Wikipedia talk:Requests for adminship/2013 RfC/2

Discussion
At some point, we'll need to move things from the main page to keep it manageable. After this has run for one week, would it make sense to move proposals that seem to have no chance to this talk page? - Dank (push to talk) 20:02, 7 February 2013 (UTC)
 * I would suggest hatting discussions where we have clear consensus in either direction, there has been adequate time to consider things and discussion has ground to a halt. Unless it is very clear that will take at least 7 days, sometimes longer.  Ϣere Spiel  Chequers  07:22, 11 February 2013 (UTC)
 * For the traffic we've got now, that works for me. - Dank (push to talk) 16:24, 11 February 2013 (UTC)
 * In around 10 hours, at the one-week point, I'll do some collapsing; unless something changes, only Tiering and Autonom look collapsible to me. - Dank (push to talk) 15:54, 12 February 2013 (UTC)
 * It might be worth summarising proposals so far on the main RfA talk page (and possibly other venues) -- the numbers participating here are rather low. Is the Signpost covering this? Espresso Addict (talk) 01:20, 13 February 2013 (UTC)
 * Yep, we're getting a mention or two every week. - Dank (push to talk) 04:20, 13 February 2013 (UTC)

This week, there was just one to collapse: "mass re-boot", at 1 and 12. - Dank (push to talk) 13:17, 19 February 2013 (UTC)

hatting
why not leave proposals unhatted for the duration of the rfc? i see the 'unbundling' proposal had minimal support when hatted, but this tactic doesn't allow for the possibility that support could grow as word of the rfc gets out, now that this rfc is included in the 'in brief' section of one of the signpost articles. this defeats the purpose of a request for comments. it almost seems like no real reform is desired. 174.141.213.61 (talk) 12:35, 13 February 2013 (UTC)
 * If I'm trying to make sure there's no real reform, then I'm very bad at that. You do have a good argument: in general, closers are expected to do absolutely nothing until a discussion is over. This RfC specifically asks the closers to be more active ... I'm looking forward to the next one when we have nothing to do :) Do you want me to self-revert on Tiering (which has 2 supports and 15 opposes) or Autonom (1 support and 8 opposes)? - Dank (push to talk) 13:33, 13 February 2013 (UTC)
 * I for one have no objection to batting the auto nom proposal, it served its purpose as being the boldest thing I could think off, and I regard the auto prospecting proposal as a good way to make it more realistic. Martijn Hoekstra (talk) 15:15, 16 February 2013 (UTC)

MediaWiki Watchlist message
To increase awareness should a Watchlist message be considered? Leaky Caldron  13:48, 13 February 2013 (UTC)
 * Eventually, we want to advertise everywhere we can, but timing is important; this RfC seems to be working, and a sudden influx might (or might not) interrupt the flow. Seraphimblade is basically in favor; personally, I'm not in a hurry. - Dank (push to talk) 15:03, 14 February 2013 (UTC)
 * This is the final staging point for new ideas from the wider community who may not have followed this RfC since its inception on December. I'm not impressed by the number of contributors or the suggestions, which do little to really resolve the perceived issues that gave rise to the RfC in the first place. The more the merrier, I would say. Leaky  Caldron  15:09, 14 February 2013 (UTC)
 * Anyone have an objection to putting in a request to add it to the watchlist at the halfway point? That's in four days. - Dank (push to talk) 13:22, 16 February 2013 (UTC)
 * Request filed at MediaWiki_talk:Watchlist-details. - Dank (push to talk) 13:12, 20 February 2013 (UTC)
 * Not shown up yet. Leaky  Caldron  20:45, 26 February 2013 (UTC)
 * I've left a message at the link above. — Mr. Stradivarius  ♪ talk ♪ 13:55, 2 March 2013 (UTC)

templates
It would be good to have a template similar to the one that publicises RFA - listing the options and levels of support in them.  Ϣere Spiel  Chequers  21:52, 21 February 2013 (UTC)
 * By levels of support, do you mean someone would be updating the template every time someone votes? If so, are there any volunteers? - Dank (push to talk) 22:03, 21 February 2013 (UTC)
 * Yes that's what I mean. If people like the idea we could post a bot request or approach whoever wrote the current RFA one.  Ϣere Spiel  Chequers  22:37, 21 February 2013 (UTC)

Possibly sooner
Note that this round closes in 30 days or "possibly sooner". Any thoughts on the "sooner"? - Dank (push to talk) 12:46, 25 February 2013 (UTC)
 * I'm surprised how little activity we are getting at present - I thought that my proposal to unbundle deletion blocking would get a few more comments. I'd be inclined to let it run for thirty days, but maybe we should seek more input - WT:AIV might be a good place to mention one of these proposals.  Ϣere Spiel  Chequers  15:19, 25 February 2013 (UTC)
 * You're talking about WP:Requests_for_adminship/2013_RfC/2, right? - Dank (push to talk) 15:46, 25 February 2013 (UTC)
 * No I see that one as more relevant to the speedy deletion crowd. I was actually thinking of Requests_for_adminship/2013_RfC/2, but then I should have said blocking not deletion:)  Ϣere Spiel  Chequers  15:57, 25 February 2013 (UTC)

Comment
"The closers will share their thoughts on what's coming in Round Three in the middle of this round" – is something being worked on? -- Ypnypn (talk) 20:29, 26 February 2013 (UTC)
 * We should have something for you before long. - Dank (push to talk) 20:35, 26 February 2013 (UTC)

The current round of voting, Round Two, will end a few days early, this Saturday. After that, everyone is invited to attempt to determine (here on the talk page) what compromise proposal will get the broadest support, based on the results of Round Two. Round Three will be a one-week up-or-down vote on that compromise. - Dank (push to talk) 03:54, 28 February 2013 (UTC)
 * What, only one proposal? Out of all the various areas of discussion? How is that going to work? Surely we need several distinct proposals, based on all the different ideas that have some kind of decent support level. — This, that and the other (talk) 00:10, 1 March 2013 (UTC)
 * The main page for this RfC, WP:RFA2013, says that that goal is to find a compromise, and then have everyone vote on it ... but that compromise might be to vote on more than one proposal :) It's up to you guys, we'll talk on Saturday. - Dank (push to talk) 01:06, 1 March 2013 (UTC)
 * There are several unconnected proposals that seem to be getting support and would help a bit, including at least a couple of my suggestions. However the only changes that would actually solve the problem are not heading for consensus. I would suggest that a better route forward would be to start spinning the popular ideas out as separate RFCs or even just implementing them if we have consensus. That won't solve the problem, but if you reduce a problem enough times eventually you may get to an acceptable situation. If we aim to present one proposal a month that we believe will get consensus then we could keep the debate going as to the next couple of reforms, someone might even come up with an idea that would get a big enough change to fix the problem and in a way that the community could accept. But IMHO if we tried to package up several changes in one big proposal we'd risk it failing because everyone would object to at least one element of the package even if they supported others.  Ϣere Spiel  Chequers  14:09, 1 March 2013 (UTC)
 * Exactly right. A lot of the more successful proposals here only have about 60 percent support. The closers will need to weigh up the arguments on both sides as well as just counting the votes. Votes like "yes, absolutely!!!!!" and "no, never ever!!!!" are not as useful as those with proper explanations. — This, that and the other (talk) 09:05, 2 March 2013 (UTC)

End of Round Two
Round Two is over. Now we're in the phase where you guys try to figure out on this talk page, considering the vote totals and especially the rationales from Phase Two, what approach or combination of proposals is likely to get the broadest support in an up-or-down vote in Round Three. You're proposing some new processes that will only work if people volunteer their time to make them work, so it's important to get as many people on board as possible. As long as you guys are making progress towards consensus, we (Ed, Todd, and me) won't interfere ... if you get stuck, we may be able to help you get unstuck. Feel free to ask questions, and we'll get back to you as soon as we have a chance to discuss the results so far among ourselves. - Dank (push to talk) 01:55, 3 March 2013 (UTC)
 * To start with, here is a list of proposals that have at least as much support as opposition:
 * Probationary period for (at least some) new admins (16 support, 9 oppose, 2 neutral)
 * This would require a policy change and would need to be well advertised. The proposal would need to be particularly well-thought out for this one.
 * Recall, the 20,000' view (16 support, 8 oppose)
 * This would require a policy change and would need to be well advertised. Also, the proposal would require a lot of careful thought and structuring, possibly presenting multiple options to the community.
 * Unbundling - some U1 and G7s (12 support, no oppose)
 * I don't think this needs to move to Round 3 - someone could just write an adminbot (or write a MediaWiki patch) and point to the existing Round 2 consensus in their BRFA/Bugzilla bug.
 * Unbundling - limited block/unblock (7 support, 2 oppose)
 * Agree a criteria for adminship (13 support, 13 oppose)
 * "Not unless" candidates (8 support, 4 oppose)
 * Concerned editors start searching for quality candidates (5 support, no oppose)
 * This is just a simple suggestion and I can't see that we need to do anything with this in Round 3.
 * Auto-prospecting (24 support, 3 oppose)
 * Project for nominators (13 support, no oppose)
 * This can just happen; see support #10 and #12.
 * Obviously the RFC was not a vote, so this list is just for informational purposes. — This, that and the other (talk) 02:15, 3 March 2013 (UTC)
 * It seems that only unbundling (U1/G7 & block/unblock), searching for quality candidates, auto-prospecting, and project for nominators reached consensus to move forward. OhanaUnitedTalk page 03:03, 7 March 2013 (UTC)
 * Depends how high you set the bar for consensus. — This, that and the other (talk) 00:12, 8 March 2013 (UTC)
 * Low end of consensus threshold (70%). OhanaUnitedTalk page 00:24, 9 March 2013 (UTC)
 * There is nothing to do about searching for quality candidates proposal, everyone can already do this. Similarly project for nominators could be started by anyone at any time.--Staberinde (talk) 13:04, 9 March 2013 (UTC)
 * Then the only thing that needs to move forward to 3rd round (if 3rd round exists) is the unbundling of U1/G7 & block/unblock. OhanaUnitedTalk page 20:28, 11 March 2013 (UTC)
 * I'll be adding a section called soon. - Dank (push to talk) 21:32, 11 March 2013 (UTC)

Recall
I think that some recall process not involving ArbCom is good in theory. However, in practice, there are 2 major problems: The point is, we need some good way to handle this, or else not go there. עוד מישהו Od Mishehu 20:23, 4 March 2013 (UTC)
 * Regarding the recall stuff, Requests for Comment/Community de-adminship proof of concept/Navbox has links to related discussions if anybody wants to see what has been flatly rejected and what had broad support. That may help narrow the discussions so people don't end up wasting time on stuff the community won't pass. The summary in the close of Requests for Comment/Community de-adminship proof of concept has the most recent views of the community and may be particularly helpful if this idea is going to be presented to the community. Cheers. 64.40.54.87 (talk) 07:22, 3 March 2013 (UTC)
 * There are two concrete proposals sitting around in projectspace: Dennis Brown's WP:RAS and jc37's WP:RRA. They are actually quite similar in many ways. Personally I think the whole idea of "sanctioning" just confuses matters (I think the two proposals were taking a leaf out of ArbCom's book), and would be pleased with a simple "keep/desysop" vote. And the process of closing the debate (by bureaucrats or ArbCom?, all the weird time limits and different "verdicts", etc) is overcomplicated. But that's what we now have to nut out. Starting afresh is looking promising to me on the admin recall issue. — This, that and the other (talk)  08:39, 3 March 2013 (UTC)
 * This is one of the areas which the existing desysopping route via Arbcom has got right. There are some cases where the admin deserves some sort of trout but not a desysop, if you constrain recall to either keep or desysop then you risk promoting the idea that any slip up merits a desysop. Aside from the awkwardness that such incidents will create, it is also a bit silly to be discussing a proposal that will deter more good candidates from running as part of a response to RFA being broken and not producing enough admins - surely it would be better to discuss this as part of some proposal to reform Arbcom?  Ϣere  Spiel  Chequers  09:18, 3 March 2013 (UTC)
 * Second-guessing the community like that is not very helpful. Although perhaps you are right that tightening up admin recall doesn't belong in a RFA RFC! In any case, I just wrote User:This, that and the other/community admin recall with my thoughts on the issue. — This, that and the other (talk)  09:34, 3 March 2013 (UTC)
 * I have trouble understanding how existence of a recall procedure is going to deter more good candidates from running RfA? Only candidates who should be discouraged by this are those who request adminship mainly with intention to jump into major controversies right away.--Staberinde (talk) 12:55, 3 March 2013 (UTC)
 * If the only times this procedure is used are ones where a reasonable person would agree that the admin involved had done something seriously wrong then this wouldn't be too damaging to RFA. However those cases are already handled by Arbcom so having a third way to get rid of admins is unneeded. All the admin recall proposals I've seen so far have involved taking out some of the safeguards involved in the current Arbcom process and thereby making things easier for various coordinated groups of off wiki trolls to harass Wikipedia admins. The more we aid the trolls to make life uncomfortable for admins the more difficult it will be to persuade good candidates to run at RFA. Personally I'm probably fairly safe, my main admin activities involve deleting attack pages, user-requested deletions, blocking spammers, a few user rights and the some other clearcut speedy deletions. But any admin who gets involved in the more contentious stuff is going to annoy some people, and without a filter like Arbcom to filter out the spurious cases this site can easily degenerate into a kangaroo court.  Ϣere Spiel  Chequers  22:34, 3 March 2013 (UTC)
 * 1) How do we prevent the "enemies" an admin has created through proper admin activity (a user the admin had blocked correctly, an author of a page the admin had deleted...) from "ganging up" on him/her and having him/her desysoped without good reason?
 * 2) How do we prevent a situation where many adins will be afraid to handle controvertial cases, for fear of being desysoped?
 * Obvious solution for personal enemies issue is getting widespread community involvement, so that relatively small group of old enemies cant force their will. Basically some sort of reconfirmation RfA, which would be located at same place as normal RfAs, but could be somewhat modified (like lower passing requirements).--Staberinde (talk) 21:20, 4 March 2013 (UTC)

Unbundling U1 and G7
We got a response from a developer that this was practical - I'll get back to him and raise a request for a mediawiki change on bugzilla.  Ϣere Spiel  Chequers  09:18, 3 March 2013 (UTC)
 * My personal preference would be that participants here hold off on bugzilla requests until we know what the final request will be; it could easily be that, as a result of negotiations to get everyone on board, unbundling winds up being more elaborate than just U1 and G7, and devs tend to get testy when you ask them to do something one way, then ask them to do it a different way two weeks later. However, I haven't checked with Ed and Todd on this (I felt this needed an immediate response), and I can't stop anyone from filing a bugzilla request, nor can I stop anyone from reading the results of Round Two and coming to reasonable conclusions on what we already have consensus on. - Dank (push to talk) 12:06, 3 March 2013 (UTC)
 * Specifically, the vote was 7 to 2 (that doesn't mean this is just a vote, but it's suggestive) in favor of "Unbundling - limited block/unblock", and that's going to require discussion ... and then people may decide they want to tie those two bundles together, or tie them to other things. If we give devs multiple requests from multiple people over several weeks, they're less likely to cooperate. I'll watch this discussion with interest, and I've suggested to Ed and Todd that, depending on which way this goes, we may want to be more hands-on with this discussion than with the others. - Dank (push to talk) 12:59, 3 March 2013 (UTC)
 * These would be radically different usergroups. Every registered editor would get a very limited deletion right re their own mistakes, and a small number of very experienced non-admins would get a limited block unblock right. I think that these are sufficiently different that they should be treated differently at Bugzilla, as for tieing either of these to other things, yes there may be proposals to expand the limited block into some sort of admin lite, but realistically if that happens we will just wind up with another unsuccessful admin lite proposal. I would concede though that unbundling limited block/unblock could do with greater discussion as the number of participants has not been great and there is some disagreement as to whether these editors should be allowed to deal with editwarring etc or only vandalism and spam.  Ϣere Spiel  Chequers  22:12, 3 March 2013 (UTC)
 * Certainly valid points. My only request here is to avoid annoying the devs ... but perhaps the best way to do that is to go ahead and get this one thing off the table so they don't get multiple requests. WSC, could you get a dev to weigh in here saying that they believe this is doable and they don't mind doing this bit before the other bits? Are there any objections (Ed? Todd? Anyone?) to WSC's proposal that he work with the devs to get this done? The vote was 13 to zip on this one, with nothing unresolved in the discussion section. - Dank (push to talk) 22:43, 3 March 2013 (UTC)
 * I spoke to one of my contacts there and she sent a techie over on the 22nd, he said that most was doable. Whether they will do it is another matter of course..... But I doubt they would see this as part of some sort of RFA reform - I'll report back to them that we have consensus and take it to Bugzilla.  Ϣere  Spiel  Chequers  01:48, 5 March 2013 (UTC)
 * Raised at Bugzilla  Ϣere Spiel  Chequers  13:45, 8 March 2013 (UTC)
 * Two of the closers of this RfC are on board and one is busy. The vote was 13 to 0 in favor. The only question remaining is whether the page could be deleted by the creator if a page with the same title was ever previously restored (un-deleted) ... this point was mentioned by Thryduulf in Round Two. Two things to consider are wheel warring (by non-admins!), and the likelihood that proper attribution to the original editor will be lost if people get in the habit of adding back previously deleted text themselves, just to avoid deletion by the original creator. - Dank (push to talk) 16:04, 8 March 2013 (UTC)

Round Three
At the moment, I'm speaking just for myself, Ed and Seraphimblade are slammed with work and are still thinking this over. My position is, before we can give people something to vote on in Round Three, we have to resolve the differences between Round One, in which there was a super-consensus in favor of making RfA more even-handed and easier to anticipate and deal with, and Round Two, which doesn't explain clearly how we're going to get there. The brightest spot I guess is Unbundling - limited block/unblock (7 support, 2 oppose), but the 2 opposers make a good point: without sufficient vetting, these "limited blockers" (can someone think up a name for this role, please?) will make enough mistakes to generate pushback against this proposal, which would be a shame, because proper use of the blocking tool would get the job done and would also help people pass RfA later on. (People don't need the deletion tool for us to know how they'll do with deletions, of course, all they need to do is tag for deletion and vote on deletions, but use of this blocking tool will give us new and potentially valuable information at RfA.) Is there any way to make the vetting process for this (at WP:RFPERM, I assume) lightweight when it needs to be lightweight and more in-depth when there are serious questions to be answered? If this proposal passes, I'm in favor of another RfC in three months (if there are big problems) or six months (if not) to examine whether this proposal is working as intended.

Since Concerned editors start searching for quality candidates, Auto-prospecting and Project for nominators all got solid support, we're probably going to have a bunch of new candidates at some point. But I can't help thinking of the time in college when a friend asked me if I thought he should ask some girl out, and I said "Go for it!" ... I thought he was just looking for encouragement, but it turned out he thought I had some insight into what she was going to say. She turned him down, and he never forgave me. I'm sure a bunch of new candidates who hadn't previously been thinking about RfA, or thought about it and rejected it, will give us some new, high-quality admins ... but others are going to fail, and I'm not going to be surprised at all if they wind up blaming us for setting them up for a fall, if they don't know what to expect, and most don't. Of the proposals that were intended to raise the passing rates at RfA, the one with the highest percentage supporting was "Not unless" candidates (8 support, 4 oppose) ... and now that it looks like we're going to have "limited blockers", that would make it easier for candidates to demonstrate what they know while the crats are trying to figure out what to do with their RfA (but it would only demonstrate their skills and knowledge, not their intentions ... anyone can "play nice" when they're being watched ... so we'd have to rely on the RfA to tell us whether the community trusts the candidate or not).

Two of the opposers said: we don't want crats making this call. Some things RfA voters do feel strongly about (trustworthiness, clue); for other questions, RfA voters already tend to rely on other RfA voters to do the research (a full analysis of deletion work, for instance). It's not inconceivable that, if a candidate is seen as trustworthy and hard-working but perhaps not as experienced as they need to be, some RfA voters wouldn't mind letting crats make the call on that at a later time. The other two objections were: we don't need this, because the candidates can always run a second time. Well, they can, but they usually don't ... that's most of the problem, the number of well-prepared candidates has been dropping. Most candidates who get 60% supporting and 40% opposed tend to believe the 60% and to have a low opinion of the 40% (see confirmation bias), and that low opinion leads them to believe that there's not a lot of point in working hard for another year just to get shot down again by a bunch of dimwits. (I want to be clear that I don't think RfA voters are dimwits, I think we do pretty well, all things considered, but I agree with the super-consensus in Round One that the throughput at RfA is suffering and needs to be fixed.) So ... I want to be very careful not to bias the voting here (or at least, not to get blamed for biasing the voting :), but we did have 8 supporters here, and it's conceivable to me that we could find a way to satisfy the opposers, if we're talking about candidates who were likely to get promoted some day anyway, and if we're limiting the role of the crats, and if we make sure the candidates at least have the limited-blocker tool to use while the crats are doing their thing. Well, I've said enough (or maybe too much) already, what do you guys think? I'll point the four opposers here in case they're not watching. - Dank (push to talk) 19:53, 13 March 2013 (UTC)
 * The abject failure to broadcast stage 2 will make the decisions reached unacceptable to the wider community. 9 comments (7/2) is not enough to stick. I oppose it so make that 7/3. Little over 50%. Waste of 3 months IMO, no real reform ideas and one that will not be accepted once the community at large sees it. Leaky  Caldron  20:00, 13 March 2013 (UTC)
 * The Signpost (every week), WP:CENT, RfC, WT:RfA ... where else could we have posted it? There wasn't even one comment in support over at the watchlist page. I share your nervousness, Leaky ... we didn't get enough participation to be able to predict what's going to happen as we move forward. But I have to listen to what people said and try to find consensus, and then let people react however they're going to react ... what else can I do? - Dank (push to talk) 20:11, 13 March 2013 (UTC)
 * I dont think RfA can be made easier to pass until adminship becomes real "no big deal", but this RfC obviously didn't achieve any breakthrough in that direction. Unbundling of the blocking tool can be proposed but I am sure it will be shot down in larger discussion, although my view of its chances to pass may be biased due my personal opposition. Only real conclusion from RfC seems to be that more effort should be put into searching good candidates that would pass RfA as it works currently.--Staberinde (talk) 14:22, 16 March 2013 (UTC)
 * With regard to "Not Unless", as one of the oppose, I would definitely consider supporting this, after the comments by Joe Decker and Dank. I believe when I commented by position was coming from trying to simply the process and not complicating the process further. My thoughts is that if we agreed on a criteria for adminship it would make it more likely that people would pass the first time, and thus, it would be unnecessary to have people second guess themselves, and removes the seemingly arbitrary votes "dimwits" from being a major concern. Although without clear RFA criteria that was actually weighted heavily in the RFA process, then I would agree that the "Not Unless" concept has merit. Tiggerjay (talk) 01:52, 14 March 2013 (UTC)

Update: we're still working on the closing statement, more soon. - Dank (push to talk) 04:17, 18 March 2013 (UTC)

Apologies for the delay in closing Round Two; sometimes it takes as long as it takes. User:Seraphimblade is on a wikibreak ... which is fine, because Ed and I (Dan) agree on a broad range of points, including everything in this section. The main thing we want to say is: we're not telling anyone "tough luck, maybe next time". We think this RfC was very successful and it will do some real good, if people will let it ... but it doesn't deal with every problem and answer every question, and there's no way it could have ... we've gone 10 years without a successful RfC on RfA, and not everything is going to get "fixed" to everyone's satisfaction overnight. Ed and I respect all of the points of view that have any significant representation in this RfC. There was a two-thirds vote in favor of something called "recall", and the arguments there were quite reasonable ... the problem is, neither the proposer nor the voters told us what "recall" is, so we don't think there's anything we can implement at this time based on this vote. I'll be happy to talk with people about setting up a new RfC on that not too long after this one closes.

The strongest support was for the various proposals to go "prospecting" for more suitable RfA candidates ... I gave the three links above. Ed's on board with what I said above about the possibility that this will create problems ... if you do outreach to people who haven't been thinking about RfA, it's really going to be counterproductive if a lot of them fail because they haven't been thinking about RfA. Fortunately, the "Not Unless" proposal got an 8 to 4 vote, and as I mentioned above, support seems to be even stronger for that now. But there was also a substantial vote for "Probationary Period" (although the opposition was substantial as well), and many voters liked the Probationary Period because it would give the provisional candidates some tools they could actually use, so they could do more useful work and the crats would have better data to evaluate them with. So our question for the supporters and opposers of Probationary Period is: can you find a compromise on where to draw the line on userrights for Not Unless candidates? That is, imagine a "prospect" who gets 65% in their RfA ... but the crat (and everyone else) agrees that the only reason the candidate failed was lack of experience with admin-related things ... otherwise, they're clueful and hard-working and clearly trusted by the community. Ed and I think (and hopefully crats and you guys think as well) that we've got consensus as of this RfC to make some of these guys "Not Unless" candidates, that is, the crat can put off closing their RfC so that they have time to work hard and demonstrate knowledge and skills (btw, you guys never said ... how long do they have before the crat has to close it one way or the other?) To anyone who has strong feelings in either direction: what userrights do you think a lot of these guys will need to put them in a better position to get promoted to adminship when the crat closes their RfC? What userrights should they absolutely not have until and unless the crat promotes them? - Dank (push to talk) 15:52, 22 March 2013 (UTC)
 * "None beyond rollback" is an applicable answer, by the way. ;-) More seriously, your only limitation for suggestions here include anything deletion-related that isn't deletion itself (i.e. undeletion, view deleted revisions) due to WMF legal limitations. Ed [talk] [majestic titan] 16:04, 22 March 2013 (UTC)
 * Right ... that would be fine with me. What I'm saying here is that, judging from the votes, there was substantial support (16 to 9) for the idea of giving the whole package to the provisional candidates ... and I'd like people to talk about this a little more and see if some accommodation can be reached before we give up. I'll go invite people who participated in the relevant discussions to come have a look. - Dank (push to talk) 16:41, 22 March 2013 (UTC)


 * Simple. No one should have any Admin userright until they have passed an RfA. "Not Unless" means not now AFAIAC and, I suspect, the majority of the community once this is given wider communication. This was a paltry 8/4 among a very small group of interested editors. Too complex to implement and manage, candidates should simply be told to come back for RfA when they can demonstrate improvement in the areas of concern. Anything else is just RfA scope WP:CREEP. Also, I cannot see 'Crats wanting anything to do with this since it involves tenuous judgement issues and they prefer to deal with solid facts. Leaky  Caldron  16:16, 22 March 2013 (UTC)
 * I half support and half reject this, Leaky. The way the RfC system works is: you invite people to come, and the ones who have the spare time to think through all the issues and are brave enough to commit themselves to a position (on a subject that's generated a lot of heat and push-back in the past ... not so much of that in this one) get to make the rules. OTOH, I'm not going to try to lay down any law that you can't launch another RfC, any time you like, on any subject. Be warned that it's harder than it looks, and the really hard part is focusing on listening ... but if everyone's on your side except for a "paltry" few, as you believe, it shouldn't be difficult for you to establish a different consensus, and I'm not going to discourage you from doing that or interfere ... in fact, I'll thank you for putting in the effort. - Dank (push to talk) 16:37, 22 March 2013 (UTC)


 * So something based on this will be implemented without further widespread notice? Leaky  Caldron  16:49, 22 March 2013 (UTC)
 * We'll follow the plan we've had since the beginning. We gave notice through the RfC system, in the Signpost (many weeks running), and at WP:CENT, WT:RFA, WP:AN and Jimbo's talk page. We'll do the same before Round Three, which will be an up-or-down vote on Not Unless (as it's interpreted in discussions here ... Limited Blocking, the U1/G7 proposal, and the "prospecting" proposals already have clear support, there's not a lot to discuss there). - Dank (push to talk) 17:01, 22 March 2013 (UTC)
 * So hows about a neutral message from the closers at WT:RFA? That should stir up interest! Leaky  Caldron  17:03, 22 March 2013 (UTC)
 * Yep, WT:RFA too, as soon as we've got a little more feedback from the voters. Regarding "neutral": neutrality required that the closers not bias the questions or the results, and we didn't. Now that the voters have supported Not Unless in Round Two, neutrality requires that Ed and I be honest about, and supportive of, the voters and their rationales. - Dank (push to talk) 17:31, 22 March 2013 (UTC)
 * Why not ref to WT:RFA now, in order to obtain more varied feedback? Leaky  Caldron  17:39, 22 March 2013 (UTC)
 * Because our first obligation to is to close Round Two, and Ed and I have a couple of questions about what the voters meant in Not Unless and Probationary Period. After they give (or don't give) us feedback, we'll close Round Two and then post notices for Round Three. - Dank (push to talk) 18:27, 22 March 2013 (UTC)

A question for people who commented at Probationary Period and Not Unless
If you voted or commented at "Probationary Period" or "Not Unless", Ed and I would like to know if there's been any meeting of the minds since the voting. The first option would have given the entire admin package to a provisional candidate in the period after the RfA and before the candidate achieves non-probational adminship, and the support was substantial, but there was much less opposition (and two of the opposers are no longer opposing) for the second option, which gives no userrights to the provisional candidate. Is it still an all-or-nothing question for you guys, or has a consensus developed that any of the admin userrights would be acceptable for the "Not Unless" candidate after their RfA? - Dank (push to talk) 18:41, 22 March 2013 (UTC)
 * I didn't/don't read the Probationary Period option as you have. My support was based on (all) candidates being passed by the 'Crat. and recalled if subsequently found to be unsuitable within a probationary period. The 'Crat. would still close as normal at the end of the RfA, not leave it open ended. Leaky  Caldron  18:58, 22 March 2013 (UTC)
 * My understanding on "probationary" was the same as Leaky Caldron's. --Orlady (talk) 19:13, 22 March 2013 (UTC)
 * The spin that the "Not Unless" supporters had is also present in many of the "Probationary Period" rationales ... but Ed and I couldn't tell for sure what people wanted. Thanks for letting us know. - Dank (push to talk) 19:18, 22 March 2013 (UTC) tweaked to "before the candidate achieves non-probational adminship" 19:28, 22 March 2013 (UTC)
 * Much like Leaky Caldron and Orlady I presumed, if a probationary system was introduced the candidate would pass as normal and be given the tools under a certain supervision (perhaps a Special:Probationary admin actions which can be viewed by admins not on probation. The key about being on probation was that the tools could be recalled easily by a crat during the probationary term rather than going through a lengthy ArbCom process Cabe  6403  (Talk•Sign) 19:34, 22 March 2013 (UTC)
 * Ayup. ^^^^^  --j⚛e deckertalk 19:35, 22 March 2013 (UTC)
 * You guys aren't answering the question I asked, but that's fine ... I'm happier to hear what you're thinking. If no one gives me an answer, I can assume the answer is "no". - Dank (push to talk) 19:47, 22 March 2013 (UTC)
 * I didn't !vote or comment about "Not Unless" because it didn't make a lot of sense to me. My support for "Probationary" was based on the notion that some "opposes" at RFAs are based on "what if" concerns that likely would become less emphatic if !voters thought it might be easier to take back the bit in the early weeks after it was granted. I guess my objection to "Not unless" was related to a perception that the reservations expressed about RFA candidates are seldom as specific as that proposal contemplates. --Orlady (talk) 20:08, 22 March 2013 (UTC)
 * I !voted in support of the Probationary Period. My intention, similar to others above, was that the RfA would finish as normal, and the new admin would have full powers and go off and start doing their job.  But if they screwed up badly in the first 3-6 months, their adminship could be revoked via a very fast/simple process.   After the 3/6 months elapse, they are vested in the job, and a full RfA de-sysop process would need to occur.   I did not !vote on the "Not Unless" proposal, and I have no opinion on it. --Noleander (talk) 20:48, 22 March 2013 (UTC)

Okay, I'm expanding the question based on the responses so far: has there been any progress on a meeting of the minds between the supporters and opposers of Probationary Period? Ed and I aren't comfortable ignoring the opposition there ... there's a lot of it, and it mirrors things that have been said many times before ... and meanwhile, opposition to Not Unless has dropped. But if you guys can work something out, this would be a good time to do that :) - Dank (push to talk) 20:21, 22 March 2013 (UTC)

As I read it&mdash;correct me if I'm wrong&mdash;the "not unless" idea puts an awful lot of discretionary power in the hands of bureaucrats, and it isn't at all what I had in mind when voting for the probationary period. I like the idea of a probationary period, especially if we designed it so that candidates with lower support margins could pass. I don't like the idea of investing a bureaucrat with the power to make the outcome conditional upon something that he or she feels is necessary; that would just shift the power in the whole process further toward bureaucrats. So these ideas look very different to me, and I don't think (unless my impression of "not unless" is mistaken) that my support for one could transfer to the other. Everyking (talk) 20:27, 22 March 2013 (UTC)


 * Okay, Dank, I'll try to answer, although I'm wasn't an opposer. I don't know if there's been any meeting of minds, I guess I'd just say that I'm open to a wide-range of solutions (probation, optional probation, not unless) that provide some middle-ground to an all-or-nothing RfA, and that I think that having some solution of that form is more important than the specific form of that solution. --j⚛e deckertalk 20:53, 22 March 2013 (UTC)

I remain opposed to both of these proposals, and I am also opposed to being asked this question--an RFA should be closed based on comments in the RFA. Otherwise those who originally commented in the RFA but are not available to answer your subsequent questions are shut out of the process. Chick Bowen 21:31, 22 March 2013 (UTC)
 * That's a fair criticism, Chick, and if we didn't need more input to increase the odds that Round Three will be successful (in some sense), we wouldn't have asked at all. - Dank (push to talk) 22:28, 22 March 2013 (UTC)


 * I'm responding here because of the message on my talk page, not that I have much to contribute. I commented in Round 2, have been kind of wondering about Round 3, and otherwise not given the subject any thought since Round 2. I still like the idea of a probationary period, and I'm not particularly comfortable with not-unless, but I've had zero discussions with anyone else about it. --Tryptofish (talk) 23:27, 22 March 2013 (UTC)
 * What I had in mind when supporting the probationary period was, say, a 3-month period of initial adminship (including full use of tools) after which a simple support/oppose "reconfirmation" vote occurs (as has recently occurred on Wikidata). However, it doesn't look like anyone else read it that way. — This, that and the other (talk) 23:45, 22 March 2013 (UTC)
 * These two are very different proposals and I don't see any logic in linking them. The probationary idea involves extra scrutiny for new admnins and a desysop for new admins who make mistakes. The Not unless proposal involves a quick route to adminship for some people who fail for reasons that can be resolved without the need for a fresh RFA. So one proposal involves all new admins the other a small minority of those who fail RFA. As for my views on the two proposals, I still think that the probationary idea is a blind alley - our problem admins are those who lose touch with the community and its standards, and we are more likely to have problems with admins who have been admins for more than three years than with our new admins. As for not unless, I'm aware of a few instances where it would have been useful. The classic example being Panyd's first RFA when her boyfriend and nominator accidentally edited from her PC and a lot of people worried that they were socking. In that case "Not unless the candidate and nominator identify to the Foundation as two different people" would have been a sensible close.  Ϣere Spiel  Chequers  09:10, 23 March 2013 (UTC)
 * Not sure if/how/when/why the 2 proposals have become conflated. Probation is simple. Every successful candidate using existing processes is considered temporary for a period (tbd) during which concerns about performance might result in desysoping (under a process to be defined). Not unless appears to provide some sort of flexibility in terms of what tools are given at the discretion of the 'Crat. It is too complex, problematic to implement and ultimately unworkable. Despite being identified for consideration as a result of the selection process used in this RfC, it will not, I think, be approved by the wider community. No need to mix and match elements of the 2 proposals. They were and must remain separate proposals. I have discussed this with no one. All my deliberations are on this page.  Leaky  Caldron  11:19, 23 March 2013 (UTC)
 * By not unless, I understood as a statement that a 'crat should give failing admins an idea of what they need to improve in order to be more likely to pass the next round - as I said eplicitly, "Probably few of the failed RFAs are truly hopeless users". By "probation", I meant that that all new adminship is conditional during the first period (half a year, I believe, was explicitly mentioned as a period), during which a community-driven request for desysoping can be done - this should help the community trust these new admins better, while the user would (reasonably) not gain many enemies (it takes time for them to resurface under IDs which won't show them as such) or handle controvertial areas (these should only be dealt with by experienced admins). עוד מישהו Od Mishehu 21:55, 23 March 2013 (UTC)

We'll probably post the close to the Round Two page today. Your feedback here was very helpful. In Round Three, we're reporting the "Not Unless" vote as what it was, 8 to 4, and later statements give us some doubt about the 4 ... because the Round Two close we should report only on the actual voting during Round Two, plus any reasonable interpretations of just those votes that we get from later discussion. (We want to give a consistent message here, applicable to Round Three and any future related RfCs ... if you have an opinion on something and you want your voice to count, you have to vote.) We're not discarding statements you guys made above about being dubious about this or that ... and you haven't missed any opportunities, you can vote in Round Three and put anything (within reason) in your rationales. - Dank (push to talk) 14:38, 25 March 2013 (UTC)

Kind of a shame
That so few were even aware of this and these decisions were made by so few in the community. A little disappointing. I suppose I have no one to blame but myself...well maybe not. LOL! (but I will stick to that)--Amadscientist (talk) 07:50, 31 March 2013 (UTC)
 * I'll say something in the closing statement about how we might get more people involved. - Dank (push to talk) 12:19, 31 March 2013 (UTC)

Project for RfA nominators
As this suggestion received almost unanimous support, I have gone ahead and created WikiProject Admin Nominators. Anyone with an interest in the nominations process is very welcome to join. Espresso Addict (talk) 22:08, 8 April 2013 (UTC)