Wikipedia:Pending changes/Request for Comment 2014/Closure

Discussion
Alright, I'm not sure why this page was deleted (despite the nonstandard name), and it can be moved if other closers like. I proposed at the talk page of the RFC that we begin by totaling all !votes, explicit or not, and then evaluate strength of arguments. I saw no objection to that at the time, so I'll get started. Pakaran 23:57, 8 March 2014 (UTC)
 * Sounds like a plan. As I think I mentioned already, I've been holding off, waiting for you both. For one thing, since I closed the last one solo, I thought it polite to let you all note your read of things first. (Not sure if that came out on paper as it was meant in my head, but hopefully you understand : ) - jc37 07:16, 10 March 2014 (UTC)

Pinging my colleagues to review table below:,  D  P  00:41, 22 March 2014 (UTC)
 * I'm not going to make excuses; RL got distracting this week. I can, however, promise to look this over before Sunday.  Pakaran 01:54, 22 March 2014 (UTC)
 * Briefly, I understand what you're saying regarding implicit opposes counting less, and that seems to make a great deal of sense to me. From the running totals kept, etc, I guess I'd gathered the impression that a lot of editors expected that (for example) an !vote for #4 would count against every option which presupposed that the feature be usable under at least some circumstances.  However, I'm a latecomer here, and willing to defer on that point.  This also raises the issue of whether a "strong" !vote in favor of 4 (again, an easy example to discuss) should count less than a marginal, poorly justified, !vote in favor of a contradictory proposal.  Pakaran 02:00, 22 March 2014 (UTC)
 * Hi, as it's now Sunday, I'll drop a ping to you both. (@Pakaran and @DangerousPanda) As I have said several times already, I don't mind waiting, but out of respect for all the work of all those who have contributed to the various discussions, we should probably get to working on the overall closure sometime in the near future.
 * I welcome thoughts on this from both of you, and look forward to working with you both on this closure. - jc37 23:55, 30 March 2014 (UTC)
 * I'm here, and agree that we need to keep pressing forwards. RL has been slightly busy for me, but I can now dedicate the time to get this done, hopefully sooner rather than later.  Thanks for the ping, and I very much understand the community's frustration.  Pakaran 18:27, 1 April 2014 (UTC)

Closer discussion only
Ok, following extensive re-reading, I have to say a few key things:


 * 1) I'm not comfortable with "implicit opposes" as having significant weight - explicit is vital
 * 2) There's a range of "total votes" per proposal - and this might be because of when they were made, and how many people came back to re-vote
 * 3) Is anyone truly comfortable with the concept of a 50% plurality?  I'd be more comfy with 65%, but could be convinced otherwise.

Reusing the table data (which I have yet to verify) I see this:

So, this leaves either 8 or 5 ... depending on whether you go with a simple majority, or a 2/3 majority. D P  00:40, 22 March 2014 (UTC)

Next steps
D P  00:40, 22 March 2014 (UTC)
 * Step 1: decide whether simple majority or 2/3 is best choice
 * Step 2: doublecheck the numbers
 * Step 3: after narrowing down to the 8 or 5 based on step 1, review the arguments for the "first cut successful" proposals for validity
 * Step 4: review each of the proposals that meet the valid argument in relation to the others to ensure that there are no duplications or (*gasp*) contradictions
 * "In general", while I don't "count votes" per se, in big community-wide "will the devs implement" sort of RfCs, 2/3 (sometimes 60%) tends to be passing, and usually anything between 1/2 and 2/3 tends to be closer discretion, based upon weighing the arguments/statements. That said, reading the comments is obviously important, and we of course don't count votes : ) - jc37 21:57, 22 March 2014 (UTC).
 * Nope, we don't count !votes :-) D  P  00:00, 31 March 2014 (UTC)
 * No we don't! I think, however, that throwing out proposals that do not pass at simple majority is only reasonable (unless we find exceptional reason to do otherwise).  This does not mean that 50% is a plurality; even nearer the mid fifties, there's a presumption of not passing.  Note that implicit !votes, even if we don't strongly weight them, certainly seem to suggest that Proposal 4 should get a by from any early elimination (should its total fall that far).  Anything else strikes me as unconscionable, and while I haven't yet checked the count, we'd have to be blind not to see this issue coming.  Pakaran 17:10, 10 April 2014 (UTC)
 * Not a fan of vote counting, but ok, as a comparison. Re-reading over each of the proposals that didn't achieve even 50%, I don't believe any of them has consensus. As far as I can tell this merely includes 5,8, and 12 (presuming I have not miscounted). So to be more specific, I see 5 and 12 having no consensus, and 8 being outright opposed. (Which incidentally, makes 9 moot, as it relied on 8 as proposed.) - jc37 18:28, 11 April 2014 (UTC)

Tally confirmation
I'll start some notes on a checked tally of explicit !votes, without looking at the above table. History since discussion ended has not changed any !votes, so we can trust the current version of the page. If we could not, frankly, I'd be astounded, but a sanity check is only reasonable.
 * Proposal 1 has 65 supports, of which it is a second choice to proposal 2 in two !votes, 12 and 21. It has 34 opposes, and zero neutrals.
 * Proposal 2 has 62 supports, of which one, #29, is a second choice to proposal 4. It has 23 opposes, and zero neutrals.
 * Proposal 4/0 has 36 supports, several of which note explicitly that they are not implicitly opposing certain proposals (there's unsurprisingly duplication with the list in section 3 of the RFC): #17 supports proposals 12 and 14, #34 supports 5 and 12. I need to go AFK now, but my next order of business is to check arguments by the other several few !voters listed as in contradiction, and total the opposes and neutrals.  Pakaran 17:55, 10 April 2014 (UTC)

Some thoughts
Just as background, the source of this set of RfCs, was the previous RfC. The result of which was: "There is only a consensus for implementation if and only if an rfc concerning criteria for its use gains community-wide consensus first".

So one thing I think we need to keep in mind is whether a.) any of the individual rfcs' proposals were successful, and b.) whether any of those (singly and/or cumulatively) fulfill that requirement.

Another thing I think we should be aware of (as I mentioned elsewhere) is that "the question" #4 asks (through a statement: "No change in current policy. PC2 is not to be used.") asks whether consensus has changed, concerning whether PC2 should be implemented. Which is why we see opposers and supporters in swapped positions (various supporters of PC2 are seen opposing #4, and vice versa). - jc37 18:16, 11 April 2014 (UTC)

I was going to leave the following comment on the talk page, but since I am not the only closer, it seems more appropriate to post it here so that you two can comment upon it : )

When closing an rfc discussion, it's generally based upon what is being requested for comment upon.

So in all but #4's case, the requests all presume that: "in the case of pc2 being enacted/implemented, then should x criteria/structure/whatever be implemented?"

So an editor choosing to not comment, or choosing to oppose any of the other proposals (besides #4) has little to do with whether pc2 should be implemented.

That said: first, if none of the proposals met the "criteria" requirement of the previous rfc, then presumably pc2 would not be enacted, and second, if in reading someone's comments (even if in a discussion section) they say they overall oppose pc2, but that person chose to not comment under #4, their comment might be something to consider for #4. (depends on the comments etc. on a case by case basis.)

I'm sorry if there are those who may not like this, but (with the current exception of when an editor is requesting a post or user-rights, like rfa or arbcomm) it's how discussions are closed on Wikipedia. The closer reads the discussion, determines the local consensus, while keeping in mind the broader common practice/policies/guidelines/etc.

We don't read minds here (even if we could : ) - We pretty much are left with what is actually on the page.

I welcome your thoughts on this. - jc37 01:17, 16 April 2014 (UTC)
 * Absolutely. I've been trying to feel things out during the process, which is grossly sub-ideal.  My collection of statistics, then, was apparently poorly founded.  I don't have a dog in this fight, and if anyone believes I do, they're welcome to contact me (as always).  One of my personal rules of thumb is not to look at numbers beyond, at most, whole percentage points (so, as a completely arbitrary example, 75.3% is not distinguishable from 74.6%).  If you're doing so, you're (even here) down to a few individual !votes or less, and that's pretty clearly a horrific way to judge consensus.  As for the "criteria" requirement, that is our task, and the phrasing in the prior RFC contains an "if and only if".  It appears to me that that means that, if there are criteria that reach consensus, then use should be approved?


 * On a more practical matter, it also appears that we're all hesitant to take the lead, for different reasons, and that is for better or worse likely to continue making our progress rather slow :/ Pakaran 03:45, 16 April 2014 (UTC)
 * Well, with all that in mind, I'll go ahead and start what I think may be 2 relevant threads as a start for us to discuss in. (Pinging you both : ) - jc37 21:34, 16 April 2014 (UTC)

Does proposal 4 have consensus?
I think we've already gone into the question/statement of #4 above, so to try to be simple in response: If I were closing this individual proposal as a stand alone RfC (following up the previous RfC), I would close with a result of No consensus, and (per what we've already noted above), that as the result is no consensus, it doesn't meet the standard of consensus has changed, so the previous RfC stands. And this without needing to go into whether this rfc or that one had a broader community turnout and thus whether this most recent discussion would be representative of the "wider community" (local consensus vs community consensus).

And now, as this is not a stand alone proposal, if we determine that the discussion of #4 did not overturn the previous close, then we next would need to determine if the "criteria" in the "if and only if" is fulfilled. (Which is the next thread below.) - jc37 21:34, 16 April 2014 (UTC)

Does proposal 7 have consensus?
I picked #7 because the result (to me anyway) appears obvious regardless if we read for content (as we should), or even counting "votes" (as we shouldn't : )

(If either of you prefer to pick a different proposal to start with, I welcome it : )

If number 7 has consensus, then I presume the "if and only if" criteria of the previous rfc is met. And then it's merely a question of looking through each of the rest of the proposals to determine if any of those also had consensus.

And yes, I think #7 had consensus, not the least of which because, as I read it, editors with rather divergent opinions supported it (albeit, some more strongly or weakly than others). - jc37 21:34, 16 April 2014 (UTC)
 * Proposal #7 has a truly overwhelming consensus, and it appears to meet the if-and-only-if criterion. If we were counting !votes (which we aren't :) ) it would also be so much stronger than the consensus in favor of proposal #4 that it would "win", but that's an entirely separate matter.  /, I'd appreciate your thoughts.  Pakaran 00:25, 18 April 2014 (UTC)
 * Yes, it has pretty massive consensus, even when adding in the "none of the above" proposal D  P  00:27, 18 April 2014 (UTC)

Proposals 1, 2, and 14
As I read these, they are mutually incompatible (unlike the other surviving proposals, which are independent of this triplet). Also, after reviewing all the arguments several times, it appears to me that Proposal 2 has the strongest consensus of the group. Of course, I welcome comments from both of you. Pakaran 01:17, 18 April 2014 (UTC)


 * Ok, to try to respond to this from my perspective, let's pretend (for now) that all three (1, 2, and 14) had consensus. I'm not saying they do, this is just for the sake of explanation.
 * So 1 states:
 * PC2 should be usable in the same situations as PC1 (i.e. cases of persistent vandalism, BLP violations, or copyright violations), but only if the situations are being caused by autoconfirmed users and blocking them is an ineffective solution.
 * And 2 states:
 * PC2 should be an option for pages that qualify for and might otherwise be fully protected instead, e.g., to deal with certain vandalizing sockpuppeteers without locking everyone out completely and letting the socks "win" a big badge of disruption.
 * So just to take these two first, the first says it's intended for situations caused by autoconfirmed users (which I think is merely a reflection of the technical usage of the tool) followed with a statement of "but only if x is ineffective"
 * The second, instead of focusing on a qualifying type of user ("autoconfirmed"), focuses instead on a qualifying type of page (which "might be otherwise protected instead").
 * I don't think these are mutually exclusive. If both are determined to have consensus, the resultant guideline could state both sentences and not be contradictory.


 * As for 14, while 1 and 2 are rather broad in definition, and merely use a few examples each, 14 - which doesn't appear to contradict 1 or 2 - lays out more specific instances of when it may be used.


 * Though I notice 14 includes more than that, for example there is a touch of a few other proposals throughout, such as #7 in 14 though it doesn't specifically specify "a year".


 * But before I go too deep into 14 and what is said there and its perceived compatibility with other proposals, I think I'll just pause and say that I don't think 14 had consensus. And there were concerns in both support and oppose about what was stated and the wording, etc.


 * So anyway, the shorter version: I don't think 1 and 2 are mutually exclusive or incompatible, and I don't think 14 had consensus. - jc37 15:35, 19 April 2014 (UTC)


 * Alright, to start with, I agree completely regarding Proposal 14. My statement regarding the "strongest consensus" was sloppy, and could as easily have read "Proposal 2 has consensus, which Proposal 14 does not, and by my reading, Proposal 1 fails due to being significantly more weakly supported than Proposal 1 and incompatible".  From your statement, we must apply the stronger reasoning of "Which, if any, members of the set, considered on their own merits, are passing".  This is not true of Proposal 14, even disregarding the prohibited !vote by a banned user recently noted on the RFC page.


 * It appeared to me that a significant number of !voters were acting under the impression that only one of Proposal 1 and 2 would be adopted, and I was deliberately attempting to avoid looking at the various "implicit opposition" data. As such, it was not clear to me whether that data had been seen as only including users who supported Proposal 4 (and as such opposed the various "PC may be used in X conditions" Proposals), or also included one or more other conflicting pair(s).  So, having read your comments, I will reassess the issue of, firstly, "are Proposal 1 and 2 compatible".


 * I can indeed imagine a closing statement in which usage were allowed in either circumstance. As perhaps an understatement, this doesn't seem to have been a universal assumption going in, however, and I would argue that we do not need to reach this issue.


 * Proposal 1, by the ~2/3 standard discussed above (which is higher than the margin I suggested initially), would be marginally if at all passing if we were counting !votes. Additionally, a great deal of its support is lukewarm, and often as a second choice to Proposal 2.  Therefore, my view is that, regardless of many of the not-so-minor details I've discussed, Proposal 2 the only Proposal named in this section which has consensus, but a case can certainly be made for the combination of 1 and 2. Pakaran 18:53, 20 April 2014 (UTC)

13
Do you think 13 has consensus? - jc37 15:41, 19 April 2014 (UTC)
 * 13 has a presumption in favor of passing due to the volume of opposition to all use and the results of past RFCs, coupled with the fact that it represents a restriction on use. It has support based on a number of grounds.  I do not feel comfortable giving significantly reduced weight to the proportion of the opposition based on (paraphrasing) "there might be a special case where we need to stop an edit war"; an RFC outcome banning use, by definition, would ban use in any such special cases that exist.  However, I believe that 13 does have consensus.  Pakaran 19:16, 20 April 2014 (UTC)

15
I'm abstaining from comment on 15. So I'll leave you two to determine the outcome of that discussion. - jc37 15:41, 19 April 2014 (UTC)

Final closure text
Based on the comments above, I think we're nearing the end. Is it possible to get this wrapped up and "closed" by the end of the week? We can pick a day and time for the 3 of us to be here if you like.
 * User:Pakaran and User:DangerousPanda

Based on the discussion "so far", I'd suggest we each write first draft closing statements, and then we work together to unite them into a single statement we can agree upon.

I welcome your thoughts on this. - jc37 19:29, 25 April 2014 (UTC)

Okay, I guess I'm up to bat
My call is that, even for you voters who gave long rationales, I don't know enough about where you're coming from to allow me to weigh your thoughts against those of other voters, so I want to start off by asking anyone who wants to respond to do so (closer input goes on this page, anything can go on the talk page) ... and if that goes well, I'll be polling other voters for their thoughts.

I'm asking for more information because it's very difficult to pick out any consensus. The one thing that shines through is: some of the supporters and some of the opposers see a possibility of harm to WP if we make the call the wrong way, and no one has given a clear argument that harm isn't possible, in either direction. Some of the supporters are concerned that damage could be done if we can't keep up the game of whac-a-mole with the more obnoxious serial offenders, and they want some kind of additional tool. (Some opposers have asked for evidence, and without getting into that question, it's enough for me at the moment that I don't see convincing counter-evidence.) Some of the opposers are concerned that real damage could be done because we're being asked to give a thumbs-up to PC2 here before we've answered any of the important questions, such as: what is PC2? (I interpret this as: what are the closest analogies to PC2 already in play on WP, and how will PC2 be similar to or different from those?) What do we do if PC2 is misused? What's the actual wiki-wide consensus on scope now, and what might it become? Who are the PC2 reviewers, and how do we vet, train and dismiss them? I get the objection of the supporters to this point ... that if we wait till we know everything before we proceed, that's the same as saying we'll never do it. I'm asking a different question, namely: do I need to ask more questions, or can I just go ahead and make a call? Is there even a small chance that I'll damage WP by acting rashly? (A "small chance" is too big. We regularly have "big" RfCs with the potential for big impacts. 20 RfCs, each with only a 5% chance of doing harm, will produce a net likelihood of harm, and that's not acceptable.)

The information I'm looking for from the voters is: what have you seen on WP that leads you to support or oppose? PC2 is a proposal to put one or more people "in charge" of an article for some period of time, vetting a subset of the edits. This happens all the time on WP already, in various ways ... we have PC1, peer reviewing, Good Article reviewing, various wikiproject initiatives, lots of RfCs, noticeboards, etc. None of these things are PC2 ... but they share with PC2 the basic idea that people are asked to make some basic calls on edits after getting editor input. We've all seen some of these processes working flawlessly, and some not so much. I'm betting that our individual experiences are influencing what we expect will happen with a tool like PC2 ... but how? What have you guys seen go right, or wrong? I need to know more about where voters are coming from. I'm not going to negate what you say ... but if some consensus develops from the discussion that people are expecting that PC2 will be more of the same of what they've seen already, and I know what it is they've seen, then I can use that consensus to help me weigh the vote rationales from the RfC. - Dank (push to talk) 19:21, 29 May 2014 (UTC)
 * To me, it sounds like you essentially just asked for a new RfC on PC2 in order for you to discern a close for this RfC : )
 * What we need to do as closers is to look at this RfC (really, this "set" of RfCs) and determine what does and/or does not have consensus. We do this in light of this RfC, and in light of the broader community guidelines/policies/common practice/previous consensus/etc.
 * That said, looking beyond that, am I correct that you are suggesting that you cannot determine consensus of this RfC due to lack of information?
 * If so, what in particular are you looking at and what information deficiencies are you noting? (this is sort of a "what and where" question : )
 * And I'm not trying to put you "on the spot" : ) - To give an example, while I think others suggested that 13 had consensus, I really didn't think it did. I don't think it precludes further discussion, but I don't think 13 as it stands right now has consensus.
 * And finally, I asked this on the discussion page, but you might have missed it: Please read over the above threads, and please comment, I'd like to discuss them with you : ) - jc37 19:47, 29 May 2014 (UTC)
 * I see #7, #13 and #15 as "what ifs", so it doesn't make sense to me to deal with those before dealing with the main question in #1 and #2. Even though #2 gets a significantly higher numerical percentage than #1 by virtue of fewer opposers, all of the oppose rationales for #1 (except for Formerip's and King_of_Hearts's) apply equally to #2, so unless I hear something different before I make a call, I'm going to assume that #1 and #2 have roughly the same levels of support. I'd like feedback on #14 as well: I'm not saying it had consensus, I'm saying it got a slightly lower percentage because it's #14 and because it was more specific and thus easier to critique, but the anti-14 rationales are similar to the anti-1 and anti-2 rationales.
 * I'm open for any amount of discussion without further feedback, although I've got a slight preference to do that with the closers by email, because I don't want to put ideas into voters' minds (until I get the feedback I want). The problem isn't that I can't draw any conclusions ... I can draw a hundred conclusions that are consistent with the information on the RfC page, because the information is too sparse. I don't need a whole new RfC, I just need to get a feel for what analogies the voters are drawing. - Dank (push to talk) 20:44, 29 May 2014 (UTC)
 * Though that's not my normal preference (I tend to favour full on-wiki transparency for most things), I think I understand your concern. So please feel free to email me (and presuming they have email enabled, copy to the other closers as you prefer). - jc37 20:54, 29 May 2014 (UTC)

I'll start working on a closing statement here, one bit at a time. It's intended to be my own statement, but I'm very open to feedback both from closers (here) and anyone else (on the talk page), and I'll do my best to take your comments on board. OTOH, note the pushback I got when I asked for more feedback on the talk page ... the point is being made that the time for discussion is past, and that taking on more feedback now risks minimizing and disenfranchising the voters who don't happen to be participating now. So ... yes, please do comment, as a way of giving me a clearer understanding of what you've said before, including in the RfC, but understand that no feedback I get now can override anyone else's comments or rationales; everyone gets a voice.

It's apparent from the discussions above that #1 and #2 are going to pass. I think the end result is going to be better if I concur on a majority opinion rather than go off in my own direction, and that's a sufficient reason for me to get on board with #1 and #2. It's a very close call anyway, in my mind ... more of a "tie" than a "no consensus" (yes, there's a difference). Many of those who do favor "no consensus" accept that that result would not actually settle anything, and could easily lead ... has in fact led ... to uses of PC2 that didn't respect our policies and community consensus (and that's all I'm going to say about that).

So, I'm on board with proposals #1 and #2, and I'll do my best to figure out what they mean ... on the surface, it seems clear enough, but the devil is in the details. As for #7, #13 and #15, I see lots of support and not much opposition. The other proposals fail.

The big problem for me is that there are a lot of points being made where I think I understand what people are saying, but I can't be sure and the discussion is kind of sparse. The closing statements of this RfC and the next RfC (mandated by proposal #15) are the best places to gather together whatever consensus appears to exist. When we get into the grind of making individual calls at WP:RFPP, re-fighting old wars will only hurt rather than help ... admins are probably going to insist, appropriately, that each case needs to focus on the individuals and the issues involved, and not on extraneous points about the big picture and where it's all leading.

More soon. - Dank (push to talk) 15:27, 30 May 2014 (UTC)

Getting started on a closing statement, jc's working on one too, then we'll negotiate. Feedback on the talk page would be great ... as you can see, all of the closers have gotten a lot of feedback on the talk page for months now, and their comments here are better for that feedback, and I've gotten feedback for almost two days that has been really helpful. And of course, anyone is free to read the talk page and agree or disagree with anything there or here, I've gotten really good advice and I'm sure I'll get more. [moved from section below]. - Dank (push to talk) 17:46, 1 June 2014 (UTC)

Dank's closing statement
I've been concerned for years that the PC2 issue has been so divisive, because productivity suffers in polarized volunteer organizations. My main thesis is that opposers and supporters have been entrenched and unwilling to listen to the other side, and still are. In most protracted conflicts, the longer the conflict goes on, the farther apart the two sides get, the less they listen to each other, and the more any attempts at negotiation come to be condemned. There's no point in calling this "bad behavior"; we all do it, under the right (or wrong) circumstances. But Wikipedia could be hurt regardless of which way we go with PC2, if we let one side "win" and one side "lose", if we base our decisions on only half of the available information and wisdom of the voters.

In support of my thesis, I can't find evidence in the support sections for proposals #1 and #2 that the supporters read anything written in the opposing sections, apart from two supporters rebutting opposition comments ("struggling to understand" and "lack of use cases"). Supporters may have read opposition comments, of course, but it seems significant that it doesn't show in those two sections. It's not because the supporters were generally in a hurry or uncommunicative ... their comments show that they're paying a lot of attention to each other, just not to the opposers.

There's lots of evidence that the opposers of #1 and #2 read the supporting rationales ... with great skepticism. Some are dismissive; others make comments that have the effect of supporting the more aggressive opposers. For example, some opposers have claimed that #1 and #2 contradict each other, which would imply that most of the supporters weren't bright enough to see an obvious contradiction in their own statements. (Of course, #1 and #2 don't actually contradict ... neither one says "only" in the main clause ... and the supporters of both #1 and #2 made it very clear that they wanted PC2 to be available in either case.)

The supporting percentages in this RfC among people who didn't comment in last year's RfC on PC2 may be indicators of what we'll see in future RfCs: 77% for proposal #1 and 84% for #2. Time will tell. I didn't use those figures to make any calls in this RfC.

...

Jc has just posted his close, finding no consensus for PC2 at this time, and that some additional work refining and clarifying proposals would be needed for consensus. That's my position as well, and I concur word-for-word on his main findings, the three paragraphs that begin "Overall" and end with "at least once a year". I agree that the numbers were close (66% support for proposal #1 and 73% for proposal #2), but all of the oppose rationales for #1 (except for Formerip's and King_of_Hearts's) apply to #2 and to PC2 in general, so #2 has roughly the same support as #1. Some supporters expressed reservations, and there were 14 votes in proposal #4 that implied additional opposition to #1 and #2 (but 3 votes that implied additional support for PC2). On top of that, the discussion among supporters of #1 and #2 about PC2 itself was lightweight, leaving me wondering if they actually have different things in mind when they think about PC2. It shouldn't be hard, I think, to come up with remedies for this last problem before the next PC2 RfC, and I look forward to those discussions.

I agree with Jc that "discussion concerning any and all criteria" in a future RfC is possible, of course, but for me, it would take a clear consensus to remove or add to the criteria selected in this RfC, and as a closer, I would tend to resist attempts to insert additional criteria in any RfC that happens this year, if the discussion concerning those criteria seems less serious than the discussions we had here. (I've been asked how I can find consensus for the criteria in #1, #2 and #7 without finding consensus for PC2, and that's easy: all but two of the opposers to #1, for instance, rejected PC2 in general, without rejecting the criteria proposed in #1. Among people who actually talked about what would be the best criteria for PC2, almost everyone was in favor of the criteria in #1, #2 and #7.) I would hate for people who struggled with this RfC for a month or more to feel that they didn't accomplish anything, and I think the supporters can be proud that they managed to get the hardest part of the discussion behind them, the selection of criteria. The proliferation of proposals on criteria was one of the main factors that made this RfC difficult for everyone.

There were only 8 or 9 people supporting proposals #13 and #15, and since we're finding no consensus for PC2 here, I don't want to impose those results on future voters. But the comments in #15 illustrate what has also been true in past RfCs: reviewership issues are hard to find consensus on. PC2 won't work, of course, if the community gives reviewers and the people who approve reviewers contradictory instructions, so some kind of consensus will be needed in a future RfC before PC2 can be approved.

I'll post suggestions for future work later this week on the talk page. I just started work as a new closer at the end of May. Thanks for the many helpful discussions; it was a pleasure. - Dank (push to talk) 11:07, 7 June 2014 (UTC)

I just posted a summary on the RfC page. At the moment, PC2 is active on six pages, and this RfC did not address these so-called "IAR applications of PC2". If a future RfC does address that question, reasonable notifications should be made (for instance, to the Executive Director of the WMF, to admins who have been using PC2, and to noticeboards where PC2 is sometimes discussed, such as WP:ANI and WP:RFPP) so that everyone who feels they have a stake will have a chance to participate in the discussion.

I like to end on a positive note, and here it is: conversations on the talk page of this page have suggested that some of the things that were discussed during this RfC may provide food for thought on how full protection and semi-protection might work better, and both supporters and opposers should be commended for their work on this important topic. - Dank (push to talk) 19:00, 14 June 2014 (UTC)

Note/update
As I mentioned in email, "Since you were busy writing your draft, I tried to refrain from posting. (Which also allowed me to spend the weekend with family : )"

I'm working on cleaning up some of the rougher edges of my draft (It was a stream of consciousness thing I wrote months ago, and needs updating, and also I've been trying to adjust it to take into account things since then (Like the various closer comments on this page, etc.) I intend to post it later tonight, and we can go ahead and discuss, etc. And hopefully complete this closure fairly soon. - jc37 16:54, 2 June 2014 (UTC)

Closure draft by jc37
Warning: This is lengthy, but I think that the results needed at least some explanation.

To start off with, let's clarify talk about consensus for a moment.

We use that word - consensus - so much that I think we may sometimes forget what it actually means.

Consensus essentially means that the group in question is resolved in that overall, whatever has been deemed to have consensus, has the most support.

And there is a place that I think sometimes people misunderstand consensus.

The "most support" doesn't necessarily mean the most number of votes stating "support", but the most supportive comments/arguments.

Another thing to note is the difference between no consensus and opposed. If something has consensus, then the premise which has consensus is carried. If the result is that the premise is opposed, then typically in Wikipedia, it is usually suggested to not start a new discussion on that specific question/premise immediately, as that can be seen as disruptive. (There are obviously case-by-case exceptions.)

But what about discussions where there is nearly consensus, of it looks like it's possible that with further discussion there could be consensus, or where the premise/question needs work or tweaking and re-submitted to the community for discussion? Then in situations like that (and others - I'm obviously not being all encompassing in examples : ) - then a "no consensus" result is typically the appropriate assessment. And No consensus does not preclude immediately starting a new discussion. Indeed, it almost never should preclude further discussion, as open discussion is considered one of the hallmarks of Wikipedia. We only frown on repeatedly restarting clearly opposed discussions as this can be seen as a waste of the community's time. consensus can change, but we've generally learned over time, that it rarely changes 30 seconds after a discussion closure.

Now I was the closer of the previous RfC. And so I've felt that it was important to wait to let the other closers give their perspective on this current discussion so to try to not colour their comments in advance.

That time appears to have passed and it looks like we have the comments that we will have on this.

Among other things, in this close, there were (mostly) 3 questions I tried to keep in mind (in no particular order):

Has the requirement of the previous RfC been met?

Over all, does this set of RfCs have an overall consensus?

Do any of the individual RfCs have consensus while keeping in mind the other RfCs? (There is a distinct though possibly subtle difference between these last two.)

Taking the last one first, I'll paste something I typed up previously:


 * 1 and 2 and 7 pass
 * 4 no consensus to override previous consensus
 * 5 no consensus (apparently for want of clarity in application)
 * 8 opposed
 * 9 no consensus (and 8 opposed, so moot)
 * 12 and 14 no consensus based on comments
 * 13 no consensus based on comments - while it may have had a majority of "votes" I don't see a consensus there, especially in light of the other RfCs on the page which had (among other things) more participation.
 * 15 abstain - I abstain(ed) on 15 due to what I said concerning transparency when I volunteered to help close this. that said, I've closed discussions I've commented in before, and if I were to close this one, I doubt many would disagree that there is consensus for 15, though it's really just a suggestion for a new RfC concerning the reviewer user-right.

And so taking the first question at face value: Yes, strictly speaking, there presumably is now criteria for usage of PC2.

But that's only strictly speaking. And I do not think that 1,2, and 7 met the requirements of the previous RfC close. (I'll get to that in a moment.)

Consensus isn't a vote (as we oft have wont to say), and I believe merely counting votes would do this set of RfCs a disservice.

The abbreviations PC1, PC2, PROT, etc all make this all sound like we're merely fitting puzzle pieces in a chart. But as many said in the past RfC and in these, PC2 is different based upon who it potentially affects - all auto-confirmed editors. These are editors who have created an account, and meet the current 4/10 requirement to be considered autoconfirmed by the wiki system. This is a rather broad swath of editors, and when added to all IPs and all unautoconfirmed editors, this becomes a rather vast amount of editors, making this very nearly equivalent to full protection - until a reviewer allows for the edit to go through. (This is important to note, since autoconfirmed editors can edit semi-protected pages.) And so those opposing have shown concern that this violates "anyone can edit". That said, this is almost always the concern when security measures are proposed. Which should not be surprising. In internet security there is usually a question of balance between open and usable, and closed and secure. With the line typically drawn at "secure, but usable", And further as we're a community which has a policy of "Anyone can edit" (with case-by-case exceptions noted), "usable" tends to have a rather broad definition.

So with that in mind, I think the question repeatedly being asked of the community is: Is the line we're drawing here acceptable? Does the security of PC2 meet the balance of "secure but usable"? Does it meet the community's expectation of "Anyone can edit, but we need to keep the community secure  so that anyone can edit undisrupted".

I didn't think that there is overall consensus for PC2 when reading over the last RfC, and I still do not see community consensus for PC2 when looking over this set of RfCs. I do think it's "close" though. With some work refining the proposals for usage criteria, I think this might eventually get to a place where the community says "yes, we can accept this, there is consensus on implementation of PC2".

Something that I felt was important to note is that some comments in this set of RfCs were of a tone of "I don't support PC2 implementation, but if it is going to be implemented despite my concerns, this X criteria is something I would like to see implemented."

And also, while proposals 1 and 2 had consensus, it was noted in the previous RfC (paraphrasing:"Usage will be pretty much like PC1"; "This is to prevent the need for "full" protection in certain circumstances, such as socking") - and the commenters still dismissed such criteria as a "blank cheque" at that time. So, while consensus can change, I don't think it has in this case as yet. 7 only deals with what happens after a page has PC2 applied, so while it is technically a criteria for usage, it isn't a criteria for initial placing.

14 was a work in progress, and was consistently being tweaked. It's possible with more tweaking/discussion/tweaking/discussion/etc. this might have had consensus. But it wasn't there yet.

As a suggestion concerning one way forward, one thing that wasn't directly a proposal, but seemed to be alluded to by some commenters is that a possible thing to also discuss might be criteria for usage depending on namespace. Article space seeming to be the leading concern about usability over security. (And this is something which goes back to the initial discussions of PC/PC1 initially - not to have certain article space edits go "live" due to BLP concerns or other circumstances.) So this may be an avenue worth exploring through discussion. And those advocating a "test period" might consider proposing such a test in a namespace outside articlespace as a way forward. But this is merely a suggestion.

I haven't looked to see who, or the circumstances, but I saw that DP mentioned that someone arbitrarily closed the page to further discussion at some point.

While I suppose that process calls for an RfC to be closed after 30 days, somehow I don't think that that is meant to mean "let's close this to discussion, then go find someone to close this". And I think that this is supported by current policies/guidelines: "Not every discussion needs to be closed". And not every discussion must be closed by an arbitrary deadline. This is true in RfA, in deletion discussions, and is true in RfCs too. Discussion can be left open until such time as the actual closers decide that the request for comment should be closed to further discussion. Anything else leads to questions of "the wrong version" and of "stifling discussion."

As I said, I didn't investigate this, and I have no problem presuming that it was done in good faith, this is merely a suggestion for the future that perhaps arbitrarily disallowing further discussion might not have been the best choice in this case.

And so after all of the above, to try to be clear:

'''Overall, there is no consensus as yet to implement PC2. That said, there is consensus for proposals 1 and 2 and 7 to be used as criteria, but only if PC2 is implemented in the future. This close does not preclude (does not prevent) an immediate follow up RfC to continue work towards working out consensus toward the implementation of PC2, nor does it preclude (it does not prevent) further discussion concerning any and all criteria for usage.'''

So (while noting the specific examples, and that there are exceptions, such as arbcom), if PC2 is implemented in the future there is consensus for at least the following as usage criteria (the text of 1, 2, and 7):

"PC2 should be usable in the same situations as PC1, but only if the situations are being caused by autoconfirmed users and blocking them is an ineffective solution. Therefore, PC2 should be an option for pages that qualify for and might otherwise be fully protected instead. And the use of PC2 for longer than 1 year on any page must be reviewed by an administrator and/or by a community discussion at least once a year."

If you didn't read everything above, and read the emboldened text at the bottom and said to yourself "Wait, what? So in the last one jc37 said there was no consensus until there was criteria, and now jc37's saying there's criteria but still no consensus for implementation?", then I entreat you to please read the above. If after that you would like to ask for clarification, please feel free.

To the other closers - As I said above, I think that there is a fine line between consensus and no consensus here. And I am open to arguments that there is consensus here. I'm just not seeing such at this time.

Added due to current events: To all of you commenting on the talk page: As someone who has closed more than a few discussions, I think it's fair to note that it looks like at least a few of you are sharpening pitchforks and lighting torches waiting for Dank's actual close, based upon your read of his stream of consciousness notes and comments so far. Before starting the bar-be-que, I ask you to please consider that he stepped up to volunteer to help you all by offering to close this : ) - and also that I think he is trying to help find consensus before closing this discussion as a way to try to save the community time and effort in that (as I note) this appears to be rather close to consensus, even though it (in my estimation at least) doesn't appear to be quite there yet. No, that is not the typical thing for a closer to do, but I think his heart's in the right place, so please don't seek out said heart with your pitchforks and torches : )

And Dank, I welcome your thoughts on the above. In particular, concerning 13 and 15. - jc37 18:21, 3 June 2014 (UTC)

Closure text
While we can point back here to our drafts, I think we should have a more concise, unified closure text for pasting to the top of the RfC.

To start with something you and I appear to disagree upon: you don't feel that the community is allowed to immediately start on discussing the next steps? I'm not sure I understand that.

That is the substance of what I wrote above concerning a "no consensus" closure. If this was opposed, I might agree, but this wasn't. Indeed, as we both appear to agree, this discussion very nearly had consensus, I don't think we should arbitrarily prevent discussion. And besides, all we can do as closers is suggest editors wait x length of time. After all, as closers we merely are to reflect the community's wishes, not to impose our will upon the community : ) - jc37 18:41, 4 June 2014 (UTC)
 * My concern is that someone might file an RfC tomorrow, and when asked why, they'll point to your language. Can we just leave that part out completely? - Dank (push to talk) 18:47, 4 June 2014 (UTC)
 * Oh I see. It's intended to allow for an RfC immediately, to discuss these things, not to have a binding RfC to (or to not) implement PC2.
 * But these are all RfCs, it's up to the community to decide. And honestly, if someone started an RfC immediately to implement or to prevent implementation of PC2, I think that there would probably be [{WP:SNOW]] in that RfC's future as (if I were to prognosticate) most editors would be saying WP:NOTNOW (yes I know that's an RFA link, but I think you get the idea.)
 * These editors waited (along with you and me) many months already, I simply do not feel comfortable looking at them saying: "Hey, you all did a great job, you're so close to consensus, but hey, you need to wait several more months before you're allowed to continue that work".
 * That seems anti-wiki, and several other things.
 * I'm not saying they "must" start it, I'm merely saying this close (as a "no consensus" close), does not prevent it. - jc37 19:16, 4 June 2014 (UTC)
 * Okay, it's not a big deal, I'll delete my objection. - Dank (push to talk) 19:29, 4 June 2014 (UTC)

Okay, I'm happy with my text, and I've changed my "draft" to my "full closing statement". (One thing I just caught: in your statement, could you change "therefore" to "also"? The voters didn't say that #1 implied #2; just the opposite. And since you like "prevent", let's go with that rather than "preclude"). For the "more concise, unified closure text for pasting to the top of the RfC", do the 3 paragraphs that I agreed with "word-for-word" work for you, with pointers to our full closing statements here? - Dank (push to talk) 15:37, 5 June 2014 (UTC)