guideline. Fresheneesz 07:42, 26 September 2006 (UTC)

To repeat, the important difference is between "approval voting" to pick the most popular design for a standard, and "majority voting" on any random motion. The former works, the latter does not. >Radiant< 08:35, 26 September 2006 (UTC)

Whats your point? Fresheneesz 18:50, 26 September 2006 (UTC)

Removed tag

Since there is clearly some contention as to the status of this article, I've removed the "guideline" tag. Not a guideline if there isn't consensus. - brenneman ^{L} 11:51, 29 September 2006 (UTC)

Where? Where are these people disputing this page? Radiant is one, are you another? This has been a guideline for a year, more discussion is needed before its demoted. Fresheneesz 00:55, 30 September 2006 (UTC)

• Once more, whenever you're talking about "promoting" and "demoting" pages you're clearly misunderstanding Wikipedia. >Radiant< 11:08, 30 September 2006 (UTC)

Mandating reasons attached to a vote

I think that this page should encourage users to write the reason for their vote right next to it like this:

Delete - because the page fails verifiability, none of the sources provide information that verifies the article. User:Your name here a:dd, the datehere year (UTC)

That way people can gauge how and why the poll is turning out like it is. Fresheneesz 19:10, 26 September 2006 (UTC)

• So what you're basically saying is that people should discuss rather than voting. >Radiant< 21:22, 26 September 2006 (UTC)
  • And also that if a reason comes up which is better than all the others, it overrides all the other opinions... sounds a lot like you're saying that straw polls are non-binding and that they're an indicator only! Then why do you insist that this is a guideline? Daily usage shows that this has been rejected time and time again. [ælfəks] 09:42, 29 September 2006 (UTC)

.... uh.. straw polls... *aren't* binding. Doesn't it say that in the article? I don't understand why you're confused. Straw polls aren't binding, and are used as an indicator. Whats your point? My point is that this page attempts to explain how to make a good straw poll. Are you trying to tell me that a straw poll would be better *without* peoples reasons for their opinions? Fresheneesz 01:49, 30 September 2006 (UTC)

Oh, btw, I insist on this as a guideline because it has been one for a long time. That, and I don't see significant consensus to demote it. Do you? Please just point me to something that would convince me. Fresheneesz 01:51, 30 September 2006 (UTC)

• A how-to is not a guideline. And if you're asking people for their reasons anyway, you're actually discussing rather than voting. >Radiant< 11:05, 30 September 2006 (UTC)

Reasons would make it BOTH a poll and a discussion. Fresheneesz 01:19, 1 October 2006 (UTC)

Page seems old, broken , and badly maintained

It seems to advocate things going in different (random) directions. We should consider reverting.

Somehow guidelines do not appear to coincide very well with best practices, we've seen it before. We may need to consider a different procedure for maintaining them.

Kim Bruning 14:54, 1 October 2006 (UTC)

• Told you so. At any rate, reverting where exactly? >Radiant< 20:36, 1 October 2006 (UTC)

Radiant wants to make sure you want to revert to a time when this wasn't guideline - a vast minority of the history of this page. But I agree, guidelines, practice, and intention have all gone in three different directions - wikipedia needs a better framework of organization. Fresheneesz 23:01, 1 October 2006 (UTC)

Returning this page back to guideline status

This page appears to have more merit than it's previous {{essay}} tag would lead one to believe with 1, 2, 3, 4, 5 other language Wikis that appear to have equivalent pages. This page merits continued development and being turned back into a guideline. (→Netscott) 05:05, 20 November 2006 (UTC)

Even WP:NOT references this page (and has done so for a very long time). (→Netscott) 05:34, 20 November 2006 (UTC)

• However, analysis shows that this page not at all matches actual practice in the English Wikipedia. We are not responsible for other-language wikis, most of which are several orders of magnitude smaller and many of which have different "rules" than we do. Even though the header has a few lines that state otherwise, the gist of this page reads as an endorsement of polling in most situations; such an endorsement is simply not a good idea given the problems that can be caused by improper or premature polls. (Radiant) 15:09, 20 November 2006 (UTC)
• If there are some adjustments we could make to get back to guideline status, that would obviously be ideal. I will see if I can find the time. TheronJ 22:24, 20 November 2006 (UTC)

Wording change

I would like to add the following to the page, under "What a Straw Poll is good for":

"Straw polls can be useful in cases where there are many editors involved in a debate, or in cases where some editors in a discussion are posting multiple times per day, but others can only participate once a day (or less). A poll structure can help ensure that equal weight is given to all editors' opinions."

Does this look acceptable? --Elonka 06:52, 2 December 2006 (UTC)