Wiktionary:Votes/pl-2009-03/Removing vote requirements for policy changes

Removing vote requirements for policy changes

• Voting on:

That the wording of Template:policy be changed from "It should not be modified without a VOTE" to "It should not be modified without discussion and consensus. Any substantial or contested changes require a VOTE."

Let's face it folks, our policy documents are generally out of date. I can't count the number of times that I've had to tell a new contributor, "Well, even if it's not written down anywhere, it's still policy." This is both confusing and frustrating, especially for new editors. This vote is an attempt to remove one of the greatest hurdles to keeping our policy documents up to date with current practice. In short, it means that people can make edits to policy documents like WT:ELE and WT:CFI without a vote. However, if there is disagreement (probably often taking the form of a rollback of the edits), then the issue is taken to the Beer parlour. If a strong consensus is achieved, then the changes become no longer "contested" and can be remade. This does not entirely do away with votes, as they will still be useful in contentious situations, and will be especially important in major switches of policy. However, it does allow us to cut through a lot of red tape to align official policy with de facto policy. Props to Visviva for making the initial BP proposal of this. -Atelaes λάλει ἐμοί 06:15, 31 March 2009 (UTC)

• Vote ends: 23:59 29 April 2009 (UTC)
• Vote started: 6:15 31 March 2009 (UTC)

• Vote created: -Atelaes λάλει ἐμοί 06:15, 31 March 2009 (UTC)
• Discussion:

  WT:BP#Revising_Template:policy

Support

1. Support -Atelaes λάλει ἐμοί 06:15, 31 March 2009 (UTC) (as nominator)
2. Support Visviva 06:27, 31 March 2009 (UTC)
3. Support Neskaya kanetsv 06:35, 31 March 2009 (UTC) Much needed. --Neskaya kanetsv 06:35, 31 March 2009 (UTC)
4. Support Bequw → ¢ • τ 06:50, 31 March 2009 (UTC)
5. Support —Michael Z. 2009-03-31 08:03 z
6. Support Eivind (t) 08:31, 31 March 2009 (UTC)
7. Support † ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 13:20, 31 March 2009 (UTC)
8. Support Conrad.Irwin 16:44, 31 March 2009 (UTC)
9. Strong support —Ruakh_TALK 23:03, 31 March 2009 (UTC)
10. Support —Rod (A. Smith) 04:18, 1 April 2009 (UTC)
11. Strong support DAVilla 19:03, 1 April 2009 (UTC)
12. Support EncycloPetey 04:19, 9 April 2009 (UTC) I have some reservation about doing this, but currently think the overall benefits outweigh the possible negative outcomes. --EncycloPetey 04:19, 9 April 2009 (UTC)
13. Support Dominic·t 05:54, 9 April 2009 (UTC)
14. Support — [ ric ] opiaterein — 17:33, 12 April 2009 (UTC)
15. Support; it is better to require visible opposition to a change to prompt a vote, rather than forcing each change to be held for a month even if it is uncontroversial. However, many changes are tricky, and it shouldn't be considered that this will magically make coming up with and clarifying these changes easy. JesseW 05:23, 16 April 2009 (UTC)
16. Support Jackofclubs 08:01, 29 April 2009 (UTC).

Oppose

1. Oppose Robert Ullmann 11:24, 5 April 2009 (UTC) very strongly oppose and I would ask everyone above to reconsider this very seriously

   The voting process is not onerous; it takes only a few minutes to set, and a proper amount of time to specify the change. If you are not willing to take this degree of time and care, you should not be editing policy documents.

   The voting process exists to document the consensus resulting from proper proposal and discussion. If that has in fact occurred (the premise of the proposal) then a vote is simple. If it has not, then a vote becomes problematic. If you are having trouble bringing something to a vote, it means that you haven't gotten the needed consensus. (Or that you don't really understand what you are doing.) The problem is not the voting process, and the premise above that there has been sufficient debate has not been met. Making an end run around the voting process to make what is then almost certainly an ill-advised change to ELE or another policy document is not a solution.

   There have been several frustrated attempts recently to propose changes. In each case the problem was not the requirement for a vote, but the requirement that the change(s) be properly discussed (before any vote), be coherent and understood. And: not be very bad changes. The voting process is not the problem.

   If passed, this will result in a devolved state of affairs in which every change to ELE or another policy document is immediately met with a revert and a demand for a vote. Why? Because in any case in which consensus has been achieved (the premise above), a vote is simple and easy. The procedure described in this proposal is only useful when someone wants to pretend that consensus has been achieved when it has not and try to force a change. Very simply, if the premise of the proposal above has been met, voting is not a problem; the only possible purpose and use of the proposal is to evade a vote that would and should fail. Robert Ullmann 11:24, 5 April 2009 (UTC)

   Two more points:

   • All policy does not have to be crammed into (e.g.) WT:ELE. It is too big already, and it is is simple to create other pages that describe something in detail, without having to have a vote or anything (until eventually making them full policy). So the argument that voting gets in the way of writing things down is completely invalid.
   • When someone considers some "minor" change to be not "substantiative", the probability that it is in fact minor and does not seriously break something is a close approximation to zero. (;-) Robert Ullmann 11:31, 5 April 2009 (UTC)

     • You make excellent points. The problem is that votes take forever, and that (with all apologies) people vote "oppose" for bad reasons — things like "yes, these changes are good, but they don't cover some case that isn't covered currently, either." Atelaes and Visviva's proposal solves this issue by removing any expectation that the voted-on proposal be perfect. Another problem is that people vote "oppose" for minor issues of wording; in theory, the process of discussion should resolve that, but in practice, there seem to be a lot of editors who don't scrutinize the wording carefully until the vote has begun, at which point they object. (It's totally understandable — who wants to spend time reading each new draft of a vote? — but it's unfortunate.) Atelaes and Visviva's proposal solves this issue by allowing wording to continue to be tweaked, as long as there continues to be consensus. And all these problems deter people from starting any votes, even ones that appear (in BP discussion) to be unobjectionable.

       Of course, the current proposal doesn't solve the biggest problem, which is that ELE and CFI aren't actually policy at all, but rather a poor approximation to policy, such that requiring >75–80% consensus doesn't work (there's no stance that has anywhere near 75–80% consensus, at least none that's as detailed as CFI and ELE currently are); but I don't have a solution to that problem, and apparently Atelaes and Visviva don't, either.

       —Ruakh_TALK 21:57, 5 April 2009 (UTC)

     God, I hate the indentation here.

     A change can be minor in one sense and not in another. Misspellings are minor, but they are also very obviously wrong. Over time examples can become outdated, for instance if a word was thought to have a single etymology and is discovered to have two, or if a dated term becomes archaic, or if a brand name becomes genericized. In such cases another example that illustrates exactly the same point is minor in that it doesn't change the meaning, but significant in that it keeps the documentation valid and current. There are improvements to explanation that do not require a change in intention. Remember that we used to have some stupid word "hruck" or something as an example? I don't think that was ever voted off, it just got changed through wiki collaboration.

     If the proposed changes you mention were the substantial ones that come to my mind, then they would not be affected by this vote, because all changes would need to be discussed, and substantial changes would still need to be voted on. DAVilla 21:38, 7 April 2009 (UTC)

   I alert you attention to this edit which you made just 10 days into your "very strong oppose" vote and which occurred without discussion that I was aware of and certainly without any vote. DAVilla 08:46, 4 May 2009 (UTC)

2. Oppose --Duncan 22:11, 5 April 2009 (UTC) (After editing conflict, but somehow I found I don't have to do much rewording). The phrasing "It should not be modified without discussion and consensus. Any substantial or contested changes require a VOTE." is so vague that it can be interpreted in whichever way whoever likes to. And anyway the current wording uses the expression should not, rather than must not, so I can't see how any de facto policy, if rewritten into eg ELE, would meet a rollback. And even if it did, so much the better for bringing about a discussion on why we claim one thing and keep doing another. --Duncan 22:11, 5 April 2009 (UTC)
3. Weak oppose per RU.—msh210℠ 20:03, 8 April 2009 (UTC)
4. Oppose — Carolina wren discussió 19:21, 9 April 2009 (UTC) I'm not sanguine about this because it encourages a "shoot first ask questions later" philosophy inappropriate for policy changes. I'm all for eliminating the need to VOTE on minor changes, but I'd much prefer that they be brought first to the Beer Parlor and then implemented, rather than the other way round. — Carolina wren discussió 19:21, 9 April 2009 (UTC)

   In fairness, the policy promotes BP discussion if anyone disagrees with the edit (which will probably be fairly often). -Atelaes λάλει ἐμοί 19:43, 9 April 2009 (UTC)

   Policy documents shouldn't be the scene of edit skirmishes, that's what BP is there to smooth out, so I strongly prefer an "ask questions first then shoot" type policy. I understand the frustration of our current policy of "file a requisition in triplicate for the bullets and then wait for them to be delivered from the supply dump before shooting", but this proposed change goes too far in the other direction. — Carolina wren discussió 20:41, 9 April 2009 (UTC)

   Well, I don't think anyone wants to see edit wars on CFI. Perhaps the wording should be "It should not be modified without discussion and consensus?" :P Bear in mind that the vote is on the wording initially proposed, not my interpretation of it. -Atelaes λάλει ἐμοί 21:09, 9 April 2009 (UTC)

   Right now the process is more like "ask questions first, get no answer, then get shot when you make the change". Admins don't rely on the policy pages as much as other users, only when it comes down to a dispute, but not as a reference on the side for creating entries. The information gets out of date, and the people most eager to make changes are the ones least trusted to do so. DAVilla 03:08, 10 April 2009 (UTC)

5. Oppose Nadando 03:20, 10 April 2009 (UTC), as per Robert Ullmann. Nadando 03:20, 10 April 2009 (UTC)
7. Oppose DCDuring TALK 01:40, 11 April 2009 (UTC) per RU. DCDuring TALK 01:40, 11 April 2009 (UTC)
8. Oppose per Robert Ullmann. The uſer hight Bogorm converſation 22:24, 28 April 2009 (UTC)

Abstain

1. Abstain Dan Polansky 06:22, 6 April 2009 (UTC) A note on wording: re: "Any substantial or contested changes require a VOTE.". This I read as "Any change that is (a) substantial or (b) contested requires a VOTE." From this logically follows a weaker requirement that "If a change is substantial, then it requires a VOTE" (regardless of whether it is contested), which only slightly tweaks the current meta-policy, only allowing changes in formatting to be done without voting. If that was the intention, then fine.

   A substantial change to the current meta-policy would be "Any change that is substantial and contested requires a VOTE", or, put differently, "If a change is not substantial or it is uncontested, then it does not require a VOTE". --Dan Polansky 06:22, 6 April 2009 (UTC)

   As intended. If a change is substantial, then we require a vote. We don't want people to go off rewriting policy pages on their own without presenting them first. DAVilla 21:14, 7 April 2009 (UTC)

   Then I don't understand the motivation for this vote, or I don't understand "substantial" in "substantial change". I quote: 'Let's face it folks, our policy documents are generally out of date. I can't count the number of times that I've had to tell a new contributor, "Well, even if it's not written down anywhere, it's still policy."' If any substantial change, disputed or not, requires a vote anyway, then the problem with policy documents easily getting out of date is not being addressed by the proposed change of the meta-policy. --Dan Polansky 10:15, 10 April 2009 (UTC)

   Well, one thing is that smaller things can be addressed without the rigors and time of a vote. Often all that is needed to keep a policy page up to date is small changes as practice incrementally changes. -Atelaes λάλει ἐμοί 10:17, 10 April 2009 (UTC)

   I think I really misunderstand "substantial". I thought "substantial" meant a change in substance of what is being said. Thus, a semantically invariant rephrasing would not be substantial, while any change that has a real effect on how Wiktionary entries are formatted would be substantial. But from your response, I now infer that by substantial you mean major as contrasts to minor. Some examples would be helpful, I think. Like, is adding the template {{t}} to WT:ELE considered substantial? And what would be an example of a change that is not substantial and thus not requiring a vote if undisputed? --Dan Polansky 10:30, 10 April 2009 (UTC)

   Certainly minor rewordings and grammatical fixes are minor, and if this vote passes, can be changed without a vote. Then again, I think we've always done things that way, contrary to officially stated policy. I would argue that adding {{t}} to ELE would not be a substantial change at this point, because it is rather clearly the current practice. Something that would clearly be a substantial change would be a change which sets a standard which does not exist in practice, or which differs from practice (such as the addition of numbered pronunciation headers, for example; this is done by some, and not done by others; there is no standard practice). As for the boundary line between the two, I think that's something we're going to have to figure out as we go along. This proposed change of this vote is to simply acknowledge that there might be some changes which don't require a vote. The current wording allows absolutely nothing without a vote. -Atelaes λάλει ἐμοί 10:46, 10 April 2009 (UTC)

   I think I understand better now. Still, your "I would argue that adding {{t}} to ELE would not be a substantial change at this point, because it is rather clearly the current practice" makes me wonder, as whether a change of a policy document is major or minor, significant of insignificant, substantial or unsubstantial, is not driven by the state of current practice AFAICT. If a policy document is frozen, the current practice can drift arbitrarily far away from it. Thus, aligning a policy document with the current practice may require arbitrarily big changes to it. Put differently, if the current practice differs from the codified policy in a substantial way, then the required adjustment of the policy is substantial.

   That said, I understand that the proposed wording at least creates a possibility for all kinds of changes being interepreted as not "substantial", and thus not requiring a vote, which was impossible before. It seems to me, though, that if by "substantial" is meant "differing from current practice", it would be better to replace the former with the latter in the wording of the meta-policy. --Dan Polansky 11:21, 10 April 2009 (UTC)

   The "or contested" bit was meant to be a sort of safety net, for helping to decide what's substantial and what's not. If someone reverts your edit, that's the best evidence that your change was substantial, which means you need to gain consensus, i.e. you need to start a BP convo and try and convince people to agree with your changes. -Atelaes λάλει ἐμοί 12:16, 10 April 2009 (UTC)

   I would say that according to the proposed wording "It should not be modified without discussion and consensus." you should gain expressed consensus before making any, however unsubstantial, edit, shouldn't you? The only difference that I can see is that the edit could be agreed upon in, say, BP, without a further need for bringing it here to Votes. --Duncan 14:25, 10 April 2009 (UTC)

   Well, we don't have to be overly dogmatic about this. Take the {{t}} thing for example. Even if we hadn't had an official "let's add {{t}} to ELE" BP discussion, if it's being consistently used by most editors, then clearly there has already been discussion and consensus (in fact, I remember when the BP was nothing but discussion about t). If someone reverts you, then clearly we need to have that specific discussion. Now, I want to make something clear here. As far as I can tell, no one, not one single person, wants policy pages to become grounds for edit wars. So, one revert stands, until significant progress in consensus has been made. I think that, after a while of being able to modify policy pages without votes, experienced editors would know when a change requires discussion and when it doesn't. -Atelaes λάλει ἐμοί 19:42, 10 April 2009 (UTC)

   Adding {{t}} to ELE is not going to be a “controversial” change because it is clearly a current practice, as Atelaes says. (Someone reverting the page would indicate we might be wrong about that, but if it had to be voted on I feel it would easily pass.) As to whether the change would be “substantial”, I say no because it better exemplifies the description than the existing wikitext. That statement again borrows on the fact it is current practice.

   Given that {t} was created with so much input in WT:BP as an intended standard, "discussion" could comprise just a statement on the talk page that an editor intends to update it, standing for a week with no response potentially. DAVilla 22:29, 10 April 2009 (UTC)

Decision

(We have 16–7 (69.565%).)

Sadly, I'm pretty sure that's a fail (no consensus), at least by the standards used on Wikipedia, although I don't know if we've officially decided to accept those standards. Anyone is free to disagree with me on this. -Atelaes λάλει ἐμοί 10:25, 2 May 2009 (UTC)

I disagree. Two thirds is the only number I've ever heard here. What are you thinking, three quarters? Way too strict. Seven tenths? Damn the decimal system, that's just too strange from a mathematical perspective. Five sevenths would be more harmonious (to borrow musical terms) though I somehow doubt that's been proposed. Yeah, even outside of the numbers it isn't all that cut and dry to say the least, but two thirds is what we would normally aim for. I'd prefer if someone who objected were kind enough to pass this though, so that it isn't contentious. User msh210 did say weak oppose, if 16-6.5 (71.111%) is any more convincing. DAVilla 08:23, 4 May 2009 (UTC)

This seems to be in the nature of a constitutional/rules change. In most institutions the requirement would normally be for more of a plurality than for substantive change. DCDuring TALK 10:13, 4 May 2009 (UTC)

I agree with DCDuring. I voted to oppose, so will not fail it at these numbers, but I think it fails.—msh210℠ 15:15, 4 May 2009 (UTC)

Yeah, I wouldn't feel comfortable passing this with less than 70–75%. Robert's edit is telling, but doesn't actually invalidate his vote. —Ruakh_TALK 21:15, 4 May 2009 (UTC)

Okay. Considering the gravity of the vote, I guess a higher standard like 3/4 isn't that out of line. DAVilla 06:05, 5 May 2009 (UTC)

Vote fails 16–7-1. :-(   —Ruakh_TALK 21:15, 4 May 2009 (UTC)

In case anyone was wondering, this was the source for numbers that I was using. It appears that 75% is consensus, less than 70% is not, and in between is a grey area. We really ought to figure out what our own standards are, for future reference. -Atelaes λάλει ἐμοί 19:04, 5 May 2009 (UTC)

Update: I've stricken my "fails" comment, in light of my remembering/realizing today that we never actually approved the current editing-is-forbidden text atop these documents, and a quick glance at the edit history of any of them suffices to show that said editing-is-forbidden attitude was never thoroughly accepted by the community, so it seems wrong that a more-than-two-thirds majority should be unable to overturn it. If someone else wants to close this as a "fails" vote, I do think there's justification for that (and obviously I won't revert you); but I've decided I'm not comfortable doing it myself. (See current BP discussion for more on this topic.) —Ruakh_TALK 01:08, 13 May 2009 (UTC)

Links to the BP discussion: Reverts at Wiktionary:Criteria for inclusion. with a "subchapter" at Moving on --Duncan 15:32, 16 May 2009 (UTC)

Vote passes. At the time, 69.565% was debatable, and this vote never got closed; but since then we have passed votes — much more significant votes — with only two-thirds' support, so I'm hereby belatedly passing this. —Ruakh_TALK 22:42, 22 March 2012 (UTC)

I'm unpassing this vote as it is being discussed at WT:RFDO#Wiktionary:Votes/pl-2012-03/Minor ELE fix. --Daniel 23:10, 22 March 2012 (UTC)

To try and please as many people as possible, I will declare this a no result, this vote has neither passed nor failed. Mglovesfun (talk) 11:42, 26 March 2012 (UTC)