Wikipedia talk:Community de-adminship/Draft RfC Archive 3

Other projects
I note that the |current proposal for Global Sysops has a section on reconfirmation which is similar to this proposal, and I recall that en-ws has similar. It would be useful if the final RFC noted down all teh other projects which had or didn't have something like this. Anyone fancy doing some research? AndrewRT(Talk) 01:07, 8 January 2010 (UTC)


 * Good point - definitely something to add to the FAQ. Ben   Mac  Dui  19:40, 8 January 2010 (UTC)

Cute. Now my disagreeing comments have been gutted from this page.
Somebody in their ardor to protect this ridiculous proposal has now gutted my prior comments from this page. Yet, this is the page where one is to come to discuss the proposal, according to Wikipedia talk:Community de-adminship. I guess opposition to this is no longer tolerated? Is that it? --Hammersoft (talk) 22:41, 9 January 2010 (UTC)


 * Hammersoft, you will notice that almost all the comments have been archived, not just yours. What remains is a summary (with some of the more detailed matters at the end still in place so that they could be evaluated for the next steps). You will be free to voice your opposition to the RfC when it is actually in process. I am sorry that this does not appear to be as clear as I would like it to be. Constructive comments as to how this can be made more obvious are welcome. Regards, Ben   Mac  Dui  18:33, 10 January 2010 (UTC)


 * Yes this is clear enough to follow, or as can be in the circumstances. Your premature "Hell no" motion had to be archived after the 4th Jan, and it had pretty much stabalised at 2:1 against curtailing/abandoning the CDA propoal process. All it did was waste even more time over the hardest period of the year to get anything done. If you want to 'vote-protest' further, do so at the final stages, in the eventual RFC poll. Matt Lewis (talk) 20:36, 10 January 2010 (UTC)


 * While I have stood aside of late to let other voices be heard, I am moved to comment. Hammersoft and those who wanted to shut this "ridiculous proposal" down before it could be polished - and widely reviewed in a finalized form by the Wikipedia community - have had their say.  Consensus was clearly against them, and we need to move forward now.  Agree with Ben MacDui that constructive comments are always welcome, as long as they are indeed constructive.  Jusdafax   21:41, 10 January 2010 (UTC)

Will this proposal include de-bureaucrating people?
One of the complaints in favor of this process is that admins are elected for life. Well, so are bureaucrats, with no process available to remove them. Shouldn't this proposal include a means of community de-bureaucrating if the bureaucrats are deemed to not be doing their job appropriately? --Hammersoft (talk) 15:38, 11 January 2010 (UTC)
 * You have made a point, but this de-admin proposal neither would nor could propose de-sysoping bureaucrats too, as you surely know. I do hope you won't keep creating these sections - you can comment in Late Comments above, but creating new sections for comments like this in a working environment will soon be disruptive if you carry on. You could have easily made this point at any time, and probably did I would guess. If you really want a separate 'CDB' - why don't you start something up?


 * This proposal is centred on bureaucrats making the final judgement in a Community de-adminship process, based on consensus (currently with 70-80% guide margins). What this type of CDA will do in my opinion is make bureaucrats more accountable, and closer to the editors. I believe the whole process will lead to greater accountability and better decision making across Wikipedia, and increased faith in both admin and bureaucrats amongst editors. Matt Lewis (talk) 16:03, 11 January 2010 (UTC)
 * Asking a question isn't disruptive. Assume some good faith here. It's an honest question. I am asking this question now because I thought of it now. I find demanding an editor not ask questions or make comments to be considerably disruptive. As to the point raised by me, I don't see a reason why this process could not be expanded to include bureaucrat removal. Bureaucrats have made severely controversial decisions before. Some have stepped down after making such decisions. Some have not. For the ones who have not, the community has no means to de-bureaucrat a no longer trusted bureaucrat. This process could readily be adapted to include the bureaucrat flag. --Hammersoft (talk) 16:11, 11 January 2010 (UTC)
 * I am in agreement with Matt Lewis' point re:bureaucrats making the final call, and also not being involved as eligible for a CDA (or actually, CDB) process. There are, for the foreseeable future, a great many more admins than bureaucrats, and this draft RfC is and has always been about administrators, and only administrators.  It is my view that trying to abruptly start adapting it in the final stages to bureaucrats would be both awkward and time consuming.  Let's stay focused on Community de-adminship, please. Jusdafax   16:26, 11 January 2010 (UTC)
 * e/c I mainly lack AGF due to your "I'm a certified idiot" etc user page - which is how you want to present yourself presumably. I have certainly not demanded that you do not ask questions! Firstly - this process is called CDA, not CDA/B. Secondly (and logically), when bureaucrats are judging cases it will be problematic if they are challenged via the CDA/B during this. If you want to change both the name and the contents, you may as well start up something new: namely a CDB. I wish you luck!
 * Also - if a bureaucrat abuses his/her admin duties, they can be subject to a CDA like anyone else. Bureaucrat duties are something else though. Matt Lewis (talk) 16:43, 11 January 2010 (UTC)
 * Adding "certified idiot" to my userpage doesn't grant all other editors the right to assume bad faith with regards to my work here. One of my biggest, if not biggest period, complaints about this project is the lackadaisical attitude with regards to personal attacks. I find calling myself a certified idiot helps to reduce the idiocy that would otherwise get tossed at me. --Hammersoft (talk) 18:08, 11 January 2010 (UTC)
 * And on that, I agree with Hammersoft. --Tryptofish (talk) 18:16, 11 January 2010 (UTC)
 * You can always add User:Hammersoft/CertifiedIdiot to your userpage :) Join the ultra-secret certified idiot cabal! :) --Hammersoft (talk) 18:21, 11 January 2010 (UTC)

Proposed simplification of page structure
I think that it is confusing to readers that we have Community de-adminship as the main proposal, but Guide to Community de-adminship as the page where the bulk of the material to be decided is located. I propose that we merge the two pages. Specifically, the list of other pages at the bottom of Community de-adminship could be incorporated into Community de-adminship/Draft RfC, and then all of Guide to Community de-adminship could be moved into Community de-adminship. Then, there would simply be Community de-adminship, with Community de-adminship/Example as a sort of appendix. OK? --Tryptofish (talk) 20:11, 11 January 2010 (UTC)


 * Good by me, and I commend you for thinking of that. Let's do all we can to prepare for the RfC. Thanks again! Jusdafax   20:26, 11 January 2010 (UTC)
 * Thanks. If other editors agree, I would suggest waiting to do it right before we make the final version, just to simplify the editing process going on now. --Tryptofish (talk) 20:40, 11 January 2010 (UTC)

Fine by me - I'm assuming the actual RfC is going to go on Wikipedia talk:Community de-adminship. Ben  Mac  Dui  17:30, 13 January 2010 (UTC)
 * Maybe I misunderstand, but wouldn't we have people come to an RfC on a page dedicated to the purpose, with a link to the page with the proposal? --Tryptofish (talk) 20:34, 13 January 2010 (UTC)
 * Yes, you are right. I have a cold and am not 100%. If you suggest a red link title I'll move the draft there ready for an off and fix the links in the FAQ and Navbox. Ben   Mac  Dui  16:49, 14 January 2010 (UTC)
 * No problem, and no hurry. I hope you feel better soon! --Tryptofish (talk) 20:06, 14 January 2010 (UTC)

I'm not sure it's possible to do serious page re-arringing now. I've put in a couple of drop-down archives (though I don't think they lower the KB of the page, but they will make it more readable), and have created an 'archive 2' for post Jan 4th stuff. I'd be obliged if someone could look at archiving into that. It's hard for any one editor to do too much, as I'm sure MacDui and others know. I haven't advertised the finalization poll yet, as I think the page is still too offputtingly long. The length of this page is making 10 minute jobs last up to an hour for me - it's excrutiating. My laptop memory just can't seem to handle what it could anymore, or maybe it's just Firefox losing pace again I don't know. Matt Lewis (talk) 14:12, 16 January 2010 (UTC)


 * Community de-adminship/RfC it is then. Ben   Mac  Dui  18:10, 16 January 2010 (UTC)
 * I don't understand that. As I've said before - please can we be clearer with each other. Ambiguity won't help any of us in any way.Matt Lewis (talk) 19:33, 16 January 2010 (UTC)


 * Sorry it is actually a reply to Tryptofish re the name of the actual page for the RfC and follows on from the discussion above your comment of 14:12, 16 January. (This earlier conversation is about re-ordering other pages rather than this one). Hope that's clear. Ben   Mac  Dui
 * I might appear to be the 'invisible man' of late, but remember I'm still here, as are my posts. That is, I'm here for the time being at least. I'm a serious Wikipedian, but I'm not so dedicated as to flog a dead horse. Matt Lewis (talk) 20:08, 16 January 2010 (UTC)

Question about "all editors"
I've made an edit to the Guide to reflect the discussion above, under. But in doing so, a question became apparent to me. In Uncle G's original version, there was to be only !voting without discussion, and all discussion was to be on the associated talk page. We have clearly agreed that there need to be comments on the !vote page, of course. But Uncle G's original language said that all editors, with or without accounts, are welcome to "discuss". At this point, that has come to mean !vote as well. So, I want to make sure: are we saying that (unlike RfA) IP editors are free to !vote in CDA? --Tryptofish (talk) 19:32, 12 January 2010 (UTC)

Based on the above question and this reply I made a simple amendment thus to the Guide. Ben  Mac  Dui  18:04, 16 January 2010 (UTC)


 * Good. Yes, that's also my reading of what was discussed, but I just didn't want to take that interpretation onto myself without having other eyes on it. Thanks. --Tryptofish (talk) 21:38, 16 January 2010 (UTC)

To Do List
A number of these are also covered at Wikipedia talk:Community de-adminship.

Publicity
Proposal 3: Community Support for changing:

"Nomination by the Community at large requires the signatures of no fewer than 10 editors in good standing (defined below), within a period not longer than 3 days. Signatures must be placed in the nomination area of the requests, as a simple signed bullet point."

to "not longer than 7 days".

Concrete percentages
Tricky one this, but the only proposal with genuine support was

5.4 Add to the current wording:


 * "Thus, for an Administrator to be de-sysopped, a bureaucrat will review the discussion to see whether both a minimum of 50 editors and a general consensus supports de-sysopping. Consensus is sometimes difficult to ascertain and is not a numerical measurement, but as a general descriptive rule of thumb, above ~80% support for de-sysopping would be acceptable; while support below ~70% would not be, and the area between is subject to bureaucratic discretion."


 * Comment: I think we are missing a very crucial point here. Admins have, by definition, a vested interest in making their own removal more difficult. At 80%; you need to muster a mere 9 other Admins to block any de-sysopping. Attila The Hun could manage that! I suggest we stop ignoring the critical factor of Admin solidarity and suggest 80% after excluding Admin votes. Sarah777 (talk) 19:09, 10 January 2010 (UTC)
 * I had similar thoughts&mdash;but not about administrators in particular, rather just about Wiki-friends of the admin under review, regardless of whether they were admins themselves&mdash;when I pushed for a lower percent to desysop. But the result of the discussion was pretty clear: I ended up as the only support for the lowest percentage. In my opinion, the version that we are now contemplating is the only one that has a reasonable chance of being accepted by the community. I'd rather get that now, and then, when there is a required review of the process, look at lowering the barrier if that is what experience dictates. No way the community will agree to excluding admin !votes, nor should they, as there are plenty of admins who are actively supporting this proposal. I suspect the alternative is a rejected proposal and continuance of the status quo. --Tryptofish (talk) 19:19, 10 January 2010 (UTC)


 * I'm not a fan of stressing any particular percentages at all - it is what attracts me to the actual 'consensus' approach. The decision to de-sysop should either be a fairly straightford one, or will need a discovered consensus - and a small party (to analyse the statements given) is better for that: ie it is the job of the bureaucrats. Having percentage safeguards is the wrong kind of thing to do in these matters, as I think they pose more problems than they solve. And people have different ideas of where they should be too. If the bureaucrats are clearly biased and corrupt - then it will now be easier to question them. The only way we are going to get anywhere 'near' them is via something like this - so I suggest not compromising things (and maybe ruining things) with percentages, and just give them the power to judge, knowing we expect them to be fair. It will make the whole place more accountable, and if beurocrates don't feel secure judging on these matters, then they really shouldn't be in there job.


 * All we need is the first line:


 * '''"Thus, for an Administrator to be de-sysopped, a bureaucrat will review the discussion to see whether both a minimum of 50 editors and a general consensus supports de-sysopping.

'''
 * It will force bureaucrats into behaving professionally, entirely based on the facts, the comments people have left for them at the CDA, and fully focused on the crucial idea of consensus - even if it is a 'discovered' one. If they don't do anything with a 'bad' admin (whether by de-sysop, or something else), the admins position will be more untenable in the future, than if he/she simply passed via a vote count. Essentially, (and this is all essential 'theory') Bureaucrats won't be able to 'game' the vote percentages so easily if they have no set values to go by at all. It might seem illogical at first, but I think that is the way it would pan out. Thus, the Bureaucrats would become more accountable themselves.


 * Also, having no 'targets' will make it a little harder for any 'bad' admin's admin-supporters/editor-supporters/meatpuppets/sockpuppets, to 'game' at the vote stage. In fact, all the problems with voting will be lessened to some degree. Matt Lewis (talk) 20:26, 10 January 2010 (UTC)


 * Given that few agree on the actual percentages (this was apparent in the voting at 5.4), I suggest we either leave them right now (and perhaps suggest they could be added after a 'trial phase') - or get another vote going - to make sure the final proposal is really something people are in unison over. Matt Lewis (talk) 21:13, 10 January 2010 (UTC)
 * I'm going to at least partly disagree, and in my doing so, please keep in mind that I did not, myself, support the percentages that emerged from the discussion. (1) I'm surprised by the talk about corrupt Bureaucrats. I really do not think that has ever been an issue. (2) Please note that the language is simply the language that already exists for RfAs, so no one is creating percentages here that do not already exist at the mirror process. (3) I think we have to be very careful at this stage about substituting the views of individual editors (me included!) who happen to be active in this stage of discussion, over the consensus views of many more editors who commented recently. (4) Let's look critically at the response at Wikipedia talk:Community de-adminship/Draft RfC/Archive 1, as well as at the criticisms that have come in from those editors who completely oppose this proposal in any form, and who can be expected to work hard to defeat the proposal that comes out of here. Option 5.1 (70% to desysop) got 19 support, 17 oppose. Option 5.2 (30% to desysop) got 1 support (please note who!), 25 oppose. Option 5.3 (>50% to desysop) got 6 support, 16 oppose. Now it needs to be pointed out that these numbers are limited in usefulness: some editors oppose all options but didn't !vote, and some editors do, indeed, oppose any fixed numbers. What this tells me is that it is right to use 5.4 (approximate percentages, nothing rigid), and that there really is no one option that has overwhelming support. But, if we go with anything like 30-50%, there will be overwhelming opposition, and that strong opposition will emerge unless we do something to spell out that the numbers are in the vicinity of 70-ish, where there is only tepid opposition. Accept the consensus we have now, and I expect the proposal will pass, and we will then find in the first few CDAs that nominated admins may well be able to game the numbers, in which case the community will adjust the policy based on experience. But make the changes proposed here, and the proposal will be rejected, and we'll have nothing. --Tryptofish (talk) 22:17, 10 January 2010 (UTC)
 * I suppose I'm thinking from the general pesky 'editors' point of view. The lack of trust in all of the Wikipedia heirarchy honestly cannot be underestimated - from what I've seen as an editor, I cannot stress that enough. I don't necessarily share it all the time, but I understand where it stems from - people normally interact in worlds that are far more advanced than Wikipedia, and the basic lack of accountability on Wikipedia simply invites distrust. I can see a stumbling block with percentages (and a very easy target to go after), and I'd love to go back to before the Christmas period (when I wasn't aware of this) and get in a proposal of my own, developed from SilkTorks 5.4: Simple consensus, and a 6-month trial period to see how the bureaucrats fare. I admit I see this whole process from a different perspective to many perhaps - for me it's all about natural evolution, cross-Wikipedia accountability, and bridging the gap between editors and bureaucrats. Matt Lewis (talk) 23:00, 10 January 2010 (UTC)
 * Agree, Trypto, with every word you say just above. Sarah777 brings up a valid point, but I don't think it should be acted on.  We have to trust the 'crats to make the final call in close and close-ish votes, and for them to discern where admins have banded together to protect their own in COI or borderline COI ways. Jusdafax   22:41, 10 January 2010 (UTC)
 * I'm personally saying that we should trust the bureaucrats to find consensus without having to consider set percentages (which we can't agree on, and could be problematic in the poll) Matt Lewis (talk) 23:00, 10 January 2010 (UTC)


 * To be even clearer - many haven't voted, and 70/80 could lose it. Will general editors go for 70/80? And even if it did pass, we can't guarantee 70% would be lowered in the future, if lowering appeared to be beneficial. Quite a few people in 5.4 said they would rather 60/80, or that 70/80 was too high. As we are working on 5.4, this is highly problematic. I'm thinking of the possiblity of a refining poll, maybe with polling the idea of 'plain consensus' as well. We musn't forget that 5.4 was supposed to be about consensus too - which clearly attracted people in itself.This polling 'run' could be actually at the final RFC. As long as the polls are logical, rational and restricted, and lead on from each other, there is no reason why there has to be only be the one. In real life, these kind of matters are often decided in a series of polls (at the public stage). It strikes me as the fairest was to get everything out of 5.4.
 * Final point - The idea of 'balance' in CDA mirroring RfC is floored I think, as an RfC is an optimistic thing, whereas a CDA is a seriously negative one: they are not really comparable at all, and are just really two points or stages in the same process. Matt Lewis (talk) 00:40, 11 January 2010 (UTC)

In agreement with Trypto and Jusdafx above. I also agree with Matt Lewis that the analogy may be more like looking through a telescope than in a mirror. The experience is different depending on which end you are looking through. I think this could be especially true for the 'crats. Some numbers: Admin 'Example' does something dubious and as a result loses the trust of the community. 75% of regular editors and 45% of admins support the desysop proposal. If 200 non admins and 40 admins participate the result is 70% thumbs down. Given the fairly small number of admins as a percentage of all editors it is not an incredible outcome. Ben  Mac  Dui  09:08, 11 January 2010 (UTC)


 * So, in the above example that uses the higher 70% threshold (ie a 70-80% 'uncertainty zone'): 150 out of 200 editors voting 'For' desysop (removing admin status) could essentially be outweighed by only 22 'Against' votes from just 40 voting admin. The maths being; 150+18 'pro' votes =168, which is 70% of the total 240 votes, and 70% is a percentage that would most likely see the bureaucrats 'saving' this admin (given the way the admin/heirarchy feel, and that consensus is tied to percentages), whereas using a 60%-80$ uncertainty zone (or judging on consensus only) would simply have to see this particular example admin desysoped (as has actually abused his status in some way, and lost the trust of the community).


 * Comment to all: I can't quite follow the 'agreement path' in the comments above. I think we all need to be really clear with other at this crucial stage - ambiguity can really impair these things. Matt Lewis (talk) 14:12, 11 January 2010 (UTC)

Possible outcomes short of full desysoping
Support for 8.2

"Instead, allow for more discussion and not simple bulleted !votes."

Clarify restoration
11: No serious opposition (although little support for) adding: "without the permission of the Arbitration Committee or another person or group empowered to lift those restrictions."

Canvassing
15 Limited support for "Parties to the CDA process may legitimately contact other editors to provide input, but must at all times do so in strict accordance with WP:CANVASS."

Improve language
Three specifics received support.

Reasons for speedy close
Suggestion to spell this out.

Certification vs Quorum
Discussion about the conflict between nominations and WP:CANVASS.

Before a CDA nomination
Suggestion and general agreement that wording be tightened: "A CDA request may be initiated only after substantial community discussion at a suitable venue, such as Administrators' noticeboard or Requests for comment/User, has failed to produce a resolution."

Proactively scheduled review
...after say 5 CDAs.

"All editors"
Not clear to me (MacDui) what this was intended to amend. Perhaps a default statement about using standard RfA procedures needed?
 * That would be fine. It was intended the amend the statement that all editors are allowed to vote, which did not exclude sockpuppets and other disruptive "editors". &eta;oian   &Dagger;orever &eta;ew &Dagger;rontiers  06:15, 15 January 2010 (UTC)

Short-circuit & Need
Possible material for an FAQ.

Comments
All the above should be fairly easy to send to WT:CDA and implement save the following:
 * Q5 - specifics include
 * we are clear that "a minimum of 50 editors must express an opinion"
 * 5.4 is the only proposal receiving positive support, but this needs a more thorough analysis.
 * what does the tilde mean in "above ~80%"?


 * Resolving the Canvassing issue
 * All editors?

I'd gratefully appreciate it if one or more volunteers would check that the above summary of the results corresponds with the detailed archive. I doubt there is any huge problem, but it would be embarrassing if a discrepancy, however small, were discovered during the RfC. Ben  Mac  Dui  19:47, 8 January 2010 (UTC)
 * For one, I'm going to try and concentrate on these questions tonight. I think people want to see the final RFC up and running so they can vote, but it has to be presented properly and without errors, you're right. Matt Lewis (talk) 20:02, 8 January 2010 (UTC)