Wikipedia talk:Requests for adminship

Feel free to join the discussion on the future of Requests for adminship process at Wikipedia talk:Requests for adminship/Reform. Everyone's comments are welcome!

The next step

The next step is to ween editors off these pseudo-vote tags such as support. Whether you support or oppose isn't as important as your rationale. --Tony Sidaway 21:10, 12 April 2007 (UTC)

• I'd thought of that too and toyed with removing support and the like from people's comments. But, I felt if I did I'd be lynched :) --Durin 21:13, 12 April 2007 (UTC)

Deeper shit than I thought indeed. Perhaps the 'crats could be encouraged to make comment on their rationale for using the community rationale? Or is it easier to divide oppose by support+oppose and if the answer is around 75% then job done? The Rambling Man 21:15, 12 April 2007 (UTC)

I'd say it's best to steer away from changing other users' comments with the possible exception of removing blatant personal attacks or WP:BLP violations. -- Black Falcon 22:35, 12 April 2007 (UTC)

• Baby steps, Tony! Let's see how this current raft of changes pans out. I'll be particularly interested in seeing the bureaucrats' response to this and how it impacts on the closing process. --bainer (talk) 00:12, 13 April 2007 (UTC)
• Until this is done at XfD, it is way, way premature to do it here. -Amarkov moo! 03:35, 13 April 2007 (UTC)
• At first I really disliked removing tallies, etc, but the more I think about it, I like it. Why not just ban people beginning their sentences with Support and Oppose and so on and so forth? Moralis RfA is a really good example of how it should be done, although it could do with being a bit neater. Of all the proposals to 'fix' RfA, this seems the simplest and the best. User:Veesicle 04:04, 13 April 2007 (UTC)
  • Ban? On what grounds? Do we really want to start down the path where we create policies or guidelines that dictate how editors can or cannot write their comments? We have policies that discourage or ban certain harmful behaviours, but hos is starting a comment with "support" or "oppose" on the same level (in terms of harmfulness to the project) as personal attacks and legal threats? -- Black Falcon 00:03, 15 April 2007 (UTC)
• In other news, I've decided to move out of my modern home into a cave. The next step will be weaning myself off the use of fire. Everyking 05:07, 13 April 2007 (UTC)
  • Don't worry, I'm sure before long someone will have figured out how to demonstrate that there is a consensus that you don't need fire. Dragons flight 09:02, 13 April 2007 (UTC)
    1. Oppose, user has not demonstrated a need for fire. --Slowking Man 09:46, 13 April 2007 (UTC)
• Support... I mean, rather, I think that this is a good idea. Ideally I'd like people to give their reasons for and against any particular candidate but perhaps that is too much candor to expect... Kat Walsh (spill your mind?)
• Strong oppose. But seriously, I think this is a bad idea and one that will be met coolly. We're not really going to disallow anyone from saying "I fully support this editor" are we? If we do and somehow determine that you can only say something on RfA if it contains a minimum of 4 coherent sentences and both arguments for and against a candidate, we're going to be drowning in confusion. Pascal.Tesson 20:36, 14 April 2007 (UTC)

So, we're going to remove the tally counters, mix up the support, oppose and neutral sections and deny contributors the ability to say either of those three words in their contributions? That would make each application an impenetrable mass of opinions that Bureaucrats would have to wade through in order to analyse and determine consensus. Not only the Bureaucrat, but each interested contributor and the applicant would also have to perform their own analyses in order to see that due process was carried out and that the consensus was correctly determined. I think that this would vastly over-complicate the system. Have people looked at long XfD discussions, which do appear in this format, and attempted to determine the balance of opinion and to see if one contributor has made a particularly compelling argument? It's a hard thing to do and opinionated contributors are always ready to call out the closer of such an XfD if the decision runs counter to their wishes. I feel that a reform that increases the signal-to-noise ratio on RfAs would be a step backwards. A Bureaucrat doesn't have to rely upon the tallies and clear divisions of the RfA/ RfB templates when closing and determining consensus but they are useful to show the trend of the discussions and any discrepancies between the final tallies and decision. (aeropagitica) 23:53, 14 April 2007 (UTC)

And because of this "reform," should the 75-80% benchmark be discarded as pointless, since you need to calculate for that? This will invite promotions that previously would have failed miserably. This "reform" is idealism (RfA ≠ vote) going against practicality and reality (RfA = ambiguous mix between discussion and vote). —²¹⁰physicq (c) 00:07, 15 April 2007 (UTC)

Nah, the next step is to ween editors off the mindset that "voting is evil" and onto the mindset that "voting is necessary in order to measure consensus of large groups of people in finite time". Since RfA discussions are finite time discussions, voting is necessary. GRBerry 02:52, 16 April 2007 (UTC)

• Since when does voting measure consensus? --Iamunknown 03:01, 16 April 2007 (UTC)

Current attempt at RFA reform

I have reverted User:Thebainer's removal of the tallies from individual RFAs. I have not reverted the same action at Template:RfA. I don't care one way or another about the tally, but don't think already open RFAs should be modified.

On a similar note, I'd like to express my ambivalence about the fact that Wikipedia:Requests for adminship/Moralis was refactored in the middle of the discussion. I also do not find the current format to be very reader-friendly (lack of separation between "S", "O", and "N" sections). Images of RFAs looking like this are not exactly reassuring.

I have written similar comments at User talk:Thebainer, but am posting these comments here so that the discussion may be centralised. -- Black Falcon 22:27, 12 April 2007 (UTC)

• Moralis's RfA is already being discussed in the thread above. A question: do you find it easier to form an opinion about a candidate based on how many people have already supported or opposed? That's what you seem to be suggesting. Mackensen (talk) 22:38, 12 April 2007 (UTC)

If this was a test formatted RFA, shouldn't there be a banner saying that this was a test formatted RFA? Real96 00:19, 13 April 2007 (UTC)

You can see the existing discussion on my talk page as Black Falcon points out. The key reason why the tallies at the top are a bad idea is that, as the first thing anyone sees when they come to an RfA, they will undoubtedly have a psychological impact on that person when they come to voice their opinion, conscious or no. It's prejudicial to have the numbers up top - people should be exposed to the nomination and the questions first, and then the opinions people have expressed, at which point they can count for themselves if they find that useful for them; they shouldn't come to the debate with the numbers already in their head. Not to mention possible peer pressure effects.

It seems that the overall response has been positive so far, I haven't noticed anyone who disagrees with the idea per se (as opposed to the method) so if there's no more disagreement I don't see why we can't do this for all RfAs. --bainer (talk) 00:25, 13 April 2007 (UTC)

Reading your comment, I just had the admittedly bold idea to abolish voting and commenting in the RfA page altogether. The page could be reserved for the nomination only, while voting comments would be sent to an e-mail address and viewable only for bureaucrats. The normal thing should be to promote, unless convincing reasons are provided not to. Promotion would then follow (or not) at a b'crat's discretion (or two or more b'crats, if there are many responses), with a statement of the main concerns registered (if any) and on the decision being forwarded to the RfA page for everyone. —AldeBaer user:Kncyu38 02:06, 13 April 2007 (UTC)

I whole-heartedly disagree with anonymous commenting. Issues can and do come up that would immediately disqualify people from adminship in the eyes of many, but if these are somewhat hidden and aren't made public, then most voters will continue to remain ignorant of them. The discussion, at least, must always remain in the public eye and under public scrutiny. --Cyde Weys 04:07, 13 April 2007 (UTC)

OK, I disagree with the idea, in the strongest possible terms. Everyking 05:19, 13 April 2007 (UTC)

I don't. In fact, I agree with it. --Iamunknown 05:21, 13 April 2007 (UTC)

• I think it's a very good idea to remove those sections. Moralis's RFA is a step forward. >Radiant< 08:11, 13 April 2007 (UTC)

  I wholeheartedly disagree. The page is currently a mess, and makes it absolutely impossible for crats to decide one way or the other. Errabee 08:47, 13 April 2007 (UTC)

  • Perhaps the crats should be the judge of what they can make a decision on? >Radiant< 09:00, 13 April 2007 (UTC)

  Oh, I'm absolutely sure some crats would love it this way. No public scrutiny possible whatsoever, and if some form of protest arises (Carnildo comes to mind), one can always hide behind this poor structure. Definitely opposed to this experiment. Errabee 09:18, 13 April 2007 (UTC)

• Part of the motivation behind the proposal was to mess up the sections which made it easy to sort the supporters from the opposers, and thereby de-emphasize the support/oppose counts during the discussion. I am still in favor of letting the numerics count for something at the final close, but I don't think they should be so prominent during the course of the discussion. I don't think the RFA is any more messy than an AFD debate, although it is considerably longer. Sjakkalle (Check!) 08:56, 13 April 2007 (UTC)

Ok, putting the question of the sections aside for the minute, is there anyone else who has a problem with removing the tallies from the top of each RfA? The reasons for doing so are outlined earlier in this thread. --bainer (talk) 09:21, 13 April 2007 (UTC)

No, I find that absolutely reasonable. Public scrutiny doesn't require it. Just to make my position clear. Errabee 09:34, 13 April 2007 (UTC)

• Ironically, and just like the last time something like this was tried, the guy is now getting opposed because people don't like the change in system. >Radiant< 10:55, 13 April 2007 (UTC)
  • Which any bureaucrat worth his salt will rightfully ignore (don't shoot the messenger). Mackensen (talk) 11:04, 13 April 2007 (UTC)
    • I hope not. The fact that Moralis consented to this experiment tells about his lack of foresight of the consequences of his actions, as everyone with some common sense could see that this experiment would turn the process into a right mess. Errabee 11:16, 13 April 2007 (UTC)
    • (edit conflict) Really? That's the platform in your RFB, which appears not to be finding consensus. In that case the bureaucrats have no mandate for ignoring what (per consensus) are valid votes. To clarify: I opposed in good part because the user clearly didn't understand that accepting was poking the ants' nest. I also don't like the formatting, but that was not the main reason for the oppose. There are several procedural reasons for opposing an RFA: Too soon after last RFA, failed to answer questions, poor answers to questions, badgers respondants, no acceptance etc. I see an oppose on the basis of the candidate's extremely unusual selection of formatting as being no different than these.AKAF 11:22, 13 April 2007 (UTC)
      • "Badgers respondants?" If you raise a point in an RfA you should be prepared to discuss it. RfA is a discussion, not a vote. Even when the totals are included it still says right at the top "Discussion." Mackensen (talk) 11:28, 13 April 2007 (UTC)
      • I'll leave the question of consensus to the bureaucrats–that's why they were appointed in the first place. However, we're talking about Moralis's RfA, not my RfB, and as several users have told you (including me) it's downright unfair to Moralis to conflate the two. Mackensen (talk) 11:26, 13 April 2007 (UTC)

I'm not saying that the reasons above are valid, just that if you look at other RFAs, you'll find opposers who use those reasons. I think that it's not unreasonable to expect Moralis to understand the effect of making this sort of proposal in his RFA, when he has the opportunity of observing the effect of you making a (different) proposal in your RFB on the same page.

As to "Badgers respondants", I think that there's a difference between a discussion and badgering which each editor can decide for himself. There's a limited amount of discussion that can take place in an RFA without it becoming repetitive, and everybody has their own line which they'll draw. The extreme end are of course editors who can't let anyone else get in the last word, and some people don't want them to be admins.AKAF 11:44, 13 April 2007 (UTC)

Inappropriate or irrelevant reasons for opposing may be ignored by the bureaucrats (see, for instance the Danny RFA). --Tony Sidaway 12:21, 13 April 2007 (UTC)

It is extremely rare for a vote based on one of the list above to be ignored as inappropriate or irrelevant. While it's true, as you say that the official lattitute of the bureaucrats is wide, custom limits them to about a 5% shift. In this case both Mackensen and Moralis had the option of a standard RF[AB] with an external porposal and straw poll. That they both chose not to use the more normal method, and to conflate the two. AKAF 12:51, 13 April 2007 (UTC)

There has been no confirmation by the bureaucrats that Danny's RFA wasn't just a one-time exception. Until we hear otherwise, it would be reasonable to assume that there has not been a significant change in the way most RFAs will be closed. CMummert · talk 12:59, 13 April 2007 (UTC)

• Errabee, opposing Moralis because of the format of his RfA is in my opinion way out of line. If all you can find to oppose him for is the format of the RfA, then he must be a fine candidate indeed. Why not spend the time to actually evaluate him. Like, reviewing his edits, his messages left to other users, etc. rather than just opposing him for something that has nothing to do with him except that he agreed to let an experiment be run. The project is in no way harmed because he was willing to be a guinea pig. This is one of the ways in which we move things forward. This sort of oppose is wholly reactionary and has nothing to do with Moralis. If you want to oppose the experiment, fine. You can do so here on this talk page.
• It is a horribly bad idea to be opposing a person for helping to make change happen. This sort of behavior casts a very cold, chilling shadow on forward movement to better the project. People become fearful of making change because somebody will oppose *them* rather than the *concept*. Even if this experiment fails, it does not mean we gain nothing from it. We do not make our way forward solely by attempting only those changes we are certain will succeed and be uncontroversial. --Durin 13:08, 13 April 2007 (UTC)

While I can see your point, it's not unreasonable to oppose an RFA for the reason that the candidate doesn't feel the need to do things the way everybody else does them. The question is whether the editor is for change or just against the status quo. One of the charges against Moralis is that he is too inexperienced to understand the way things are done on wikipedia. Thus I don't think it's too unreasonable to assume that his acceptance of this unusual RFA is a strike against the status quo, whether intentional or not. AKAF 13:22, 13 April 2007 (UTC)

The job of an administrator is not to maintain the status quo--if anything, rather the reverse is true. An editor's decision to adopt a new format has no bearing on his suitability to be an administrator. Such arguments will be discounted. --Tony Sidaway 13:42, 13 April 2007 (UTC)

@Durin: There is a big difference between attempting those changes we are certain will succeed, and attempting those changes anyone with common sense is certain will not succeed. With the former I don't have a problem if the experiment fails, but in my opinion this experiment falls into the latter category. Anyone who supports experiments that are doomed to fail is either not capable of evaluating the consequences of these experiments which casts a shadow about his or hers capacity to take well-considered decisions (best case scenario) or is actively trying to frustrate other editors. This raises serious doubts about adminship.

I also don't agree with AKAF, because an RfA candidate who doesn't feel the need to do things the way everybody else does them, is not necessarily a bad thing. Thinking out of the box may produce some beautiful unexpected results. If Moralis is too inexperienced to understand the way things are done on wikipedia, then he can never launch a strike against the status quo by accepting a request to do an experiment. Errabee 13:52, 13 April 2007 (UTC)

• You presume that the only possible outcome of a format change is whether it is accepted or not. This is false. Sorry. --Durin 14:15, 13 April 2007 (UTC)

I'm curious to know from Tony and Durin which other oppose !votes based on a change of format you think should be discounted? If a candidate chooses not to answer any of the questions, but only to answer in threaded replies, would you strike all oppose comments? If a candidate moved all text not "support" or "oppose" to the talk page, would that be an invalid reason to oppose? AKAF 14:59, 13 April 2007 (UTC)

I have been refraining from commenting, because I don't necessarily feel that it's my place to involve myself in a discussion about my RfA. There is, however, one thing that I feel needs to be mentioned, and I don't intend to take part in this discussion again.

AKAF, your argument, if I understand it correctly, is that my choosing to allow the reformat shows that I A) support the change, which is no longer a subject of discussion as it has been established that I'm neutral on the subject, or B) didn't understand the consequences of my actions when I agreed, due to inexperience, poor judgement, whatever.

I think your position is a self-fulfilling prophecy. You are opposing me because I failed to see that my actions would see my RfA opposed. In other words, you oppose because I didn't realize that if I did what I did, people would oppose. Please don't think that I'm being snippy here, because it is most certainly not my intention, but the only way I can think of to paraphrase this is:

"Mora shouldn't be an admin because he didn't realize that if he let his page be reformatted, I would decide that he shouldn't be an admin." I just don't see how that works.

As stated, I won't be commenting here again, as I feel I'm stepping out of line just by commenting now. I just wanted to bring that up. --Moralis (talk) 15:23, 13 April 2007 (UTC)

Your summary is a fair statement of my position. There were plenty of other options for trying out/getting an opinion on the new format, a straw poll on the reformatting of a closed RFA an an example. However I've realised that a large number of very experienced editors disagree with my position, so in the interests of consensus-building I've changed my !vote. AKAF 15:35, 13 April 2007 (UTC)

The biggest problem with RfA is I believe that there are no consistent criteria about whom to pass and whom to fail. Mixing support and oppose comments makes things more confusing without making the process better, but otherwise is just a cosmetic change, unlikely to fix anything. Oleg Alexandrov (talk) 15:44, 13 April 2007 (UTC)

As someone who has made numerous suggestions aimed at reforming RfA, I suppose I'd be hypocritical to say something like "I'm too set in my own ways to change now". Whether or not this format can work, however, I simply do not know at this point. It is fairly easy to point out advantages and disadvantages to pretty much every single aspect of it — example: this format would make it more difficult for people to support/oppose "per User:X"; many will say that this is a good thing. On one hand, it can be. On the other hand, however, it can also be said that it takes away from the process in the sense that a key aspect of presenting a rationale and diffs/links is that if someone brings forth relevant information on the candidate, this will help to inform other participants better, and possibly people will make their decision taking into account the information/evidence presented; but if it becomes too difficult to find those key posts in a neverending string of comments, then a very relevant post will not serve the purpose of helping the larger community make up its mind on the candidate. But we can't really know which way it will go, not at this point at least. Like with everything that is new, it will take a trial run to determine whether it will work out or blow up in our faces.
As a Bureaucrat, my main concern is that in heavily commented on RfAs, for instance with 130 supporters and 50 opposers, the process of assessing the community decision on the candidate may become difficult, and for that matter, perhaps less transparent in the sense that it may not be easy for other people to realize how the decision was made without them having to go through every single comment themselves. But as I said before, it would take a test for this to become apparent, so I'd rather not make any judgement until I've been able to go through at least one RfA in that system (and preferably, close it).
I'd just like to make one thing clear, although most certainly already know it: if later on it is decided that the system doesn't work and it should be abandoned, that will not change the final outcome for any RfA closed that was run with this system. If the candidate got promoted, then it stands; if the RfA fails, then s/he will have to run again at a later time. Redux 02:45, 14 April 2007 (UTC)

The format used for Moralis' RFA is just ATROCIOUS.Rlevse 17:32, 14 April 2007 (UTC)

There are a lot of comments above and I wish to respond to a couple of them.

• To Mackensen: No, I do not "find it easier to form an opinion about a candidate based on how many people have already supported or opposed". I do, however, find it easier to read a discussion when all "support" and "oppose" comments are in their respective section (it also cuts down on unnecessary repetition).
• Regarding anonymous commenting: I full agree with Cyde Weis. Comments on candidates (both positive and negative) should be made publicly so that others may view them and use them to make decisions.
• Regarding Moralis' RFA: All comments of the rightness or wrongness of modifying one active RFA without discussion at WT:RFA aside, I must agree that the format is nothing short of disastrous. The discussion is convoluted and also overly, excessively, redundantly, and unnecessarily repetitive. If the goal of this format was to get people to actually read the comments instead of just looking at numbers, I think it has been a spectacular disaster (think along the lines of the sinking of the Titanic or the Great Fire of London). Few if any will read through a discussion that already has 3 arbitrary section breaks; at most, they'll skim through parts of it and ignore the rest. -- Black Falcon 21:19, 14 April 2007 (UTC)

Threaded discussion

This isn't a reform question, but a current procedural one. Should threaded comments be allowed on the project page? As in this example: Wikipedia:Requests for adminship/Kzrulzuall. I would think discussion AFTER commenting or about someone else's comment should be done on the talk page, but I could be wrong. Anynobody 09:42, 13 April 2007 (UTC)

I currently think it would be nice to have a threaded discussion section (where people make arguments that are then discussed) and a separate voting section (where no additional comments or threaded discussion is allowed). That might allow us to have simple voting by people who don't have anything to add to the discussion and useful discussion of arguments at the same time, without putting oppose and support arguments in different sections and without making a hard-to-read mess of the RfA like in Wikipedia:Requests for adminship/Moralis (everybody who wants to comment but doesn't have a new argument will make the page harder to read; if these people participate at all, perhaps voting would be better than a comment that doesn't really add anything useful). Kusma (talk) 11:14, 13 April 2007 (UTC)

This doesn't seem to be a step in the right direction. A voting process on Wikipedia is very susceptible for trolling, and separating votes from comments would only make things more difficult. I would like to say that if you don't have anything to say, then don't express your support or opposition either. That's why I'm absolutely not a regular in this process. Errabee 11:42, 13 April 2007 (UTC)

I think that discussion processes are much more susceptible to trolling than voting processes. "Not enough image talk edits" is a troll comment, "oppose" is not. Kusma (talk) 11:47, 13 April 2007 (UTC)

I think a simple oppose without explanation is entirely unhelpful, as a support without explanation would also be unhelpful. Furthermore, it allows people to oppose or support every nomination, because they are opposed to the process of RfA. And finally, finding sockpuppets is that much harder to do when explanations are not required. So please no separate voting section. Errabee 11:53, 13 April 2007 (UTC)

I don't think it is possible to have a meaningful "discussion" about whether somebody should become an admin if more than 10 or 20 people participate. If we do not turn to having a committee determine who should become admins, something resembling voting is therefore a likely result of all efforts that allow for wider community participation. I think we need to move in either the direction of wide bureaucrat (or other committee) discretion given some input from the community, or towards more pure voting without much bureaucrat discretion. The current state of affairs is just a mess that makes nobody happy. Incidentally, if we allow people who are opposed to the process of RfA to vote, and that leads to too many failing nominations, we can simply adjust the percentage necessary so we produce enough admins. The goal of any RfA reform needs to be to triple the throughput of RfA, and to promote admins quickly enough that we can tackle the backlogs effectively. How we do that I don't care, but I don't think making the process more confusing is going to help. Kusma (talk) 12:03, 13 April 2007 (UTC)

A request for adminship is a consensus-building process. Threaded discussion is the primary means of building consensus on Wikipedia, and is encouraged. Treating a comment as a vote (or even a "!vote", whatever that is) is deprecated and will probably not have the effect you anticipated. There's no reason why 200 or more people shouldn't contribute to a discussion, if only to express their agreement with a good argument expressed by another editor. But it isn't necessary for 200 or more people to participate if they don't want to. Years of poor formatting of the RFA page has tended to foster the belief in some editors that RFA is a vote and that the sheer numbers alone matter. Once freed from that misapprehension, and seeing that their opinions are adequately represented in the discussion, it's quite probable that they'll wander off and find something else to do instead of contributing to useless pile-ons. --Tony Sidaway 12:27, 13 April 2007 (UTC)

I can't tell you how many times I've seen RfAs fail because the nominee dared to respond to the opposition. You just have to sit there, shut up, and take it. If you disagree, you are "whiny". If you provide diffs with your disagreement, you are "argumentative". And even more oppose votes come rolling in because of it. Very, very, very few people are interested in having their opposition questioned by others, and even fewer will actually change their minds. As long as it's considered valid to oppose a person for responding to doubts (or even attacks, which are common) in his own RfA, it will never be a consensus-building discussion. Kafziel ^Talk 13:51, 13 April 2007 (UTC)

I think the manner of addressing opposition matters. On my RFA (admittedly about two years ago now) I systematically identified those whose opinions, in my opinion, could be swayed by further discussion, and discussed matters with them on their talk pages. A significant number of people changed their opinion in my favor. I don't recall anyone reacting in a hostile manner to my personal and very polite approach, but if they had I would not have pressed the matter. I would have been prepared to wait for another month or two. --Tony Sidaway 14:11, 13 April 2007 (UTC)

But tiptoeing around like that isn't conducive to consensus building. The people who oppose you don't worry about being honest, so if you have to treat them with kid gloves then it's not a real discussion. Kowtowing to the opposition is not the same as working toward a fair decision. Kafziel ^Talk 14:18, 13 April 2007 (UTC)

Part of showing that you're administrator material is demonstrating the ability to persuade. I did that. --Tony Sidaway 15:27, 13 April 2007 (UTC)

Persuading people to support you is still just campaigning for votes; there's no honest discussion there. We don't expect administrators to be obsequious, so we shouldn't expect it from administrator hopefuls, either. Nominees can participate in real consensus building if we stop allowing oppose votes based on their arguments. If we don't let everybody participate equally (including the nominee), then it's just a vote. Kafziel ^Talk 15:41, 13 April 2007 (UTC)

Consensus building is almost impossible in a format where everybody is allowed to say aye or nay without explanation. When a candidate applies for adminship, we should all assume Good Faith and the default action should be to give the tools to the candidate, unless severe enough objections have been raised that have not been answered to satisfaction. This means of course that objections have to be discussed, and anyone objecting to a candidate should be prepared to discuss this objection with the candidate or others. Consensus is then achieved when all objections have been answered to satisfaction. Errabee 14:34, 13 April 2007 (UTC)

All discussions on Wikipedia are run this way. Someone can just say "I agree with this." It helps to build consensus, but obviously it helps more to engage in proper discussion. --Tony Sidaway 15:29, 13 April 2007 (UTC)

I believe that whenever someone supports or opposes something without an explanation, he simply agrees with anything previously said by other supporters or opposers. We shouldn't discourage someone to voice his opinion because he can't come up with something original. Michaelas¹⁰ 15:44, 13 April 2007 (UTC)

If he can't come up with anything original, why even participate at all (except to add a tick mark to the tally)? The point he agrees with has already been made, and it should have the same weight whether one person says it or 100 people do. (Unfortunately, we don't live in a perfect world, so that's not the case. Bureaucrats are only human, and they will tend to side with the masses. Which is why I think the tally should stay; if it matters how many people agree on a certain point, then it's a vote and we shouldn't pretend otherwise.) Kafziel ^Talk 15:52, 13 April 2007 (UTC)

Although agreeing with another's argument might not weigh as much as the original argument in the bureaucrat's discretion, it's still helpful in establishing consensus — whenever someone thinks that a certain argument is satisfying or not, he may explicitly illustrate so by either answering him in an opposite action, or doing the exactly the same without a further comment. I concur this isn't a vote, but we don't want ending up with an RfA consisting of one oppose and two supports. Bureaucrats should be affected by other people's judgment towards the arguments. Michaelas¹⁰ 16:50, 13 April 2007 (UTC)

I agree an RfA with 2 supports is not something we want, but I wouldn't object to an RfA with just 1 oppose (and no supports). If the objection is rebutted satisfactorily, then the candidate should be sysopped because there are no objections. See also my proposal at Wikipedia talk:Requests for adminship/Reform. Errabee 16:57, 13 April 2007 (UTC)

It absolutely matters how many people think a certain way, regardless of whether they have something original to say. This is why sockpuppetry is a big problem. In measuring consensus, there is a huge difference between one person saying something ten times and ten different people saying something once each. The former is just one person's opinion shouted many times, and the latter represents ten people. Sockpuppetry prevents us from recognizing the difference. RfA is about whether the community trusts someone with the tools; knowing how many members of the community do or don't trust is critical to the decision. GRBerry 17:29, 13 April 2007 (UTC)

I couldn't agree more. That's why I support a straight vote in RfA. Consensus and debate has its place in articles, but when trying to gauge how much of the community trusts a person, all we need to know is how many people do or do not support the nomination. It's nice to have lofty ideas about consensus, but in the real world, in the end, it has to come down to numbers, not arguments. Kafziel ^Talk 18:55, 13 April 2007 (UTC)

If you don't trust the canidate, all you really have to do is give your argument for why, and the B-crat reads that. "xxx is an editwarrior on articles yyy and zzz as a result I cannot trust him with the tools", etc. —— Eagle101 ^{Need help?} 20:58, 13 April 2007 (UTC)

In that case, it works fine. But what if my concern is "I don't like when yyy made this decision xxx against consensus, and I don't trust them to not keep doing that", but most people like that decision? Assuming we don't have a voting model, either we give bureaucrats the power to unilaterally decide that a decision was bad, which is not a good idea, or my lack of trust has to be ignored. -Amarkov moo! 23:20, 13 April 2007 (UTC)

To clarify my question, I had pictured something like this:

Support

1. Support - Great editor Editor 1 00:01:01 (UTC)
2. Support - I have seen mostly good decisions from him, plus his knowledge of x is outstanding Editor 2 00:10:17 (UTC)

• [[Wikipedia talk:RFA/Nominated Editor|Editor 3 has a question or comment]] about your statement, please see the talk page if you wish to answer.

The discussions can be valuable for various reasons so they should be allowed, however they can get in the way of the voting/comment section. This is especially true for new editors, so logically it stands to reason that said discussions be carried out on the talk page (after all discuss/talk are synonyms.) Anynobody 23:45, 13 April 2007 (UTC)

The problem is, even without the wording of the comment, that encourages only discussing your reasons if you feel like it, instead of discussion being part of the process. We have enough of the mindset that people don't have to respond to challenges of their votes, it doesn't need to be encouraged. -Amarkov moo! 19:06, 14 April 2007 (UTC)

Requests for Bureacratship

Is it possible for anyone who isn't an admin to RFB? Or is this only reserved for admins? Simply south 17:37, 13 April 2007 (UTC)

I believe that there is no technical restriction against that, but in practice, RfB for a non-admin, no matter how long contributing, would be completely impossible to pass. -Amarkov moo! 17:38, 13 April 2007 (UTC)

Indeed. Each RfB candidate is expected to adhere the highest standards in order to be trusted with such tools. Being an administrator beforehand is very improtant. Michaelas¹⁰ 17:42, 13 April 2007 (UTC)

I'm not 100% certain, but I believe that in Mediawiki the user rights for a bureaucrat are a superset of an admin's. Therefore, a bureaucrat has all the permissons of an admin plus a few more. An RfB for a non-admin would essentially be a combined RfA/RfB. Someone please correct me if I'm wrong. ChazBeckett 17:51, 13 April 2007 (UTC)

I am certain. A burocrat hasn't automatically admin rights in mediawiki. Snowolf ^(talk) _CON ^COI - 18:42, 13 April 2007 (UTC)

Correct. Depending on how the software is configured, adding +bcrat might be essentially adding +sysop, but if it's configured the way most people do it (which I would expect), then it would only add access to Special:Renameuser and Special:Makesysop, not adding the other permissions. ^demon^{[omg plz]} 19:24, 13 April 2007 (UTC)

It is possible to have a Bureaucrat that is not a sysop, but a Steward would have to promote, Bureaucrat's can promote user -> admin : admin -> crat : user -> admin/crat, nothing else. Crats don't automatically have admin rights, at least by default. Prodego ^talk 19:51, 13 April 2007 (UTC)

In MediaWiki (on a different wiki using the default setup) I was able to give a regular user the bcrat flag w/o having sysop flag. This user was then able to give themselves the sysop flag. This would make me assume the same is true on Wikipedia since I dont think the devs have really changed much in the way of how flags are applied (just merely created more flags).  ALKIVAR™ ☢ 04:03, 15 April 2007 (UTC)

• User permissions in MediaWiki are additive; however, as ^demon brings up above, it depends on the site configuration. The only way to know for certain is to actually test it in Wikipedia, which is probably not very likely to happen. We can't look at the LocalSettings.php file for here, as it contains database passwords and information of that type, so my intuition tells me that it is configured in an additive fashion. (For example, oversight users are not bureaucrats by default.) Titoxd^{(?!? - cool stuff)} 15:58, 14 April 2007 (UTC)

It should be possible for a non-admin to run for bureaucratship, afaik :-) --Kim Bruning 16:03, 14 April 2007 (UTC)

But that is the point, it is possible, but it is very unlikely that they will pass. Cbrown1023 talk 16:57, 14 April 2007 (UTC)

Anyone who could pass an RfB could pass an RfA. Anyone running for RfB who is not an admin first would only be making some sort of vapid political statement. —Centrx→talk • 16:59, 14 April 2007 (UTC)

As a trivial but possibly interesting point, on another wiki I made for myself, I made my main account a bureaucrat without making myself a sysop, not realizing it wasn't automatic. Go figure. So, assuming WP's is the same way, that's the technical answer. Not that it has any impact on the way it works in practice. Heimstern Läufer 17:06, 14 April 2007 (UTC)

Why would that be a vapid gesture? If a retired admin wants to now run for bureaucrat, that shouldn't be a problem, should it? It's a different set of responsibilities. You'd think people would welcome less "conflict of interest", or what have you. :-/ --Kim Bruning 18:09, 14 April 2007 (UTC)

I believe the question has been answered. It is technically possible to give an account Bureaucrat access without giving it admin access as well, but only a Steward can do that. Bureaucrats cannot promote like this. We can only set Bureaucrat rights only if the account already has admin access, otherwise it would result in a "double promotion". That is because the Makesysop tool for local Bureaucrats was created as a simplified version of the Makesysop tool for Stewards. It would be possible to conjecture that the configuration of the tool would indicate that it was never intended for anyone to be made a local Bureaucrat without having adminstrator access first, but that's just speculation. The bottom line is that policy does not require that only admins can request Bureaucrat access, but from the practical point of view, it is indeed unlikely that the community would ever bestow this access level, which is widely considered a more sensitive set of responsibilities — but mainly due to the technical ability to grant admin access to any other account — to anyone that thus far hasn't been trusted with any kind of restricted tools. Redux 20:06, 14 April 2007 (UTC)

In the case of a retired admin? --Kim Bruning 20:11, 14 April 2007 (UTC)

Instant ratification, obviously! (if their first name has a K or an M or an I, at least!) El_C 20:16, 14 April 2007 (UTC)

Why would the retired admin not get back admin tools first? What is the point of not having admin tools? They don't put the person into some special existential class where he would suddenly have a "conflict of interest" with the tools, and no conflict of interest without them. —Centrx→talk • 20:17, 14 April 2007 (UTC)

Because not everybody wants admin tools. -Amarkov moo! 20:20, 14 April 2007 (UTC)

Why not? You are perfectly free to not use them, and I don't see why you would not want to be helpful if you come across some attack page on Special:Randompage or rampant vandalism on an article, by deleting or protecting. Having admin tools does not create an obligation to start clearing backlogs. Usually the reason cited for people not wanting admin tools is they don't want to go through the onerous process of RfA, but an RfB would be no less onerous. —Centrx→talk • 20:23, 14 April 2007 (UTC)

Maybe they just want to limit themsleves to those particular functions; I really see no reason for any restrictions for them. El_C 20:23, 14 April 2007 (UTC)

What's the point though? If you happen to come across some vandalizing IP, why create additional work by reporting it somewhere and then waiting for the IP to get blocked while he continues vandalizing? —Centrx→talk • 20:26, 14 April 2007 (UTC)

I admire some of the ideas of Cincinnatus (though not his ideas on plebs, I don't think). People should only take on those responsibilities they intend to fulfill right now, and release any others. This also allows one to avoid David Gerard's law, by only maintaining a small subset of duties. --Kim Bruning 20:38, 14 April 2007 (UTC) "On Wikipedia, the reward for a job well done is another three jobs" --D. Gerard

In the default configuration, the only right that bureaucrats have (in addition to the standard ones) is "userrights", the ability to edit user rights via Special:Userrights. The local configuration here is certainly different, since only stewards can access Special:Userrights here, and bureaucrats use the lesser Special:Makesysop and Special:Makebot instead. --bainer (talk) 04:33, 15 April 2007 (UTC)

A Problem

I have noticed that someone (I do not know who) removed the vote counters on the RfAs of TonyTheTiger and Byrlcreem2. If this was done with the consensus of the community, then I guess I am worring about something that I should not worry about. However, I appear to not see anything that would suggest that anyone was allowed to do this. Just letting people know about something I noticed. Captain panda 20:41, 14 April 2007 (UTC)

They were created when the template did not have vote counters. -Amarkov moo! 20:42, 14 April 2007 (UTC)

[1] — Malcolm^{Use the schwartz!} 20:43, 14 April 2007 (UTC)

Thank you both for clearing that up! Captain panda 20:55, 14 April 2007 (UTC)

• Just for perspective; tallys were added to RfAs without any input from the community. It isn't a matter of being "allowed". It's a matter of attempting to do what is right for Wikipedia. --Durin 23:05, 14 April 2007 (UTC)
  • The tally isn't very helpful to me, I look at an RFA monitor... but removing it in the idea that it makes RFA less vote-like seems kind of silly to me. You really think b'crats don't do some kind of calculation? Obscuring what's going on doesn't change what's going on... it just makes the learning curve steeper for new people. --W.marsh 23:58, 14 April 2007 (UTC)

The tallies were removed for a number of reasons:

• The main reason is that having the tally as the very first thing that is visible in the RfA is surely a significant psychological factor. Rather than reading the nomination and the statement/questions and then the expressions of opinion with an open mind, with the tally there, people read them with the numbers already in their head. It's certain money that this has some prejudicial effect on people.
• Another reason is that the tallies promote the misconception that RfAs are only pure votes. That the weight of numbers is an important part of any analysis of consensus is not doubted, but to quote SlimVirgin, "The numbers incline but do not necessitate, to paraphrase Leibniz."
• Finally, the tallies are available elsewhere for those who wish to seek them out, and manual counting is of course possible if people find that useful.

I have seen no reason why the tallies should be kept as they are and exposed to everyone whether they want to see them or not. I would invite anyone who wishes to address the reasons I've offered, or suggest some reasons of their own why the tallies are necessary? --bainer (talk) 00:55, 15 April 2007 (UTC)

Endorse. Ben Aveling 01:10, 15 April 2007 (UTC)

What I object to is the idea that this kind of obfuscation really changes anything... RFAs will still be closed the exact same way, which is usually based on the numbers. All things like this do is give people a false sense that things have changed, just because we've obscured some of the evidence. --W.marsh 01:41, 15 April 2007 (UTC)

For the record, I strongly agree with W.marsh. Moreover, I know that people promoting change have the best of intentions so I don't want to criticize them too much but I do believe that making the changes and then arguing that anybody questioning the changes is opposed to evolution towards "what's right" is the wrong way to go about the whole thing. The whole debate around Danny's promotion has instilled a false sense of urgency into RfA reform that's not going to help us make good choices. I have already been very critical of the Moralis experiment (and, at the risk of repeating myself, this has nothing to do with my evaluation of Moralis as a potential admin) and however frustrating the slow nature of change on Wikipedia might be, it's something we have to deal with. There's an irony here: while the proponents of the changes feel they are moving RfA towards more productive discussion and less vote count, they are trying to impose them without proper discussion. Pascal.Tesson 02:20, 15 April 2007 (UTC)

The tallies are not necessary, but removing them solves little. Yes, it does make it so people can't just vote with the crowd, but it's still a vote, and will still be closed as such. You have to attack the problem to solve it, not just the obvious evidence that it exists. -Amarkov moo! 04:37, 15 April 2007 (UTC)

As I said above, I do not doubt that weight of numbers plays a part in any determination of consensus, and I don't pretend that removing tallies will change how bureaucrats close RfAs. But if you concede that it will make it less likely that people will simply vote with the crowd, and use their brains when approaching an RfA, then what is the problem in removing the tallies? That's all I'm setting out to achieve. --bainer (talk) 05:52, 15 April 2007 (UTC)

Fair enough. Note however that you've been reverted three times by three different people in the last couple of days. Perhaps you could give {{RfA}} a break for the time being, or try to talk things out more. Oleg Alexandrov (talk) 06:22, 15 April 2007 (UTC)

Yes, the changes were undone by people who were concerned that it hadn't been discussed enough. Several discussions ensued, in which only Everyking disagreed and he didn't provide a reason. Many other users responded positively. I then repeated the change, but it was again undone some days later, by someone actually offering a reason (convenience) and so I've discussed it again. --bainer (talk) 08:49, 15 April 2007 (UTC)

I think that if we want to stop people from voting with the crowd, we should stop asking people for their oppose rationale. I haven't seen pile-on voting occur that didn't use other people's rationale, so it might well be that the "discussing" is as harmful as the "voting" bit. Oh, and ceterum censeo we need to redefine "consensus" to be 66% so more people pass RfA. Kusma (talk) 07:18, 15 April 2007 (UTC)

I am not really into trying to regulate how people make decisions... e.g. make trying to make the masses use their brains. Actually that seems like an assumption of bad faith to call them mindless, but whatever. I mean we can do all kinds of silly tricks to supposedly discourage people from being mindless, like require !votes to be delivered in riddles or use a minimum of 150 words and all kinds of silly stuff... but it is just window dressing and like I've said, fools people into thinking something has changed, when nothing has changed, or things were just made more confusing and less accessable. So I continue to oppose these non-solutions. --W.marsh 13:41, 15 April 2007 (UTC)

Just came in on this discussion & have had a similar one with bainer on his talk page. I suggest a simple solution - Why don't we have a straw poll? We had one for changing the Main page & this obviously involves a lot of people. There's lots of arguments on both sides, so I don't see why we can't settle this once & for all in a legitimate & obvious solution to this argument? Thoguhts? Spawn Man 08:08, 15 April 2007 (UTC)

Oh no not again. You don't use straw polls in this situation. (Both the guideline/policy/tag of the day pages (both pro and con) explain that that's a bad idea.). I mean fine. Learn things the hard way once, but isn't this about the umpteenth time? And it hurts everyone else who already *has* learned their lesson too. :-/ So No. I'm not that much of a masochist, thank you. --Kim Bruning 13:55, 15 April 2007 (UTC)

Straw poll? Does anyone else see the irony in using a straw poll to decide whether RfA is too much like a straw poll? -Amarkov moo! 20:48, 15 April 2007 (UTC)

not a vote: hypothetical example

Just a hypothetical for people who really want RFA to not be a vote. 100 people participate in an RFA in a fantasy non-voting Wikipedia where even using the word "vote" is liable to get you a community ban. 96 people express support (of course not using the word support, but in long, well-reasoned, non-mindless comments) because the candidate has written FAs, been active on policy pages, and so forth. 4 people express opposition on the grounds that the candidate didn't use edit summaries enough. Now, as we know, some people really think edit summaries are critical to being a good Wikipedian, and if you don't use them, you aren't fit to be an admin. As it happens, the first b'crat along shares that opinion, and closes the RFA as a non-promotion, as he didn't think consensus was that the candidate would make a good administrator.

Is this really what we want? Does that kind of thing, day in an day out, really sound like a better system to you? Because I think it's all in line with people wanting the numbers to be unimportant compared to the opinions expressed, as weighed by the b'crat. --W.marsh 13:51, 15 April 2007 (UTC)

Not promoting someone becuase of low count on edit summaries is a weak argument. Are summaries important-yes? but hardly what makes a good admin. A good admin is someone who is dependable, knows policies and looks up what they don't know, and applies them fairly. The poorest argument for not promoting someone is to say "he/she" doesn't have a need for the tools"...no one needs them, but wiki needs people who have them and the more admins we have the faster our backlogs will be cleared. If an admin only uses their tools once, wiki will be better off.Rlevse 13:56, 15 April 2007 (UTC)

That's my point though... the b'crat is really just another Wikipedian, and we might totally disagree with them on some things. Certainly we've seen people oppose even the most respected Wikipedians merely over edit summaries, so there's precedent here. Is closing RFAs really something we want to let people basically just do whatever they feel like on, as long as they can talk the talk and say consensus was for them to do that? --W.marsh 13:58, 15 April 2007 (UTC)

I agree there should be firm guidelines and the closer should follow them. Personally I like the current system with the % standards. People who zoom in on only something like edit summaries just don't see what really makes a good admin. Rlevse 14:01, 15 April 2007 (UTC)

(after edit conflict.... re w.marsh)

You started out great, up to the point where you made assumptions as to how the Bureaucrat would act. Closing as not promoted doesn't have consensus here. You said it yourself already: "Is this really what we want?". Well no, so a Bureaucrat worth their salt won't do that. :-P

Rough consensus in this case is to promote already, so a bureaucrat should really close it as such.

... but hey, that's interesting, all the opposers are in a cluster. Perhaps we could enter into discussion with them, perhaps before the closing date of the RFA?

So... Perhaps they know something we don't, like that the editor in question is doing something really scary and hiding them with false edit summaries... in which case everyone should switch opinion to oppose.

... Or perhaps the candidate can promise to use better edit summaries in future, in which case we actually have everyone supporting, and the whole thing could even be closed unanimously! :-)

Now here's the interesting thing... haven't RFA's always been conducted like this?. What's the discretion zone? Why do people strike out their opinions and take a different one? And perhaps you have noticed people discussing... typically in "oppose sections" (though I've sometimes seen supporters being challenged too) ? Now why do they keep doing that? Hmmm... think about it :-)

--Kim Bruning 14:07, 15 April 2007 (UTC)

Any good idea stretched to ridiculous proportions will appear ridiculous. That being said, 'crats do have discretion, they just don't get much opportunity to use it as RfA's are more often clear cut. The recent Danny case shows they don't just vote count, but they also don't stretch this discretion to ridiculous levels. ^High_InBC^{(Need help? Ask me)} 14:12, 15 April 2007 (UTC)

I don't see how it's ridiculous... but then again I thought of the Danny RFA quite differently. To say that consensus is an objective thing is not correct... everyone thinks consensus is something slightly different. My point is that we might not really be realizing what it means to give one person so much power to just declare what consensus was, since that basically is just what their opinion on the whole RFA was most of the time. --W.marsh 14:15, 15 April 2007 (UTC)

A single person gets to judge consensus all day long here, see the AfD's for example. The fact is that even 'crats are responsible to arbcom and Jimbo. So if they truly abused their powers(such as using their opinion of the case instead of judging consensus) we have something to do about it. ^High_InBC^{(Need help? Ask me)} 14:17, 15 April 2007 (UTC)

Though to answer your question, if 100 people voted support without good reason, and one person had a good reason(not edit summaries, but more like abuse of trust), then I think in a perfect world that oppose should out weigh the supports. But I have never even once seen such a bizzare ratio of reason to foolishness. ^High_InBC^{(Need help? Ask me)} 14:14, 15 April 2007 (UTC)

Anyway, it's easy to say that b'crats are just perfectly following consensus as long as you agree with them (or assume they'll act in the way you agree with, i.e. of course they won't care about edit summary usage because I think that would be a bad idea). But when you're one of the disenfranchised whose opinion doesn't count because the b'crat doesn't agree with you? It's a lot harder to agree with the claim of consensus. Ultimately, the more people involved with an AFD, RFA, whatever, the more your opinion of what consensus was is just your opinion in general. --W.marsh 14:26, 15 April 2007 (UTC)

If you feel disenfranchised at all, something may have gone wrong. Is it indeed the case that you feel disenfrnachised in some certain situation? --Kim Bruning 14:31, 15 April 2007 (UTC)

Anyone's going to feel disenfranchised if they know they could express their opinion and it won't count if the b'crat doesn't agree. --W.marsh 14:34, 15 April 2007 (UTC)

(ec)People can feel disenfranchised even when everything goes right. Put enough people together and someone will feel disenfranchised no matter what happens. The point is to get a good admin, not to make people feel good about how effective their vote was. ^High_InBC^{(Need help? Ask me)} 14:36, 15 April 2007 (UTC)

And some people think a good admin can't be one who doesn't use enough edit summaries. I guess this is turning out to be about as productive as any discussion on this page... now I remember why I took this off my watchlist. Ah well. --W.marsh 14:38, 15 April 2007 (UTC)

This hypothetical falsely assumes that there is no ground between a pure vote and a pure do-I-like-this-argument-or-not situation. --bainer (talk) 14:54, 15 April 2007 (UTC)

Now, to go from the hypothetical situation to the real world (so off-topic to this discussion), some people do get pissed off if the candidate does not use edit summaries (I do!). However, I have not seen people continuing to oppose after the candidate promised to use them in the future and changed his preferences to that effect. Oleg Alexandrov (talk) 15:22, 15 April 2007 (UTC)

Per HighInBC, "such a bizzare ratio of reason to foolishness" is unlikely to occur, so the hypothetical situation is a bit suspect. Nevertheless, if a bureaucrat genuinely believed that there were compelling reasons not to promote despite apparent numbers to support a promotion, a number of outcomes are possible. Already listed have been:

• 1.) Go with the votes. The bureaucrat should count votes, and a clear supermajority has spoken. Promotion must occur.
• 2.) Go with your reasoning. The bureaucrat was chosen by the community to exercise judgement and determine consensus. Promotion denied.

Neglected, however, have been a number of 'middle ground' options. The following two have already been used on occasion; I expect that they will appear from time to time when a 'crat is uncomfortable imposing his sole judgement on an RfA.

• 3.) Extend the discussion period. If a critical argument has come to light near the end of the normal discussion period, the discussion can be extended by a day or two to allow the participants to weight the new evidence.
• 4.) Open a discussion with the other bureaucrats (on- or off-wiki). Establish a consensus among bureaucrats before making a decision on promotion.

In either case, it is no longer a single bureaucrat making the call. Both allow for the community to see and understand the reasoning at work, and 'crats are supposed to be among the community's most trusted members. TenOfAllTrades(talk) 17:20, 16 April 2007 (UTC)

Reformat fails to go far enough

1 boldface, 17 Comment, 1 Endorsed, 1 Fnord, 1 Further comment, 9 Neutral, 38 Oppose, 1 Oppose Neutral, 1 Opposse, 1 Quick return, 1 Reply, 1 Strong Oppose, 2 Strong Support, 30 Support, and That took less than a minute. I think this comment speaks for itself. --Random832 03:12, 16 April 2007 (UTC)

• I endorse the fnord. >Radiant< 09:02, 16 April 2007 (UTC)

You ain't seen nothin' yet :) --Durin 12:50, 16 April 2007 (UTC)

How about the contents of the comments/opinions, how many of them actually speak of the candidate's suitability for adminship? :-)

How many comments are answered/challenged but remain unreplied/undefended?

--Kim Bruning 13:56, 16 April 2007 (UTC) Hmm, actually allowing a challenge to negate a comment if unanswerd might be an interesting rule for future rfas, and also might reduce (allegations of) bureaucrat bias...

Oh, and most importantly: what lessons can Moralis learn for future RFA's? I think that a 2nd RFA in a month or two will pass, right? What should Moralis be doing before then? This might be a useful question to ask on future requests for adminship too. --Kim Bruning 14:25, 16 April 2007 (UTC)

One of the good things that has come out of the reformat is that the support !votes (and I feel able to use the exclamation mark when referring to RfA for the first time since I've joined Wikipedia) suddenly seem more reasoned on average than before (before, many support votes gave effectively no reasoning; now, the support !votes often say something useful). Even if the negative consequences from this one-time (so far) change mean that it isn't eventually adopted, this is something that should be encouraged somehow. (Strangely, opposing votes - in the existing system - almost always used to give their reasoning, and ones that didn't were objected to.) One of my biggest complaints about the current RfA/RfB system is that it effectively compares the size of a pileon (the support section) to a random number (the oppose section in an RfB, and to a lesser extent an RfA), rather than looking at the arguments; this reformat neatly avoids that trap (although it may have other problems, as indicated by the discussion above). --ais523 15:56, 16 April 2007 (UTC)

Funny analogy

I ran across this article today and couldn't help thinking it was an apt analogy to the current state of RfA. The old 10,000 monkeys bit doesn't work as we're not writing anything here. Just a bunch of furry, cute mammals running around dropping pellets everywhere :) --Durin 14:19, 16 April 2007 (UTC)

Do Behave! :-P --Kim Bruning 14:27, 16 April 2007 (UTC)

I still believe that people are acting more intelligently than that. ^High_InBC^{(Need help? Ask me)} 14:31, 16 April 2007 (UTC)

I guess so --Kim Bruning 15:45, 16 April 2007 (UTC)

Oh boy. The RFA talk has turned into a joke room of sorts. bibliomaniac15 00:13, 17 April 2007 (UTC)

Would these current admins still be accepted today?

We all know that RFA standards are improving all the time, so our admins are just getting better and better. Right? So, by edit count, would you support these users?

John Doe #1

Username	John Doe #1
Total edits	2823
User groups	<hidden>
Image uploads	6 (3 cur, 3 old) (browse)
Distinct pages edited	1098
Edits/page (avg)	2.57
Avg edits/day	1.28
Deleted edits	103
First edit	2001/03/27 20:47:31
Edits by namespace 	Namespace	Edits
(Main)	585
Talk	420
User	168
User talk	1185
Wikipedia	283
Wikipedia talk	114
Image	13
Image talk	3
MediaWiki	12
MediaWiki talk	7
Template	8
Template talk	2
Category	4
Category talk	19

Comments

Comment here.

• Comment - Well, is this a trick question? I ask because I note the 250 deletions, 52 Blocks, etc. I'm guessing that this person is (or was?) a current admin? - jc37 15:28, 16 April 2007 (UTC)
  • Yes, these are current admins, and I'm wondering if we'd still support them today? Let me remove those sections, to reduce confusion. --Kim Bruning 15:35, 16 April 2007 (UTC)
• Oppose, lack of experience with images. >Radiant< 16:38, 16 April 2007 (UTC)
• Refuse comment per my comment for #3. -- Black Falcon 16:54, 16 April 2007 (UTC)
• It's not hard to tell who this is - Yeah, I can't !vote because it's pretty clear who this is anyway. This probably isn't such a good example, due to the fact that <cough> <cough> <cough>. The ones below are better. // Sean William (PTO) 02:23, 17 April 2007 (UTC)
• Oppose Well, I might get bit for saying this, but I think that this user has too few Wikipedia edits for my liking. Captain panda 02:40, 17 April 2007 (UTC)

Jane Doe#2

Username	Jane Doe #2
Total edits	7460
User groups	(hidden)
Image uploads	85 (78 cur, 7 old) (browse)
Distinct pages edited	2071
Edits/page (avg)	3.60
Avg edits/day	4.19
Deleted edits	202
First edit	2002/06/02 17:22:30
Edits by namespace 	Namespace	Edits
(Main)	2451
Talk	1001
User	901
User talk	1703
Wikipedia	858
Wikipedia talk	329
Image	166
Image talk	2
MediaWiki	25
MediaWiki talk	11
Template	9
Template talk	1
Category	1
Category talk	1
Portal talk	1

Comments

• Comment here
  • Appears generally qualified, but not enough portal talk edits. Newyorkbrad 15:38, 16 April 2007 (UTC)
    • Not nearly enough Image talk or Category edits to work well as an admin, but I'm sure that's not important ;) Majorly (hot!) 15:47, 16 April 2007 (UTC)
      • I don't believe I had any of either one of those either. Amazing that I squeaked through. Newyorkbrad 15:53, 16 April 2007 (UTC)
• Oppose, not enough featured articles. >Radiant< 16:38, 16 April 2007 (UTC)
• Refuse comment per my comment for #3. -- Black Falcon 16:54, 16 April 2007 (UTC)
• Absolutely oppose, this one is Karmafist, who we all know went on to become a terrible admin. --Cyde Weys 02:01, 17 April 2007 (UTC)
• Support I had better tell the truth here: if Cyde Weys had not mentioned who this user is, I would have supported without question. (And yes, I get the point this is trying to make) Captain panda 02:43, 17 April 2007 (UTC)

Jane Doe #3

Username	Jane Doe #3
Total edits	44266
User groups	(hidden)
Image uploads	42 (32 cur, 10 old) (browse)
Distinct pages edited	23103
Edits/page (avg)	1.92
Avg edits/day	29.20
Deleted edits	2312
First edit	2003/02/20 20:21:23
Edits by namespace 	Namespace	Edits
(Main)	19026
Talk	1731
User	1634
User talk	6321
Wikipedia	11170
Wikipedia talk	2395
Image	549
Image talk	20
MediaWiki	282
MediaWiki talk	134
Template	803
Template talk	111
Help	14
Help talk	2
Category	56
Category talk	12
Portal	6

Comments

• Add comment here
• Storng oppsoe has bda speling and note nouhg edit smumaries. >Radiant< 16:38, 16 April 2007 (UTC)
  • Did you also notice ... 0 portal talk edits! That's just shameful. This person is clearly not suited for admniship. -- Black Falcon 18:24, 16 April 2007 (UTC)
• Refuse comment. Insufficient information is provided for me to make a decision. Firstly, there are no responses to questions; secondly, I have no idea what their actual contributions are (just a summary of them); and thirdly, there is no information about the distribution of edits. Maybe all 2800 edits were accumulated in one month? Maybe they're all vandalism? If they are, what does it say about Wikipedia that a person can accumulate 45000 vandalism edits and still not be blocked? ;) As I said, a breakdown of edits by namespace is simply not enough (for me) to come to a decision. -- Black Falcon 16:54, 16 April 2007 (UTC)
  • Good points. Is there an edit counting tool that can provide such a distribution to your satisfaction? --Kim Bruning 18:02, 16 April 2007 (UTC)
    • I never pay any attention to the count that gets posted on the talk page. I look through the user's actual contribs myself. It's time consuming, so I don't participate in very many RfAs, but when I do I've done my homework. Kafziel ^Talk 18:12, 16 April 2007 (UTC)
      • Interiot's "wannabe kate" counter provides a distribution by months (and also other information like most edited pages). Of course, a more advanced summary of edits is still not enough to make a decision. The distribution of edits is not a factor I use in judging RfAs, but rather serves as a rough indicator of whether closer inspection may be needed (e.g., if all edits are made within the last week). Finding out that an editor has made 1000 edits a month for 20 months is not of much use without having a general idea of the content of those edits (e.g., maybe half of those edits are insults to vandals or the like). -- Black Falcon 18:24, 16 April 2007 (UTC)
        • Sure. These questions are mostly for folks who purely use edit counts as their criteria. --Kim Bruning 18:33, 16 April 2007 (UTC)
          • I'd say anyone who solely uses edit counts probably isn't involved enough in RfA to bother with reading this talk page. Kafziel ^Talk 19:00, 16 April 2007 (UTC)

We get that people have edit counts. Already. We get that they contain useful information only when viewed in a wider context. Already. Splash - tk 00:30, 17 April 2007 (UTC)

• Support Unless, this user is some banned user, I would support due to obvious experience. Captain panda 02:46, 17 April 2007 (UTC)

  Heh, she's definitely not banned. // Sean William (PTO) 03:03, 17 April 2007 (UTC)

  So ... would someone please share the identities of #1 and #3? I've already run 30 or so usernames through the edit counter and ... well, I'm hoping we can avoid any feline casualties. -- Black Falcon 03:57, 17 April 2007 (UTC)

  GodKing, Karmafist, Angela.

John Doe #4

Username John Doe #4
Total edits 8125
Image uploads 13 (13 cur, 0 old)
Distinct pages edited 5043
Edits/page (avg) 1.61
Deleted edits 302 (browse)
First edit 2004-10-08 01:50:15
Edits by namespace Namespace Edits
Articles 3943
Talk 559
User 302
User talk 1070
Project 1714
Project talk 207
Image 31
Image talk 3
MediaWiki 7
MediaWiki talk 14
Template 152
Template talk 36
Category 39
Category talk 2
Portal 46

Comments

• Who cares? These random edit counts are a pointless exercise, unless you really want to say that edit count is the only factor that should determine adminship. Titoxd^{(?!? - cool stuff)} 04:32, 17 April 2007 (UTC)

Need more then edit counts

I'm not sure what has started this, but I've never used a number of edit-counts only to determine my !vote on RfA. While they are useful to determine what areas somone is active (or inactive in) more information can usually be garnished from edit summaries! — xaosflux ^Talk 02:58, 17 April 2007 (UTC)

Alot of people do so, or are perceived to do so at least, in ways that can seem capricious. One user is currently opposing a candidate partly for a lack of edits to WikiProjects, for example. --bainer (talk) 04:30, 17 April 2007 (UTC)

• Frankly I find this exercise a bit silly since I never look at those edit counts on RFA to begin with. >Radiant< 08:04, 17 April 2007 (UTC)
  • Not only that, but some of our more infamous administrators are being cherry-picked for their high edit counts (or Jimbo with his edit count) to trick us into "supporting" them (or opposing Jimbo.) This is a useless exercise trying to make a point, and failing miserably (to me at least.) Edit count really isn't important to everyone, and some of our problem administrators had very clean records leading up to their RfA. Grandmasterka 10:39, 18 April 2007 (UTC)

Methods of finding consensus

I was reading our consensus decision-making article today and found some interesting stuff in the tools for consensus section. While of course it is not possible for us to hold up cards and wiggle our fingers during an RFA discussion, there are some ideas that parallel our current processes:

• Tools are used to speed up and ease the process.
• It is important to be able to point out breaches of procedure.
• It is important to have the ability to clarify points or answer questions.
• When using color cards and hand signals, there is a simple way to denote consent/reservations/objections.

Is it really possible to make RFA (and other processes) any more like consensus decision-making processes? I frequently see people say "consensus is not a vote", but what we're doing now seems to be the textbook definition of consensus. If our processes don't work, perhaps consensus is not what we really want? --- RockMFR 15:32, 16 April 2007 (UTC)

It takes time to explain how our consensus finding systems work, and people keep coming and going. Some folks get frustrated at having to explain each time... but at the same time, our project namespace is being camped by people who (as a self-defeating aggregate) don't want to see accurate descriptions of our processes. It's quite frustrating. :-/ --Kim Bruning 15:38, 16 April 2007 (UTC)

It bothers me that people talk about "finding" or "determining" consensus when a better phrase would "forming", "developing" or "seeking" a consensus. This is not just being picky about words. Consensus is a process as well as a state. It's not nearly as important what the current state of opinion is as what the final state of opinion is.

When the majority of people want something but there are a significant number of holdouts, the question is whether the holdouts will give in to the majority or the majority will decide to respect the opinions of the holdouts. The decision of which of these two options will prevail is determined by the process of forming a consensus and that can only be done via discussion and persuasion. This is why voting is evil but surveying to determine consensus in advance of and after discussion and persuasion is not. !voting is OK as long as the process doesn't fixate on the precise numerical outcome while ignoring the consensus formation process. You can see consensus formation at work in RFAs where people change their !votes based upon the discussion.

--Richard 17:18, 16 April 2007 (UTC)

I think it would be helpful if we all dropped this silly jargon term "!vote". There is a very good English word to use in its place: opinion. --Tony Sidaway 17:29, 16 April 2007 (UTC)

I second that! Furthermore, in a lot of contexts, people writing "!vote" actually mean "vote". Just say it, for crying out loud - nobody is going to block you. :) Kafziel ^Talk 17:37, 16 April 2007 (UTC)

I try to carefully keep a distinction (for instance, see my comments a few sections above). A 'vote' is something that's merely counted and used to come to a decision by seeing whether there's more of one type of vote than another type of vote (something which should probably be avoided on Wikipedia, but which is seen all too often on sectioned and tallied RfAs). A '!vote' is something that superficially looks similar, and generally has a bolded statement at the start, but whose content ought to be read when coming to a decision, and the arguments it's making considered rather than just its existence. 'Opinion' has a similar connotation (and is a better word than 'vote' for AfD, for instance), but carries with it less of a sense of the importance of the point the !vote is making. Of course, this distinction is pretty useless if people don't know what '!vote' means, and the fact that it isn't a real word with a well-known meaning may be reason enough to deprecate it. --ais523 17:46, 16 April 2007 (UTC)

I always thought "comment" was much better. Kafziel ^Talk 17:47, 16 April 2007 (UTC)

I agree. "!Vote" can be more clearly expressed as "argument" or "comment". In AfD discussions, people should not "vote" or "!vote", but rather argue for an article's deletion or retention. If "argue" brings up negative connotations of an argument between argumentative editors arguing various arguments, then "comment" works just as well (although I usually reserve "comment" for neutral posts, where I make no recommendation). RfA, I think, should be part-comment, part-vote. A comment is useful only as long as someone can judge its validity; my personal opinion is that bureaucrats should not judge the validity of RfA comments (except for obvioius vandalism, of course). If a bureaucrat finds certain arguments to support or oppose a candidacy unconvincing, he or she should not wait until the RfA is over and then discount those vote-comments. Instead, he or she may participate in the RfA and, if desired, try to convince people to change their positions. -- Black Falcon 18:13, 16 April 2007 (UTC)

Consensus forming it is then :-) --Kim Bruning 18:01, 16 April 2007 (UTC)

Wikipedia:Requests for adminship/Matt Britt

Like the Moralis RfA, this RfA has been refactored. However, in this case it has been refactored to be more like an RfC. The refactoring was done with the knowledge and consent of User:Matt Britt. The intention here is to break away from the obsession with vote counting and focus on discussion and relative merits of the nominee. Please keep meta discussions about the format of that RfA here, rather than on the RfA itself. Thank you. --Durin 22:00, 16 April 2007 (UTC)

Interesting experiment. I believe this one will turn out to be ok but I'm not sure we've got the right guinea pig. I have a hard time believing that this format for a more controversial candidate would not rapidly turn into an incoherent mess whose interpretation will be difficult. B'crats are supposed to evaluate consensus and it would be nice to have a format where we have some sort of confidence that different b'crats won't make different decisions. Pascal.Tesson 22:09, 16 April 2007 (UTC)

• Finding a guinea pig for this experiment is harder. It's a very radical change, far more so than Moralis' RfA. If another user wants to be an additional guinea pig for this format, who might be more controversial, feel free to ask :) --Durin 22:22, 16 April 2007 (UTC)

This is a bad, bad concept, an application for a North Korea travel visa would be simpler than this. Wikipedia is not a bureaucracy. I can't see any advantages in this model, and I can't figure out how is a bureaucrat going to determine any community consensus with so many parameteres.--Húsönd 22:19, 16 April 2007 (UTC)

This is a good notion, but I think the format is entirely too byzantine. A Train^talk 22:23, 16 April 2007 (UTC)

• Readily granted. I admit having a bit of a time getting my head around this sort of format, and coming up with a reasonable formatted RfA in this form. It can probably definitely :) be evolved. --Durin 22:34, 16 April 2007 (UTC)

If I believe that the candidate has all five or six of the favorable qualities listed, what would have been a "support" with a comment becomes five or six separate entries. Multiply that by the number of supports, and then similarly with the number of opposes, and one envisions the page quickly becoming unwieldy. Newyorkbrad 22:25, 16 April 2007 (UTC)

• Not unlike some RfCs :) --Durin 22:27, 16 April 2007 (UTC)

• On those, it's sufficient to endorse the "Outside view by Newyorkbrad" and then move on. :) Newyorkbrad 22:29, 16 April 2007 (UTC)

Newyorkbrad has hit this nail on the head. The structure of the RfA requires discursive answers from contributors over several measured dimensions. What are the advantages and the practicalities of this structure over the current model - I know about the vote-counting and the pile-on issues. This model is obfuscatory. I might support in three areas and oppose in two areas or any combination of the same, as well as the 60+ other regular contributors to the RfA articles. 60x5=300 comments to read through and balance out. The result would have to be issued as a Supreme Court ruling, showing the Bureaucrat's working out of the answer as evidence is weighed up. I'm not convinced that this is the best way to establish consensus and measure the opinion of the contributors on the evidence in-hand. (aeropagitica) 22:43, 16 April 2007 (UTC)

I like this format quite a bit better than the one chosen for Moralis - that one had the advantage of making it less "vote-y", but also made it very difficult to follow the discussion. I'll be interested to follow the development of this RFA. JavaTenor 22:31, 16 April 2007 (UTC)

The only word that comes to mind is "idiotic". No offense to whomever decided upon this format for an experiment (as I have no idea which editor(s) decided upon it, and I would not hold it against him/her/them, anyway), but this is the worst possible way to run an RfA. It's one thing for an RfC, when there are only a dozen people expected to contribute to the outcome, but it's another for an RfA, where there could be hundreds of opinions. It also doesn't leave wiggle room- with an RfC/arbitration/whatever, there are findings of fact; with an RfA, everything is opinion. Everybody has a different answer to "Is this person experienced?", "How much do I trust this person?", "Does this person need the admin tools?", "Will this person abuse his/her newly-found powers?", etc. An RfA should never be about agreeing or disagreeing with pre-written statements. It's about looking at a candidate's background and arguing for why or why not someone should be an administrator. There is no reason to make the page this long and unwieldy, and there's no reason to expect half a dozen comments from a single user when one would be sufficient. I don't expect Matt's RfA to get much action, and that's a shame, but I know that I, for one, am certainly discouraged from contributing under this format. -- Kicking222 22:35, 16 April 2007 (UTC)

• I'm the blithering idiot. For proof, see the upper right of my userpage. :) --Durin 22:38, 16 April 2007 (UTC)

This is another interesting idea, thanks Durin for these experiments. This one may provide a challenge to the closing bureaucrat as the head-counting heuristic will be difficult to apply, but it should also be more useful in exposing widely-held concerns about the candidate, if they exist. Christopher Parham (talk) 22:44, 16 April 2007 (UTC)

Good to see the discussion is already in full swing. :) I was coming by just to say this version needs another category: "Nomination is the craziest dang thing I ever saw and I hardly know where to begin." Users who support this view: Kafziel ^Talk 22:46, 16 April 2007 (UTC)

The idea is fascinating, and good luck to the candidate, but I really feel that in some cases it is going to be totally impossible for the closing bureaucrat to determine consensus. If this format is adopted we are going to fail to promote an unknown number of potentially good candidates. I assume that the policy of >80% promote, <70% no (except User:Ryulong), between the two is bureaucrat discretion is going to go, because when opinions are given seperately on seperate parameters this percentage calculation will become wholly meaningless. For the record, in the light of recent applications, I would suggest that if this same format were to be applied to bureaucrat selection we would never, ever select any. And if it were not to be applied across the board, would that not be discriminatory? Tell me I'm wrong - I don't mind.--Anthony.bradbury 23:03, 16 April 2007 (UTC)

Mmm, well I've had a good go at using all the features of the new format. To summarise my experience:

1. Best bit - I like the way the discussion focuses contributors on the various skills of the candidate, it feels like a more in depth process
2. Worse bit - amount of time it took me to understand the template enough to create my own section for the comment I wanted to make
3. Concerns - judging the consensus may be difficult. But more importantly I think there's a risk that we can get too focused on a candidates lack of experience in one are- it took me a while to consider the candidate as an overall package rather than as a collection of skills that were present or absent.

I'm not sure its enough of an improvement on the present model to justify losing the convenient elements and it is horendously complicated. That said its an interesting experiment. Perhaps we should have another go at making the first X days of RfA discussion only and then the !vote happen for the rest? WjBscribe 23:09, 16 April 2007 (UTC)

The first time I read the idea of formatting RfAs as RfCs it felt like a good idea but once you actually see it, it's hard to really like much about it. As WJBscribe notes, this will make the candidate a collection of skills and non-skills. As I said earlier, the guinea pig is pretty solid so we might not get to see how disastrous this can become but WJBscribe's new section on "no XfD experience but not a problem" is (definitely not taking a jab at WJB here) a mild example of perverse effects of the format. People who feel that the summary of others is not quite right will keep adding sections with slightly tweaked formulations and there won't be much to conclude in the end. Also I'm afraid we'll arrive at impossible situations where pretty much everyone agrees that the editor needs the tools, does great work but, say, is uncivil. Unless we have two sections that are "the positives outweigh the negatives" and vice versa, it will be the b'crats job to assess that balance. And if we do have these sections, well, we're just back at voting. Pascal.Tesson 23:18, 16 April 2007 (UTC)

Exactly. Why do you think that large-scale surveys are of the 'Choose to agree or disagree with A,B,C,D' and not 'Give us your opinions about A,B,C,D'. The former is easy to analyse, the latter nigh-on impossible. This might just work with a dozen-or-so contributors but with the numbers that regularly participate the discursive nature of the RfA will not serve its subjects in the long run. (aeropagitica) 23:29, 16 April 2007 (UTC)

• I'm all for modifying and experimenting with RFA, but this format is just awful. RFA's fundamental issue is whether a user has the community's trust or not, and this format is not conducive to gathering consensus for that. It is useful for gathering evidence for an eventual lynching, but its layout is in many places liable for self-contradictions. I can agree with a user, for example, about the nominee not having enough experience in the Portal talk: namespace; that doesn't mean that I agree it is a substantive issue. If, as Pascal brings up above, we are going to end up with sections like "No XfD experience but not a problem", "No XfD experience and it is a problem", "No XfD experience but there are more relevant issues" and "No XfD experience but nobody cares", then this experiment will fail to scale and will throw no usable results. Titoxd^{(?!? - cool stuff)} 23:37, 16 April 2007 (UTC)

The results of this format.

Durin, I beg you, please stop these experiments before you accidentally invent a virus that wipes out the human species. I commend you on not taking the criticism by Kicking222 personally, but also intend to add to that criticism. As I see it, the only effect of this format change was to make turn RfA from one clear yes/no semi-vote into half a dozen overlapping, convoluted votes. Also, by discouraging many people from participating, I expect it will reduce transparency.

That's the practical side of the issue. I am also morally opposed to this format (for this, see my comments here). This format not only increases bureaucrat discretion, but does so in a way that accountability becomes impossible. What would we do with an RfA with this format that has 150 participants and 40 overlapping subsections? -- Black Falcon 23:45, 16 April 2007 (UTC)

We'll need to institute a whole other bureaucratic process to try to figure out what the bureaucrats were thinking, like a wise man once said! Kafziel ^Talk 00:05, 17 April 2007 (UTC)

• I'm not quite done with the virus that wipes out humanity. I'll let you know right before I release it. You might have time for a quick lunch at the local pub, and maybe hitch a ride with a vogon.
• Ok jokes aside...come on, it's just an experiment. There's no damage to the project being caused by this. For far too long, RfA has wallowed in a complete inability to come to any agreement on whether RfA is broken or not, whether we should reform it or not, what that reform would entail if anything, and etc..etc..etc.. The amount of discourse on these subjects could fill a small library. It's really rather absurd.
• So, I got tired of hashing these endless debates out. Instead, I decided to DO something about it and actually try doing something different for a change. Unless someone can show me how these experiments constitute some threat to the project (especially a threat that's worse than the normal, already damaging RfA formats), then I intend on trying others if I can find willing guinea pigs.
• In the very least, you have to acknowledge these experiments are fostering a considerable amount of discussion on actual attempts at reform rather than theoretical notions of how something might work. --Durin 00:59, 17 April 2007 (UTC)
  • After you patent the virus, will you at least donate some of the proceeds to humanitarian causes? ;) Yes, I'll agree that there's no damage caused by these experiments ... as long as they are limited to just a few instances. And I do have to admit that your bold actions have stirred debate considerable debate. And you are right that the existence of almost 90 RfA talk page archives, many about "RfA reform", is ridiculous. However, there is a simple solution ... something about "if it ain't broke" ... but I digress.
  • Here's the thing: I don't see that the current RfA formats as damaging and I don't view RfA to be broken (also, I really do find this format to be exceptionally terrible ... no offense). Of course RfA isn't perfect, but then, what is? Yes, some candidacies fail for reasons you and I consider trivial (e.g., too "few" talk space edits). So what? We can't always get our way! Just because we consider those reasons to be trivial does not mean we should try to impose our standards on everyone else.
  • On the whole, about half of RfAs succeed. RfA usually sees 10-15 nominations per week, of which 5-8 succeed. If you feel that we need more admins, just nominate more people. Instead of 10 nominations a week, let's have 20! -- Black Falcon 03:15, 17 April 2007 (UTC)
    • Please believe me when I say that this is meant with no arrogance; I, more so than probably anybody here at RfA, know about the history of RfA. 50% promotion rate is absolutely absurd, given the history here. I'm not going to argue, yet again, whether RfA is broken or not, whether we need more admins or not, etc..etc..etc.. ad nauseam. If you truly want my opinion on this subject, along with the opinions of literally hundreds of other contributors to RfA, read the archives. I'm not saying this to be hostile. I'm pointing out that this is very old ground. This is precisely one of the reasons why I embarked on these experiments. --Durin 13:05, 17 April 2007 (UTC)
      • Not to worry ... your statement is neither hostile nor arrogant. I too am not particularly inclined to get into a lengthy debate on the subject. I have one question, though: when you wrote that "50% promotion rate is absolutely absurd", did you mean that my estimate is inaccurate or that it is accurate but that the rate of promotion is too low? Thanks, Black Falcon 16:29, 17 April 2007 (UTC)

I think this format is a huge improvement over the previous experiment. Problem is of course that it is still a vote, and the decision can be very difficult to reach if some of the views pass and others not. Another (bigger) problem is that anyone who finds a new ground on which to oppose has to write a new view. That means that constant vigil is required from people who want to express their opinions, because after they have done so, a new view could have been added for which they feel the need to endorse or not to endorse. Furthermore, I worry about the quality of the views. In this RfA, a couple of views have been worded badly. E.g, User has no experience with XfD: if I don't endorse this view, does this mean that I think the user has enough experience with XfD? Furthermore, if I agree with some, but not all the points in a view, would I have to say I don't endorse, and write a view of my own? It all becomes very fragmented this way. Errabee 10:07, 17 April 2007 (UTC)

Harakiri

Oh great! Instead of just one vote, we now have several and, indeed an unlimited number. Discussion may occur within each voting section, but that's blatantly the case with current, sensibly-arranged RFAs. So would both i)the nominator and ii) the nominee please explain to me how this is better than before?

Once they've done that, I'd like them both to tell me how they think that presenting a series of leading questions for people to vote on (oh hush, they're discussions like its snowing in Africa) is in some way going to demonstrate consensus. Would they have people i) oppose the leading questions, ii) write opposing leading question or iii) do both? If not i) alone, then ii) is just comedic since people will oppose and support with opposite meanings both of them.

If you're going to wail at the altar of consensus, and whip yourselves on the back with a nine-tailed whip called "RfA is broken", please try not to put entertaining make-up on before you do so. (Copied here because someone invented some rulecreep where I'm not allowed to discuss this on the discussion page of the RfA. Heaven forbid we might actually have a negative statement anywhere near it). Splash - tk 23:50, 16 April 2007 (UTC)

Abort?

In all fairness to the candidate, I believe we should abort the experiment before it blows out of control. Now I understand some will disagree and say it's too early but I'm not sure this is very helpful for Matt Britt. Pascal.Tesson 00:07, 17 April 2007 (UTC)

I second the motion. It's turning into a foundering morass in here, making it very tedious to look at every view and endorsement, not only for an evaluator but for a bureaucrat. bibliomaniac15 00:11, 17 April 2007 (UTC)

The amount of text in the nomination at the moment is much less than any contentious RFA in the normal format. What about this format makes reading less text so tedious? Christopher Parham (talk) 00:16, 17 April 2007 (UTC)

You realise it's been running for all of about 2 hours, right, when you make that calculation? Splash - tk 00:27, 17 April 2007 (UTC)

Yes, and comparing to other RfAs at the two hour mark, and the difference is not significant. Moreover, as time goes on the fact that there is less need to repeat similar reasons/conversations over and over may allow this format to save space; we don't really know. Christopher Parham (talk) 02:26, 17 April 2007 (UTC)

(edit conflict) I'll pile on as being this seriously opposed to this format. This is far worse than the Moralis format, which is itself far inferior to the standard format. What is completely missing from the format are the sub-sections for "I think this view is reason to promote" "I think this view is reason to not promote" and "I think this view should be completely ignored because the issue isn't relevant". And of course, since these are almost independent of whether the view is agreed with or not, everyone gets to vote again in this sub-section of each view. This doubles the needed effort level from the horror stories described above. GRBerry 00:59, 17 April 2007 (UTC)

Agreed with GRBerry that more nuance than "support" and "oppose" is needed to avoid an excess number of views. JavaTenor 01:00, 17 April 2007 (UTC)

green card --- RockMFR 01:54, 17 April 2007 (UTC)

• Oppose aborting, unless Matt Britt wants to. You don't abort an experiment just because it seems to be going badly. We're just a few hours into it. Let it run. What's to lose? --Durin 00:54, 17 April 2007 (UTC)

  I agree with Durin. Although I don't think format is likely to be the solution to our problems, we are discovering a lot of reasons why it doesn't work so next time someone says "Why not have RfC style RfAs?" we'll have something to point to rather than just discussing it in the abstract. And good ideas may yet come of it. Unless Matt Britt wants to end it/change the format I don't see the harm in letting it continue. WjBscribe 01:03, 17 April 2007 (UTC)

Ok, I guess it is Matt's call although he should be made aware (if he isn't) that this is definitely an option. Sure it's nice to have a "see, it doesn't work" example but this is his RfA and he should be able to make that decision. Pascal.Tesson 01:37, 17 April 2007 (UTC)

Agreed. I favour aborting this particular experiment, but think deferring to the candidate is appropriate. I also don't want dislike of the format to be mixed with sentiments about the nominee, who is an excellent editor. -- Black Falcon 03:19, 17 April 2007 (UTC)

• Please abort this. It's a nice idea but the result is a complete mess. >Radiant< 12:07, 17 April 2007 (UTC)
  • There isn't any reason to abort. The nominee has not asked for it to be aborted, and it's not causing harm to the project. It's not as if people are being put to a torture rack and made to edit the RfA. Nobody's time is being wasted against their will. Contrast; the experiment is producing helpful feedback.What reason is there to abort? --Durin 13:09, 17 April 2007 (UTC)
• Abort. This is ridiculously close to WP:POINT. New format wasn't discussed enough to implement it in real life. Now result is a complete mess. Stop it before it's too late, because it's unclear if bureaucrats should promote after a RFA which format has no community consensus. MaxSem 17:48, 17 April 2007 (UTC)
  • Format should have no bearing on whether or not there is consensus to promote, unless you're arguing that the format is more important than the opinions presented therein. It may make it difficult to evaluate consensus, but that's not the same thing. -- nae'blis 17:58, 17 April 2007 (UTC)

Enough with the voting already. --Tony Sidaway 17:58, 17 April 2007 (UTC)

Arbitrary section break

• I oppose the reformatting of this RFA. I'm willing to let it continue for one simple reason: Matt Britt is a highly qualified candidate, well beyond the average, and he's likely to pass regardless of the format (unless he climbs the Reichstag). My main opposition is per Newyorkbrad. The standard RFA format is extremely user-friendly. You type in your vote and your reasoning, and that's it. You know exactly where all the comments belong. In this RFA, everything is confusing. I had to slog through endless lines of "noinclude" and "includeonly" and fully spelled-out diff links until I could find the section for voting on proposition one, then again for proposition two, and after that I gave up. More than any considerations of bureaucrat discretion, voting is evil, etc., an RFA must be user-friendly. I oppose the Moralis format for other reasons, but at least we can agree that the Moralis RFA is user-friendly.

Durin, I congratulate you on you boldness and good intentions. I hope you can find a more user-friendly way. For example, what if there were sections for "support" and "oppose" for people who just want to leave a single general comment, and a "discussion" section where the specific issues can be hashed out? It's probably worse than the current system, but at least it's reasonably user-friendly. YechielMan 02:17, 17 April 2007 (UTC)

• If we did that, we might as well have the current RfA format. The idea in this format is to get away from voting, not support it.

At least we now know what course of reform we should not take. And, of course, we shall thank Durin for pointing this part out, whether it was intentional or not, and I mean it with all sincerity. —²¹⁰physicq (c) 02:37, 17 April 2007 (UTC)

• Of course its intentional; this is just one of the potential outcomes of the experiment. We do these things not because we necessarily expect them to work, but to see if and how they will fail. That's what an experiment is all about. --Durin 13:11, 17 April 2007 (UTC)

My impression from looking at this format is negative, there are altogether too many locations where we need to enter our opinion, and I don't think that is truly needed. But the experiment is well-intentioned, and are a better way of realizing possible benefits and drawbacks with the proposed system than a priori guesses at the significance of the pros and cons. Sjakkalle (Check!) 07:41, 17 April 2007 (UTC)

The following discussion is an archived debate. Please do not modify it. Subsequent comments should be made in a new section.

• As this section has been signifignatly edited and outmerged since creating, please comment in other sections. Historical information may be found in this history. — xaosflux ^Talk 02:05, 20 April 2007 (UTC)

This method of RFA is so confusing that I am unable to participate.

Editor's note: I have move this view from Matt Britt's RfA to here, as meta discussions regarding the format of the RfA are inappropriate for the RfA and are better directed here. --Durin 12:51, 17 April 2007 (UTC)

• I strongly disagree with Durin's summary. This was a list of users who were 'unable to participate' in the RFA due to the reformatting. I don't think that one should necessarily consider all of these as opposing editors, but it is a reasonable option.AKAF 15:28, 17 April 2007 (UTC)
• Please move this back. It is extremely significant that the view that this process is invalid has nearly as many endorsements as the most popular view of the candidate. The alternative for me would be to disendorse every single positive view of the candidate in that RFA. >Radiant< 14:24, 17 April 2007 (UTC)
• I agree - this should be moved back to the discussion. Metadata or not, it is significant. Kafziel ^Talk 18:00, 17 April 2007 (UTC)

• Moved back. It is very relevant to the RFA. >Radiant< 08:04, 18 April 2007 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made in a new section.

Determining consensus

Considering this is meant to be a way of making it easier to determine consensus, I'm surprised to see that there are no guidelines on how consensus will be determined. RFCs work in this format because they aren't intended to reach any kind of conclusion, they are just a way of finding out people's opinions. An RFA needs a definite result, and I see no way to derive one from this format. It's easy to determine if there is consensus on any individual view, but how are the crats meant to determine an overall result? --Tango 17:04, 17 April 2007 (UTC)

I absolutely agree with Tango. Currently we have several views that definitely have consensus:

1. Matt has demonstrated a need in sysop tools.
2. He is trustwothy. (there are de-facto two views about this, WTHeck?)
3. He sometimes misuses automatic reversion.

Now what should bureaucrat determine:

• Does #1 means that people would like to see Matt using these tools?
• Does #3 means that some people wouldn't like Matt to be an admin?
• Do #1 and #2 overweigh #3?

That means that most RFAs, except for really obvious ones will rely heavily on bureaucrat's discretion, much heavier than now. Is it what was intended? MaxSem 18:01, 17 April 2007 (UTC)

I was just going to create a new thread. In a RfC, there is a "council" that offer solutions and determine which ones should be applied. If RfA are going to have an RfC format, bureaucrats should work like in Danny's case, using a public page to discuss and reach consensus. No more decisions made by a single bureaucrat. If it looks like a RfC, formats like a RfC and ends like a RfC, it is a RfC.

As a side note, were us to adopt RfC format, people would not be able to dismiss a RfB candidacy because "we are enough." -- ReyBrujo 04:40, 18 April 2007 (UTC)

Council? -Amarkov moo! 04:46, 18 April 2007 (UTC)

I have an idea

Since it's apparently experiment week at RFA, does anyone want to run for adminship with my idea, which is an arbitration-workshop-like structure (proposed principles, findings of fact, etc) with the discussion divided into comments by bureaucrats/by candidate/by others. Anyone up for it? --Random832 02:41, 17 April 2007 (UTC)

That makes it incredibly hard to analyze support and oppose, which is what bureaucrats have to do until there is a consensus to change that. If it's just intended to be an experiment, though, with no actual effect, then I'll do it. -Amarkov moo! 02:57, 17 April 2007 (UTC)

• RfA is not a vote. RfA is not a vote. RfA is not a vote. Bureaucrats are not expected to count votes. They are expected to evaluate consensus. --Durin 13:15, 17 April 2007 (UTC)

• Perhaps RfA should be a vote, though. It is very hard to get consensus about a "yes"/"no" answer, and we should think about promoting based on numbers in non-consensus situations with 66-90% support. Kusma (talk) 13:22, 17 April 2007 (UTC)

Go ahead. But I can say that it's going to be even more convoluted than the current standard template, though. —²¹⁰physicq (c) 02:58, 17 April 2007 (UTC)

I need a victim guinea pig volunteer, though... --Random832 02:59, 17 April 2007 (UTC)

Choose a suitable one from the Category:Wikipedia administrator hopefuls list. Note the emphasis. —²¹⁰physicq (c) 03:01, 17 April 2007 (UTC)

I am considering accepting your offer, but I would like to see an example of this method of gaining adminship first. Captain panda 03:04, 17 April 2007 (UTC)

I do suppose it could limit the amount of obviously unqualified candidates with an "opinion on hearing this request" type of thing at the beginning. bibliomaniac15 03:28, 17 April 2007 (UTC)

Um, I am interested in trying out with this method. AQu01rius (User &#149; Talk) 03:33, 17 April 2007 (UTC)

shudder... that sounds even more complicated than the Matt Britt experiment. And I don't mean that as a compliment! :-) Pascal.Tesson 03:35, 17 April 2007 (UTC)

There's no harm in trying. At least we know what doesn't work so that said unworkable ideas won't come up again. —²¹⁰physicq (c) 03:37, 17 April 2007 (UTC)

Well my gut feeling is that this is a step in a direction which is not turning out to be too convincing. Pascal.Tesson 04:01, 17 April 2007 (UTC)

I also share the same sentiments, as I have indicated above, but if people want solid proof of the unworkability of this proposal, might as well show it to them via hand-on experience. —²¹⁰physicq (c) 04:29, 17 April 2007 (UTC)

RFAr is an even worse format than RFC for this purpose, unfortunately. -- nae'blis 14:05, 17 April 2007 (UTC)

I like the idea and I will volunteer to be the guinea pig although I am not sure that my RFA will pass. Or, perhaps it is because I am not sure that my RFA will pass that I would be a good guinea pig. Presumably, the "Decisions" section will consist primarily of "Recommend for adminship" or "Decline adminship" with some possible additional recommendations. --Richard 19:49, 17 April 2007 (UTC)

Withdrawn. I'm convinced that this isn't going to be as easy as i thought, based on the results of the "RFC-like" experiment. If someone else wants to do it, go ahead, but it's not worth the trouble for me. --Random832 03:27, 18 April 2007 (UTC)

Ahem...

I've seen quite a few people (including myself) voicing the opinion that the current format is fine, or at least that there's no suitable replacement at this time. How come our opinion is just being ignored? I have no hard evidence, but I'm willing to bet there are more people who support the current version than there are people who support any one alternative. So if these alternate versions don't have significant support beforehand, why all the disruption? I know eventually (once there's support for them) we'll want to give these new versions a real-world try, but couldn't they at least go through some peer review-type troubleshooting sessions before being forced on us in a real RfA? It seems to me that in the stampede for "consensus building", you guys are actually trampling consensus by throwing these other formats out there almost unilaterally. Kafziel ^Talk 03:42, 17 April 2007 (UTC)

• For three+ years, your opinions have been listened to. The opinions were not in any respect ignored. They were talked about, hashed out, lathered, rinsed, repeated, spindled, folded, hung out to dry, shat upon by passing seagulls, sent back to the laundry, re-washed with a few applications of a stain stick, rehashed, then put through a conventional dryer, put in a laundry basket and dumped in a closet somewhere. If you doubt me, have a look at the archives. Shockingly, somebody decided that...*GASP*...we've already had enough talking :) and decided to try something different for a change. Of course, Jesus was crucified for suggesting wouldn't it be great to be nice to people for a change too. Not equating myself to Jesus here, but the reluctance to change... --Durin 13:23, 17 April 2007 (UTC)

I'm the one saying "wouldn't it be great to be nice to people". If you have an idea, don't make some other guy be your guinea pig, with his own actual RfA on the line. Do it in your own space, with your username and history, without dragging anyone else into it. It's not binding (obviously), it's not disruptive, nobody is forced to participate, it can weed out the obvious bad ideas... there's no down side to it. Kafziel ^Talk 13:33, 17 April 2007 (UTC)

• Except of course less participation. This has been done Kafziel, with little result. --Durin 13:38, 17 April 2007 (UTC)

Then the problem is the advertising, not the lack of disruption. Kafziel ^Talk 13:41, 17 April 2007 (UTC)

• Except of course that advertising was done in previous attempts. Further, Moralis' and Matt Britt's RfAs are not causing any disruption to the project. Disliking an experiment, even vehemently, is not the same as disruption. Nobody else's RfA is being affected by these experiement. Nobody's. Nobody is being forced to comment against their will. --Durin 13:45, 17 April 2007 (UTC)

Okay, if I have to ask, I guess I'll ask. What attempts? Kafziel ^Talk 13:46, 17 April 2007 (UTC)

Um... these are intended to be real RfAs, with a binding result? That's stupid. Are we going to start running AfDs with weird formats and then declaring the results (which can't be interpreted well because the process is different) binding? -Amarkov moo! 03:44, 17 April 2007 (UTC)

• Absolutely. Why not? It's not as if an admin can cause irreparable harm to the project. If he is not trustworthy, it'll still show up in the RfA. What's the big deal? --Durin 13:23, 17 April 2007 (UTC)

I'm not sure why you are so keen in trying to stop experimentation. No one is disrupting anything here. I am of the opinion that the current system does not work, except all the other systems are even worse (in short, there is nothing wrong with the current system). Just as long as there are willing volunteers for this, let it go. Obstructing new ideas is only going to cause more "overthrow the current regime system" agitation. Rather, let these ideas come to reality, and when we see the abject failure of these proposals, we shall be able to prove that the current system is really the best we can get (so far). —²¹⁰physicq (c) 03:46, 17 April 2007 (UTC)

No one is disrupting anything? Have you seen the Moralis RfA? I haven't seen such a swarm of disorganized input and pissed-off editors since... well, maybe since one of my own RfAs.

I'm not saying experimentation is impossible, but it shouldn't be so wanton, either. If you come up with an idea, set up the format in a subpage of your user space with yourself as the "nominee" and give it a week to see how it goes and get feedback on the talk page. Kafziel ^Talk 03:53, 17 April 2007 (UTC)

• Interesting. Could you please point to some negative effect on the main namespace because of the Moralis RfA? A few diffs would suffice. Let's remember the goal of this project is the encyclopedia, not RfA. The project is not being harmed by this. --Durin 13:26, 17 April 2007 (UTC)

Well Moralis' RfA is pretty awful but I tend to agree that it could be a blessing in disguise. That is, of course, if the few supporters of that format are willing to recognize that it has created more problems than it has solved. Pascal.Tesson 04:04, 17 April 2007 (UTC)

But what about the next one, and the next one, and the next one?

A similar result could have been had in a test run on a user subpage. It wouldn't have gotten quite the same level of response, but we would have seen a similar pattern. If someone comes up with an alternate format, he should be willing to put his own dignity on the line for his experiment, rather than someone else's. If things seem to be working out after a week or so, then he can go looking for a willing test subject for RfA. If not, no harm done to anyone. Kafziel ^Talk 04:12, 17 April 2007 (UTC)

• I doubt that to be the case. A subpage experiment lacks the ruffled feathers and "they moved my cheese!" effect we're seeing by having these be actual RfAs. Subpages get considerably less attention. --Durin 13:28, 17 April 2007 (UTC)

One problem: historically, RFA tests on pages away from WP:RFA receive little attention, if any. Titoxd^{(?!? - cool stuff)} 04:28, 17 April 2007 (UTC)

Everyone on this talk page will know about it, and given the recent activity here, that should be a decent number of people. We can keep a list of links at the top of this page to the various test pages, so people dropping by can see where to go. Kafziel ^Talk 04:38, 17 April 2007 (UTC)

The number of people here right now is not drastically different from the number of people that watch it routinely. However, why should I (to pick an example) care to go into a test page, when I could spend my time better in other community matters? At least when a real nomination with a potential promotion is concerned, there's an incentive to comment. Again, if the part is any indication, until someone is bold and breaks the window, no one bothers to comment. Titoxd^{(?!? - cool stuff)} 04:46, 17 April 2007 (UTC)

Why should you? Well, that's kind of my point. Whether you care to go to a test page or not, consensus appears to hold that the current format is better than any other format, so these new versions shouldn't be forced on the community by one or two people. It's similar to moving an article to a different title when consensus has already held to leave it where it is. Being bold does not include being reckless. Kafziel ^Talk 12:04, 17 April 2007 (UTC)

• You do know that a recent survey had a large majority of people indicating RfA was broken, yes? Your estimation of consensus might need adjusting. --Durin 13:30, 17 April 2007 (UTC)

I think not. There was no consensus found for any particular format, let alone one that outweighed consensus for the current one. It's easy to say something is broken in a vague way, but it's another thing entirely to actually come up with a viable alternative. Kafziel ^Talk 13:45, 17 April 2007 (UTC)

There's nothing wrong with experimentation per se, but I'd like to see the list of alternative RfA structures proposed to-date, along with their pros and cons listed alongside too. Are the alternatives just Moralis' and Matt Britt's? It's brave of them to try but it's clear that it is much harder to interpret the balance of opinion in them compared to the other running RfAs following the established structure. Is this the way that a new structure is to be drawn up, by performing live experiments and rating the comments on WT:RFA? I'd like to be able to read the rationale behind each new structure as it is field tested, I think that this would be useful to see the method that the proposer is using to attempt to establish consensus for the candidate. A link to this on the RfA would be an excellent idea. (aeropagitica) 05:06, 17 April 2007 (UTC)

• The hypothetical discussions went on forever. The pros and cons were very difficult to evaluate without concreate examples. --Durin 13:35, 17 April 2007 (UTC)

1. Support. I support the current format, and see no need for change.AKAF 06:44, 17 April 2007 (UTC)

• The cure is becoming worse than the problem. Wild experimentation isn't helpful. >Radiant< 12:09, 17 April 2007 (UTC)

  But as you know experimentation is sometimes the only way to effect change. Haukur 12:28, 17 April 2007 (UTC)

PROD didn't disrupt AFD. It was tried out on a separate page, people were (and still are) free not to use it, and it didn't wreak havoc on anyone's chances at adminship. Kafziel ^Talk 13:02, 17 April 2007 (UTC)

• The experiment isn't disrupting RfA. No animals or plants, or even other RfAs were harmed in the making of this experiment. And, whether or not it wreaks havoc on Moralis' of Matt Britt's chances at adminship is irrelevant. They agreed to the experiment; the consequence is for them to decide, not for us. Personally, I think it's patently absurd to hold the format of an RfA against a candidate. --Durin 13:37, 17 April 2007 (UTC)

It may seem irrelevant to you (by the way, I'm sure they appreciate your sympathy) but I doubt a failed RfA seems irrelevant to them. Kafziel ^Talk 13:45, 17 April 2007 (UTC)

• Yes, I'm a cruel and harsh task master who forced these guinea pigs into evil experiments where we drilled holes into their skulls, amputated toes, and tied them to a rack while sending electrical shocks through their testicles. You presume too much Kafziel. These are willing guinea pigs who know full well what the potential consequences are. Please stop trying to defend them against the great injustice of performing an experiment intended to help RfA evolve. If they want to end the experiment, there is nothing stopping them from doing so. Don't make decisions for them. They made the decisions. Not you. --Durin 13:49, 17 April 2007 (UTC)

Fair enough. I wouldn't say the current experiment is wreaking havoc with Matt's chances, though. If the experiment goes to the dogs I'm sure he can file a regular RFA and pass with flying colours. Haukur 13:19, 17 April 2007 (UTC)

• Yeah Matt Britt is a very good candidate by current RfA standards. It perhaps makes him less of an ideal candidate for the experiment :/ Still, we're getting lots of discussion on the format, and it's less than a day old so maybe it's fine. --Durin 13:40, 17 April 2007 (UTC)

)CTU( 7002 lirpA 81 ,83:30 ^{(em ksA ?pleh deeN)}_BCnI^ghiH .sdrawkcab stsop enoyreve erehw AfR na od ew tseggus I

--

<sarcasm>Really? I think if we did it backwards it would fix all the problems, because of course, changing the style of the page will change how people think and act.</sarcasm> ^High_InBC^{(Need help? Ask me)} 04:11, 18 April 2007 (UTC)

But the Sapir-Whorf hypothesis says... Err wait. Actually, if you change the procedure described on the page, people will act differently. Though I can think of ways where changing the style of a page might have effect.

For instance on WP:VAND, if we change the style of the word "not" in "Do not vandalize" so that the word not is rendered in the background color... Wait...

/me stops, before I start stuffing beans up people's noses, too ;-) --Kim Bruning 04:16, 18 April 2007 (UTC)

leizfaK - (: ?AfR ta tupni reffo meht tel ot uoy rof hguone "dettimmoc" si resu a evorp sdrawkcab gnitirw t'ndluoW ?miK ,ton yhW

Nope, if we want to have our RFAs written backwards, we could use a bot. Doing it by hand wastes everyone's time! :-P --Kim Bruning 04:22, 18 April 2007 (UTC)

Subpages not working; partially to Kafziel

It has been suggested that a better method of testing things on RfA would be to create a subpage, advertise it, and allow me to comment on the subpage. I've argued that this has been tried before, with dismal results as the subpages receive little in the way of attention. To support this, I offer some examples:

Recently, I created User:Durin/What is wrong with RfA. I duly advertised it here on this talk page, and also on IRC. The page has been in existence for a week. During that time, the number of people who have contributed to User:Durin/What is wrong with RfA: 16.

Some time ago, effort was made by User:Werdna to try to run an RfA as an RfC Wikipedia:RFA as RFC/Werdna. It too was duly advertised [2]. The number of people who have contributed to Wikipedia:RFA as RFC/Werdna: 40 (and only 11 on the experiment RfA itself; the rest was discussion).

Contrast;

• The number of people who have commented at Moralis' RfA: 121.
• The number of people who, in the first 18 hours, of Matt Britt's RfA: 43
• The number of people who have participated at WT:RFA in the last three days, which has almost solely been focused on these RfA experiments: 61

The simple fact is, subpages do not generate as much traffic as doing the experiments as we are. It's been tried before. Matt's RfA has, in less than a day, exceeded Werdna's experiment in the diversity of people commenting on it, much less all the discussion that has happened here on WT:RFA. --Durin 14:26, 17 April 2007 (UTC)

• I disagree. Werdna's experiment was a good one. (I don't really see what your subpage has to do with my suggestion; everybody has a subpage about something they'd do differently, but you can't expect anyone else to care about it.) 11 isn't a bad turnout. 20 would be better, and probably could be had if it was repeated today. You can't base it solely on the number of participants. I'd rather have 10 participants with something to add rather than 50 saying "this format is stupid" or "the purple sparrow flies at midnight". Kafziel ^Talk 14:44, 17 April 2007 (UTC)
  • Thank you for making my point; "everybody has a subpage about something they'd do differently, but you can't expect anyone else to care about it" People have and do care when it's on the main RfA page. Suddenly, it's a topic. Suddenly, there's rampant interest. Suddenly, it's actually being tried. Werdna's experiment failed in large part because few people actually tried it out. Other than Werdna, just ten people actually tried the format. In the first 18 hours of Matt Britt's RfA, we've got FOUR TIMES that amount of people attempting to use the format. That's a good thing. IO still fail to see what great harm has been caused by trying this experiment for real.
  • Look, simulations are all well and good. They're a necessary part of most development process. But, occasionally, you have to actually send something into the wild to see how it behaves in the chaotic environment it is expected to exist in. Apollo 8 for example was not the first ever launch of the Saturn V. It was the first ever manned launch. They experimented, and learned much from it. And please, don't make comparisons of man/unmanned to undermine the analogy. Matt Britt and Moralis won't be killed by agreeing to be a guinea pig. :) --Durin 14:54, 17 April 2007 (UTC)

Durin: first could you please lay off the caps lock and the bold? And second, not that I want to spoil your fun but comparing your experiments to space exploration is maybe, just maybe, a sign that you're taking yourself a tad too seriously. Sure, there's no horrendous disruption going on but why shouldn't we seriously restarting Matt's RfA when a number of editors are saying that the format prevents them from participating in the debate? You seem to object to "disruption" but if the experiment is getting in the way of the community's ability to make a decision, surely this outweighs the benefits of experimentation. Pascal.Tesson 15:09, 17 April 2007 (UTC)

• Using emphasis is hardly unusual Pascal. Further, I'm rather far removed from taking myself too seriously. Look at the upper right of my user page if you doubt that :) As to the analogy, it should be blatantly apparent that experimentation is a good thing, and using the Apollo program was a great example of that. Yet, people here still seem to think experimentation is disruptive and damaging to the project. I've asked several times now for people to provide feedback on this, or even diffs supporting this supposed damage...with nobody providing anything. Rather funny :) These experiments do not get in the way of the community to make a decision in any respect. --Durin 15:13, 17 April 2007 (UTC)

You're getting all these responses because you are disrupting the RfA process. Is it for a good reason? Maybe. But it's still disruption. The fact that a practically unheard-of editor has 121 comments on his RfA should be evidence enough.

To clarify what I meant about yours: Everybody has a subpage where they publish their manifesto, and nobody cares. Very few people are going to read it, let alone offer suggestions or copyedit it for you. An experiment that actively invites participation is quite different. People do participate on user pages that seek simple input like a comment or a vote.

To stick with your analogy, what I'm suggesting is that you do some testing before the unmanned launch. NASA doesn't just launch a rocket without testing it under controlled conditions first. They don't just let each scientist come up with his own design and take it to the launch pad. Kafziel ^Talk 15:22, 17 April 2007 (UTC)

• The testing was already done on a subpage. So you're suggesting more tests on a subpage? And having 121 people comment on Moralis' RfA is a bad thing? How? --Durin 15:26, 17 April 2007 (UTC)

What subpage, and where? Who knew about it? What changes were made? What about all these other proposed formats? Where have they been tested? That's what I'm asking.

The comments themselves are not bad; they are a symptom of the disruption. Contentious situations draw crowds. Intentionally creating a contentious situation with the purpose of drawing a crowd is disruption. Kafziel ^Talk 15:31, 17 April 2007 (UTC)

• Kafziel, we're talking right past each other. Have a nice day. --Durin 15:33, 17 April 2007 (UTC)

• • I think that with respect to Matt's nomination, Apollo 13 may be a more apt comparison :) >Radiant< 15:15, 17 April 2007 (UTC)

You have 9 editors expressly saying on this talk page that the format is too confusing for them to participate in the debate. Why shouldn't that be considered as "getting in the way of proper decision-making"? As for using emphasis, well, WP:TALK#Good_practice might be a sound reminder. Pascal.Tesson 15:18, 17 April 2007 (UTC)

• Not a single non-experimental RfA is being affected by these experimental RfAs. No articles, no policies, no protals, no projects, nothing is being affected by these experiments except the experiments themselves. Any "disruption" is illusory. --Durin 15:29, 17 April 2007 (UTC)

At some point in time, I guess we just need to try stuff, else you end up stuck in a rut. So this method is too tricky for some folks? Ok, in that case, propose ways to fix it, and see how far it goes. Worst case we don't promote this candidate. (And hopefully the next experiment will go better! :-) )

RFA can use some spring cleaning. And besides, isn't this fun? :-)

--Kim Bruning 15:25, 17 April 2007 (UTC) hey we're volunteering for this project. If it's not fun, what are we doing here? :-P

I must slightly disagree regarding your call to {{sofixit}} ... if you have a severely gangrenous leg, you don't try to fix it, you cut it off. Likewise, if you adopt a cuddly hamster that turns into this, it's probably better to reconsider that decision rather than trying to de-mutate the thing. One last example to follow this space exploration theme: if, during the final countdown, you spot a fire on the shuttle, you don't just try to put out the fire, you abort the launch. If the goal is "spring cleaning" maybe we ought to wait until September so everyone in the Southern Hemisphere can catch up? -- Black Falcon 16:51, 17 April 2007 (UTC)

<grin> Well, this is an experiment. Blowing up is as interesting a result as not blowing up. So try to figure out what's making this one blow up, so that the next experiment doesn't. :-) --Kim Bruning 02:21, 18 April 2007 (UTC)

what?

i requested for admin but it doesnt show up on the page, what am i doing wrong? The juggsd86 15:42, 17 April 2007 (UTC)

• Because it does not automatically transclude to WP:RFA. You have to manually add it. However, you should note that with less than 50 edits it would be impossible for you to pass RfA currently. --Durin 15:44, 17 April 2007 (UTC)
• I also strongly advise you to drop the idea of requesting adminship right now. You have made about 30 edits to Wikipedia since the creation of your account 4 months ago and this is clearly insufficient experience. There's a lot you can do here without being an admin: please do continue to participate as an ordinary editor and if in a few months you believe you have gained sufficient experience and feel you are ready, you can reconsider applying. Pascal.Tesson 15:47, 17 April 2007 (UTC)

Reduce onerous requirements: note bad opposition reasons

Danny's RFA gives precedent for bureaucrats to just ignore irrelevant oppose reasons. To that end, I've started noting when a given oppose is irrelevant to the question "is this person likely to cause damage with the admin tools?" and suggest it be ignored.

Remember that adminship should be No Big Deal - thus, the only question at RFA is "will this editor cause damage as an admin?" 6000 edits, five featured articles or being put forward by three WikiProjects is utterly irrelevant and any bureaucrat should pay it the appropriate amount of attention, i.e. none whatsoever.

Others, e.g. you, are welcome to do so as well. This alone should help reduce irrelevant overrequirements - David Gerard 16:14, 17 April 2007 (UTC)

I ask that you stop your highly inappropriate postings. I have looked over some of them, and they come off as disrespectful and arrogant. If your standard is, "is this person likely to cause damage with the admin tools?", then you are more than welcome to use it. Please have enough respect for your co-editors and do not call for them to be ignored just because you disagree with them. If you think a particular oppose reason is "irrelevant", then you are free to ignore it; please do not call for bureaucrats to do so. -- Black Falcon 16:56, 17 April 2007 (UTC)

Certainly. The first things I'll be marking as inappropriate are the comments claiming that other comments are inappropriate. It is up to each RfA participant to determine the issues they consider relevant, and the community has not placed its trust in you to determine that. GRBerry 18:18, 17 April 2007 (UTC)

Determination by the group that a particular opposition to consensus is not held of value by the group is fully within the purview of the process, however. "This user does not like the color blue." is not a deal-breaker. While I take issue with David Gerard's standards and don't believe they represent the consensus of the 'pedia, you are also incorrect that it is inappropriate to object to objections. -- nae'blis 18:25, 17 April 2007 (UTC)

There is a difference between participating in a discussion by objecting to, or questioning, an objection, and by claiming the right to tell the closing bureaucrats "you should ignore this". Having just gone through one RfA where David did the latter, in every case he was not only unreasonably arrogant, he was wrong because the issues were relevant. GRBerry 18:44, 17 April 2007 (UTC)

David, please do stop that experiment. Your comments come off as completely inappropriate. No good can come out of this way of doing things. Pascal.Tesson 19:49, 17 April 2007 (UTC)

• I certainly hope the bureaucrats are capable of ignoring votes as appropriate. If they're not, we're in serious trouble. Take an extreme example; "Oppose: Likes the color blue too much. --User:JoeWikiUser9999" Should that carry as much weight as "Oppose: Has been blocked for WP:3RR violations several times, including twice in the last week"? Of course not. As we get into more grey areas, there's not as clear answers. David's position is absolutely valid. The only question that RfA asks is whether a person can be trusted with the tools. It does not ask if they contribute to XfD. It does not ask if they have a featured article or a featured picture. It does not ask if they have 2000 edits. If does not ask if they've been here for six months. Trust. That's all. --Durin 17:05, 17 April 2007 (UTC)
  • As I'm sure you're well aware, the comments at hand are not the obvious vandalism that you refer to in your post. I love that you've written, "Trust. That's all." Well, you should realise that people have different criteria for trusting others. When the issue comes to down something as basic and as personal as trust, it would do well to respect the fact that not everyone is and thinks alike. People will disagree, and it is not appropriate to imply that they necessarily do so out of bad-faith malice (e.g., personal grudges) or good-faith stupidity. -- Black Falcon 18:33, 17 April 2007 (UTC)
    • Where did I imply bad faith malice or stupidity!?!?! I brought up a silly example to express the outlier areas. I'm sorry you're deeply offended by that. It had nothing to do with you or indeed any person here. --Durin 18:37, 17 April 2007 (UTC)
      • That's completely unnecessary, I assure you, as I neither am offended or find your comments offensive. In fact, allow me to apologise for the confusion my comment created (part of it is that this discussion bled over from the one I had not long ago with Mackensen at his RfB and on my talk page). I have clarified my somewhat lengthy thoughts in a new section at the bottom of the page. -- Black Falcon 20:21, 17 April 2007 (UTC)

People misnamed admins early on. People who actually administrate wikipedia are called "developers", or "foundation employees". Adminship is more like a drivers license, (an editing license if you will). With the admin bit set, you are allowed to do some more traditionally destructive kinds of edit. In fact we could possibly drop adminship at the moment, or grant it to many more people, because a lot of the issues that previously required restrictions (delete an image and it was gone, for instance) are now not so big a deal anymore. (In the case of our example, commons pushed to have reversible deletion on images) --Kim Bruning 17:28, 17 April 2007 (UTC)

• In cases where a user asks me to do some administrator-related task, I've begun saying, "I'm sorry, I cannot do that because I do not have the sysop flag", or something like that. I totally agree that the term "administrator" is a misnomer. One can do so many administrative-related tasks and not be an administrator. Likewise (and I think that this is perfectly acceptable), one can do so little administrative-tasks and be an administrator. It all depends upon what type of work you want to do on Wikipedia. --Iamunknown 17:33, 17 April 2007 (UTC)
  • I've tried to stop using it as well. It's a small measure, but "has the sysop bit" is less characters than "is an administrator", so I'll take any benefits I can get out of this process. ;) -- nae'blis 18:02, 17 April 2007 (UTC)

I don't think there's anything wrong with looking at the vote count per se. If an editor gets 100 opposes, then the bureaucrats will piss some of these people off when they promote him to admin; they should weigh the advantage of the additional admin versus the potential damage to the community. (Carnildo's last RfA is in my opinion an example where the promotion (which I supported) was not worth the resulting commotion). So bureaucrats should probably have some range outside of which they just do what the vote count says: do not promote below 60% (66%, 70%, whatever), always promote above 90% (80%, whatever). Inbetween, I don't mind bureaucrats using their discretion if they do so all the time. There were perfectly reasonable oppose votes in Danny's RfA, and there are many failed 70%+ RfAs of less well known people that could have been closed as successful by a bureaucrat applying actual discretion (by which I mean: look at the arguments, make up his own mind, and make an informed decision; I do not think that discounting sockpuppets or new accounts has to do with discretion: that part should be automatic for any crat worth his salt). But there needs to be some consistency in what the bureaucrats do. That makes people believe in the system and helps against accusations of cabalism. If the bureaucrats are unable or unwilling to use their discretion in a way that is perceived as mostly fair, we should use a pure voting system instead, which will likely cause less disruption of the community. It would also be faster, leaving more time to write an encyclopedia instead of arguing whether "only 28 Wikipedia space edits" is a "valid" oppose reason. Kusma (talk) 18:25, 17 April 2007 (UTC)

• Hmm, I think Carnildo 3 was worth the commotion; now Carnildo is an administrator and, frankly, I think that is a good thing. S/he does quite a bit of work that is necessary for the pedia. I do, however, agree that there needs to be some semblance of order. But, likewise, if we were going to give total discretion (or, less radically, more discretion) to bureaucrats, what might make sense to one person might totally outrage another. In general, I think the best criterion for adminship is, "Is this user trusted?" If yes, then admin 'em, we can use all the help we can get to clear out backlogs, monitor AIV, etc. If no, then no. --Iamunknown 18:30, 17 April 2007 (UTC)
  • Well, but "is this user trusted?" is a personal question; there is no way we can find out whether the greater part of the community trusts him other than by trying to hold a representative poll about it (i.e. vote). Kusma (talk) 18:40, 17 April 2007 (UTC)
    • (ec) I agree. In real life, trust corollates with the mutual understanding and, in particular, realising that you are safe when in the company of a particular person. Additionally, in real life, actions are not easily reversible. On wiki, however, actions are, so, in my mind, trust may be given more liberally. Perhaps the distinction between on wiki trust and real life trust is not readily apparent or acknowledged when people participate in RFA discussions? Or perhaps the distinction is non-existent? Thoughts? --Iamunknown 18:49, 17 April 2007 (UTC)
• Without commenting on Carnildo or that closing, Kusma hit one of the nails on the head. The way the bureaucrats have been acting has been reinforcing the appearance of a Cabal. The effort made and standards applied in closing the Danny RfA were far different than those applied to the typical discussion. If they aren't prepared to put that level of effort into all close calls, then they shouldn't apply them for their friends, and we should just move to a straight vote. GRBerry 18:48, 17 April 2007 (UTC)
  • If they [bureaucrats] aren't prepared to put that level of effort into all close calls, then they shouldn't apply them for their friends, and we should just move to a straight vote. Agreed. See my response to Kusma above for thoughts about trust. --Iamunknown 18:50, 17 April 2007 (UTC)
• Ah yes, Kusma is once more being his eloquent self. maybe it's time to acknowledge that save a few whackos here and there, a vast majority of RfA participants have pretty sound judgement and are expressing honestly their sentiments. If many of them are saying that they don't trust this user as an admin, chances are that... they don't. For the common good, it's reasonable then to avoid promoting. The current attempts at moving towards something that's as far as possible from voting has huge drawbacks. If we lose the little transparence and simplicity that's left, we might get a system that promotes candidates who know how to play the game rather than candidates who have the community's trust. Pascal.Tesson 19:58, 17 April 2007 (UTC)

Need for tools as a criterion

In Matt Britt's RFA, Tony Sidaway seemed to indicate that the "need for the admin tools" was not a necessary criterion for adminship. I found this interesting as this is a very common objection to RFA candidates. I'd be interested in hearing Tony's explanation of his stance. --Richard 16:38, 17 April 2007 (UTC)

Not answering for Tony; whether or not a person needs the tools is not a reason to oppose. The admin rights are being treated as a status symbol when they are simply additional rights. If an editor finds they use them once a year, that's enough for them to have the admin tools. One admin action is still a benefit to the community, and is equivalent to one admin action from an editor that uses the admin rights a thousand times a day. It's all helpful. One of the editors I nominated for adminship, User:Edcolins, has since being promoted barely used the admin tools. Is he any less trustworthy, any less capable as an admin? No. Is he an inherent danger to the project because he so rarely uses his tools? No. Read Ed's RfA, specifically his answer to question 1 where he says his behavior won't change. Adminship is not about how often you might use the tools, it's all about whether you are trustworthy enough to not abuse the tools when you do use them. That's all that matters. --Durin 16:43, 17 April 2007 (UTC)

OK, thanks. That's enlightening and I think I agree. It's really hard to figure out what are good and not-so-good criteria for adminship based solely on watching WP:RFA. I would never have thought to question this criterion until Tony made the comment. --Richard 16:55, 17 April 2007 (UTC)

I would disagree. We run a risk every time another editor is given the admin tools. Since we don't have a functioning community based system for removing those tools, it is quite reasonable to decide that the risk isn't worth running because they won't be actually using the tools to benefit the project. So the concern is relevant. (On the other hand "need" is the wrong wording, "intention to regularly use the tools to benefit the project" would be more accurate, if longer.) GRBerry 18:51, 17 April 2007 (UTC)

Please enlighten me of this "risk". Everything an admin can do is reversable, and there are almost always stewards around, should there ever be a need to urgently remove the tools - which there hasn't been iirc. Majorly (hot!) 19:00, 17 April 2007 (UTC)

Always worth unpacking an argument that people miss the middle steps in. 1) Wikipedia is built by a community of editors. 2) Administrative actions have effects on the perceptions of other editors and of readers. 3) The ability to reverse the technical steps taken by an admin does not include the ability to reverse the effect that those actions had on other editors prior to their reversal. 4) It is false that everything an admin can do is reversable, and they can drive away other editors and make the encyclopedia worse for a significant amount of time, at least in specific areas. If you want relevant recent specific examples, consider Wikipedia:Requests for arbitration/Darwinek, Wikipedia:Requests for arbitration/Daniel Brandt deletion wheel war, and the Essjay scandal. (Not acting also carries risks; if I recall correctly an admin that declined to take action with regard to Philosophy has probably made the philosophy section of the noticably worse by contributing to the loss of a high value add contributor.)

• It is a common exception, but also not a very relevant one. An editor that doesn't use the admin tools by definition doesn't abuse them either. >Radiant< 08:31, 18 April 2007 (UTC)

Consistent availability as a criterion

As a follow-on to the above discussion about the "need for admin tools", would you agree that consistency of editing is equally not important? If someone has made lots of edits but has also shown a fall-off from wherein he/she has not edited for a while, this should not be a criterion for opposition either? The idea being that the trustworthiness of the candidate and the good use of admin tools are more important than logging on every day and being available. I mention this because this is another criterion that is often used to oppose RFA candidates (i.e. rate of edits is less than x/day or y/month). --Richard 16:55, 17 April 2007 (UTC)

I would agree that "rate of edits" or "consistency of activity" are both poor standards by which to judge candidates. Quality of edits is far more important. However, (and this is addressed more to Durin than you) I also think that we should not try to impose our standards on everyone else. Most of the regulars at RfA are smart enough ... let's let them make their own decisions as to their standards. -- Black Falcon 17:00, 17 April 2007 (UTC)

• No, everyone else is stupid and I'm the only one with the right answers! ;) If I see something that I think is wrong, I'm willing to stand up and say it's wrong. David Gerard did so above himself. This is part of how we encourage a process to evolve. I don't see anything wrong with that. I recognize that the defacto social currency system we have here actively discourages people from ruffling feathers. That's one of the primary reasons why I ran my RfB knowing it would fail, and why subsequent to that I stepped down from adminship...to intentionally remove myself from the social currency system. People might not like what I have to say, but there's little they can do to hold it against me unless I violate Wikipedia policy (and I'm not). Thus, if I see something that I think is rubbish, I'm going to call it into question. This isn't a matter of trying to enforce my own standards. It's a matter of me expressing my opinion. Nobody is forced to go along with me. --Durin 17:11, 17 April 2007 (UTC)

• It's one thing to challenge arguments that you perceive to be wrong. It's another to write (and I paraphrase), "That's the stupidest thing I've ever heard! I don't want to listen to you. Lalalalala ... Hey everyone, IGNORE THIS IDIOT!!!". The difference may be subtle, but it's there. :) It's not particularly conducive to the idea of "discussion" when people decide to ignore what others say. If we feel that an argument is wrong, we certainly ought to challenge it. I wrote that people's comments ought to be respected, but that doesn't mean I hold them to be sacred dogma which is to be worshipped under the full moon. I'm just saying, I think our challenges of various arguments ought to consist of more than just this. -- Black Falcon 18:47, 17 April 2007 (UTC)

• There's a difference between "ignoring" and "considering, rejecting, and not accounting for in decision making." If someone opposes (or supports) a candidate using a line of argument that the community as a whole rejects, the community should not account for that input in making its final decision. (i.e., the closing bureaucrat should ignore it.) This avoids the situation where community action is dictated by elements and positions that the community rejects. Christopher Parham (talk) 20:08, 17 April 2007 (UTC)

• Consensus can change. The only valid evidence that the community continues to reject a particular line of argument is the people who, having seen it attempted in a particular case, nonetheless hold the opposite opinion about what to do. GRBerry 20:12, 17 April 2007 (UTC)

I think there's a question of degrees -- if I ask somebody why they deleted an article or blocked a /13 range (500,000+ IPs), I'd rather not wait six months for an answer. I personally wouldn't make too big a deal about this, provided somebody makes reasonable efforts to respond to communication, but taking an extreme example as above, I could see it becoming a deal-breaker at some point. – Luna Santin (talk) 05:33, 18 April 2007 (UTC)

Comment on the Durin/Matt Britt RFA format

On the one hand, I like the format because it shifts the focus away from a straight "up or down" vote/!vote on the candidate. On the other hand, I hate the format because it destroys the readability of the WP:RFA page. Matt Britt's RFA is so long that I can't just page past it to look at the RFA's below it. (Yes, I know I can use the TOC to navigate but that means that I have to page up to the TOC to get to each RFA.)

Could we consider having the Durin/Matt Britt RFA format as a supplemental discussion page? In other words, the Durin format discussion would be a separate page that is linked to but not transcluded onto the WP:RFA page. Before !voting on the standard format RFA page, editors would be asked to read and comment on the various assertions made on the Durin format discussion page. A committed participant in the RFA process would read and participate in the discussion on the discussion page. However, editors could choose to ignore the discussion page and just !vote on the RFA page as they currently do.

Thus, using my idea, you get the best of both worlds. With the Durin format discussion page, you get focused discussion that assists bureaucrats in focusing on what the contentious issues are for a particular candidate (if there are any). But you also have the readability advantages of the currrent streamlined RFA format.

What we have to ask ourselves is what happens if the discussion on the discussion page diverges from the consensus on the RFA page. I think this would only happen in those "discretionary" cases where the consensus is between 70% and 80%. If a candidate has less than 70% support, the discussion page is likely to have at least a few negative points raised. If the candidate is above 80%, it's likely that there will be few negative points if any. When the consensus falls into the "discretionary" range, there will be a mix of positive and negative points. In this situation, I think the Durin format will help the bureaucrat sift through the many opinions and decide which are the critical issues for this particular candidate and how much weight to give the objections.

--Richard 17:08, 17 April 2007 (UTC)

Note: Haven't read the above; just want to donate my 2c to the discussion somewhere, then go to bed. This header covers it.

Have to say, the Matt Britt format is excellent and exactly how RfAs should work in the real world. Sadly, this isn't the real world, this is Wikipedia. The mark-up is horribly complex for those of use who dread wikimarkup and all of its devil-like ways. There's no easy way to comment in each section - I defaulted to editing the subpage directly and, spending time finding, considering, finding again, checking I was in the right place, finding again and finally commenting, I was continually terrified that the world was about to end and I'd get an edit conflict after putting all the time in. I would have walked away at that point, my dead important views lost to an uncaring world. This is the ideal format, as long as 'crats can gain a consensus from it of course, but the mark-up and the layout need simplifying for those of us who are hard of thinking. Bed now!   REDVERS ↔ SЯEVDEЯ  22:17, 17 April 2007 (UTC)

Determining/forming consensus

• Please understand that percentages of support have nothing to do with consensus. I'll give you an example of how voting fails in determining consensus. Let's suppose, for the moment, that an RfA was about to close with 37 in support and 0 against. Here's such an example (except for the about to close part) [3]. Now, shortly before close an editor comes along and votes oppose citing a diff that he finds quite objectionable [4]. A few minutes later, time is up on the RfA. Has consensus been achieved? It would seem so; 37-1...that's 97% in support! Wow! Except, it wasn't consensus [5]. Oops. Subsequent to the first oppose, only four of the original 37 changed their votes. Voting across seven days does not give any indicator of agreement or disagreement. It's not a static state. --Durin 17:18, 17 April 2007 (UTC)

Yup, I think that is consistent with what I wrote just a little bit earlier on this page about "forming consensus" rather than "determining consensus". I think we are in complete agreement on this and, if I wrote anything above that seems to indicate otherwise, then it was probably poorly stated.

The point that I was trying to make above is simply that the !vote count gives a bureaucrat a high-level idea as to whether a consensus has formed or not. Your example is an exceptional case in which the consensus changed at the last minute due to new information.

In general, if the !vote count is over 80%, all it takes is for the bureaucrat to scan the Opposes to see if there is any really troublesome evidence. I'm not a very consistent RFA participant but I would wager that there are darn few RFA's where the count was over 80% support and yet the RFA failed because the bureaucrat thought the opposes were significant enough to outweigh the supports. Of course, the politic way to handle such a situation is to extend the RFA and let the editors shift the consensus appropriately. The Carnildo 2 RFA was highly contentious because the bureaucrats did not or could not help the RFA community shift the consensus and so they just made the promotion decision by fiat.

If the !vote count is below 70%, most RFAs will fail but it has not been made crystal clear whether the failure is due to the count being below 70% or because the 30%+ editors raised valid objections. In the case of Carnildo 2, the b'crats seem to have decided that the objections were not valid enough to outweigh strong reasons to re-admin Carnildo. I applaud the new effort to provide transparency of b'crat discussions regarding candidates with <80% support. I think the Durin format will also help shift the focus away from numerical percentages and onto the substantive issues. I envision the bureaucrats using the Durin format as supporting evidence for their discussion of RFA candidates whose !vote counts fall in the "discretionary" range.

--Richard 17:42, 17 April 2007 (UTC)

Tallies

A question: can voting tallies be automated? Or can someone create a bot that fix them automatically periodically (30 minutes, e.g.)? I just wonder. --Neigel von Teighen 18:03, 17 April 2007 (UTC)

No, tallies should be removed if anything; RfA is NOT a ballot. Majorly (hot!) 18:08, 17 April 2007 (UTC)

We already have a bot that does its best, but regularly is in error, so it updates a different page instead of the RFA pages. However, Majorly is just plain wrong, with the large number of participants that RFA gets, the tallies are the best tool available for measuring the consensus of the participants. GRBerry 18:12, 17 April 2007 (UTC)

In fact Majorly is plain right. The tallies are the worst tool at gauging consensus. Majorly (hot!) 18:24, 17 April 2007 (UTC)

• Agreed. The tallies should be removed. They were arbitrarily added, without community discussion or consensus, back in February of 2004 by User:Ed Poor. It's time to do away with that decision. Vote counting gives zero indication of consensus. --Durin 18:13, 17 April 2007 (UTC)
• Yes, remove them. Even if the format of RfA remains the same, remove them. They are unnecessary and set up a ballot-like mindset. --Iamunknown 18:17, 17 April 2007 (UTC)
• Tally should remain. RfA is not a ballot and that's all we need to know. It doesn't become a ballot just because of a tally that is merely an indicator of the number of participants in an RfA and their stance towards the respective admin candidate. Such nitpicking is unnecessary.--Húsönd 19:03, 17 April 2007 (UTC)
  • Why are such numbers relevant? Majorly (hot!) 19:08, 17 April 2007 (UTC)
    • Not relevant at all. But after I express my position on a particular RfA, I do check the tally every now and then just to have a quick look at how's the RfA going on. It's handy and I can't see anything wrong about it, so I believe it should be kept unless some sound arguments are provided against it. Let's not make simple things complicated.--Húsönd 19:18, 17 April 2007 (UTC)
      • The progress of an RfA can not and should not be judged by how many votes it has in support as a percentage. It's irrelevant, as an earlier example I noted showed. The tallies complicate things, and have made the environment at RfA exceptionally caustic. They were added without community input. It's a change that should have been undone a long, long time ago. --Durin 19:23, 17 April 2007 (UTC)
        • • It's a change that was inserted a long, long time ago and which nobody complained about for a long, long time which means that it was widely accepted as useful. The tally in no way interferes with the outcome, not unless participants are careless enough to go with the tally flow without analysing a candidate by themselves and making their own judgements. But even for this kind of users, the tally removal would not change anything. Furthermore, I can't see how could the tally be considered complicated, especially if compared with the newly created RfA format that is so afraid of tallies but has so little regard for simplicity. --Húsönd 19:36, 17 April 2007 (UTC)
        • I think that the tallies are in there so it's easier for a crat to judge consensus. If the RFA is failing, it saves time. But, it would not be bad if there was no tally. If there is one, I think it's a waste of time to get rid of it and consquently discuss it over and over again. Evilclown93 19:28, 17 April 2007 (UTC)

I realized yesterday after readin this page that I regularly ignore candidates based on the bot-generated tally at WP:WATCH. That's illogical for a consensus-based model so yes, once again, remove the unhelpful tallies (I'll go through and find all the places people have objected to them later, Husond). -- nae'blis 20:28, 17 April 2007 (UTC)

I don't think RfA can work by consensus, so the tallies serve a useful purpose helping us to see the current vote count. I agree with Nae'blis though that deciding whether or not to vote on an RfA based on the current tally does bring some problems; I ignore too many RfAs where I feel my vote won't have an effect either way. Kusma (talk) 20:38, 17 April 2007 (UTC)

The problem is this: if we keep tallies, they're almost always wrong and should be automated (somehow). On the other hand, I think they're absolutely useless: you can easily watch the last number from the list. I remember there was an experiment to implement tallies on AfD and was a dissaster. --Neigel von Teighen 11:15, 19 April 2007 (UTC)

RFA is not a vote... yeah, right

I wandered off to go shave and, while doing so, I got an insight about the Durin format. While I stand by what I wrote above, my new thought on the Durin format is that it still retains the !vote aspect of the current format except instead of one big !vote, we have a lot of little !votes.

Look, if substance is what counts over pure quantity of votes, why not dispense with the !voting altogether? Just have a discussion page that has two major sections... "Arguments in favor" and "Arguments against". Editors would be encouraged to enter their opinions in each section, maybe even in both sections. However, editors would be discouraged from repeating the comments of other editors. We don't need 20 people saying "civility problems". Either there is a civility problem or there isn't. If there is, provide diffs to support the case. Having 19 extra "yup, I agree, civility is a problem" !votes doesn't change anything. What would change things is if there are people who say "Yeah, and here's another example". Additional evidence supporting or refuting an asertion is valuable. "Me, too" comments are not. If you know that your "me, too" comment will be discounted (maybe even deleted), hopefully you won't waste your time writing it.

The bureaucrat could then read through the discussion and decide if there are any substantive issues. This would allow a bureaucrat to discount issues related to edit-count, use of edit summaries, frequency of edits, lack of image or category experience, whatever.

In such a situation, the example of the 37-1 vote wouldn't be an issue. The question would be whether, in the bureaucrat's judgment, the evidence provided by that one sole opposing editor outweighed the opinions of the other 37 editors.

This, of course, is a difficult call and, especially in a post-Carnildo 2 era, it would take a lot of courage on the part of a bureaucrat to support one editor over 37 others. This is, perhaps, why we have !voting as a crutch for the bureaucrat to leave the responsibility for the decision on the RFA community.

This proposed approach of not counting !votes at all changes the nature of RFA and the role of bureaucrats dramatically.

But, if you believe RFA is broken, then such radical changes might be what is needed.

--Richard 18:10, 17 April 2007 (UTC)

• Sounds good to me. I'll be happy to create an RfA if there's a willing guinea pig... --Durin 18:15, 17 April 2007 (UTC)

Just to be clear, this would be a change completely away from having a consensus based process. For consensus, it matters how many people think a certain way, not just what the points mentioned are. GRBerry 18:13, 17 April 2007 (UTC)

• Consensus is not how many people think in a particular way, unless there is unanimity. --Durin 18:15, 17 April 2007 (UTC)
  • We're not looking for consensus. And that seems to be a keystone in this lark about unanimity. We're looking, on AfD for example, for "rough consensus", and on RfA for "general consensus" (until some well meaning but foolhardy copyeditor removed the word "general" and demonstrated they share the same flawed understanding of RfA). Nobody said unanimity, ever. The kind of consensus sought on RfA is very simply influenced by how many people think a particular way: if there is a large body of support behind a well argued opinion, the size of that body obviously matters. If, for example, a whole bunch of people say that biting newbies is bad, then it's bad, even if some crazy group of outlaws say it's fine. Splash - tk 22:07, 17 April 2007 (UTC)
• I'm not sure where you get the idea that consensus is about numbers in any way, shape, or form, and I say that as someone trained in formal consensus-seeking process in the 'real world', not just the Wikipedia definition of "rough consensus". -- nae'blis 18:22, 17 April 2007 (UTC)

Let me try to understand this: editors provide evidence, bureaucrat makes decision, the end. Umm .... where the hell's the consensus? This proposal is no different than allowing bureaucrats to promote at will (e.g., see user -- promote user), only with the aid of a few research assistants. Again, what happened to the role of editors who are supposed to aid in developing consensus? They're just lowly research assistants now (I write this having been a research assistant)? That's not consensus! That's centralisation of power and responsibility. Does the portion of WP:NOT that states "Wikipedia is not a democracy" somehow imply that Wikipedia should be authoritarian? Am I missing something? -- Black Falcon 18:59, 17 April 2007 (UTC)

Yes, actually you are.

Refer to the title of this section... If numbers matter, then RFA is a vote just not a majority vote with every vote counting equally. If RFA is not a vote, then what is it? Right now, it's a vote requiring supermajority to pass with bureaucrat's having discretion to determine where exactly the supermajority threshold is (usually 80% but sometimes as low as 69%) and also discretion as to which votes to count. Like it or not, this kind of !voting is the way we determine consensus in RFA today.

I think we all agree that a major objection like edit warring, incivility, etc. should stop an RFA candidacy. The questions to ask are: when does an objection rise to the level of being "major" and do a lot of minor objections (e.g. use of edit summaries) count as a major objection? Also, who decides whether an objection is "valid" or not?

My proposal clears the table by saying that the bureaucrats should decide all this and that they should do so on the basis of quality of the objections not just on the quantity of the objections.

--Richard 15:34, 18 April 2007 (UTC)

One problem with the proposal from you is that there are not usually very good reasons for supporting a user apart from 'Hasn't done anything wrong', 'Won't misuse the tools' and 'Very helpful editor'. There are far more reasons to oppose somebody, and so, surely, most RfAs will fail like this. -- Casmith_789 (talk) 15:39, 18 April 2007 (UTC)

I disagree. First, the nominator will have given a lot of good reasons. Second, the proposal rests on the principle that "adminship is not a big deal" and therefore the default assumption is that, failing a very strong objection to promote, the candidate will be promoted. Finally, your comment still betrays a "voting" mentality because you are looking at the numerical quantity of objections over the numerical quantity of reasons to promote. My proposal focuses only on one thing... the quality of the reasons to promote versus the quality of objections to promote. A single, verifiable act of recent vandalism or incivility should be enough to derail a candidacy. In practice, it usually does if only because editors will switch from support to oppose if evidence of such misbehavior is provided.

That said, I am not saying that my proposal is the absolute best way to do RFA although I think it could work. What I am saying is that this is one way that it could work if we want to honestly assert that "RFA is not a vote". Otherwise, we should just be intellectually honest and say "Well, RFA is a kind of a vote with special rules".

Either it's not a vote and the weighing and discounting of for/against reasons is done in the mind of the bureaucrat without considering how many people have the same opinion or it is a vote and the weighing of for/against reasons in the minds of the "!voting" editors with the bureaucrat giving strong weight to the number of people with the same opinion. Let's grapple with the issue and describe what RFA really is.

--Richard 16:13, 18 April 2007 (UTC)

My thoughts on ignoring "oppose" comments

As I see it, there are four classes of "oppose" comments:

1. Obvious vandalism opposes. For instance, "Oppose. Anyone who wants to be an admin needs to get a life." I'm quite sure that everyone agrees that these types of comments, whose purpose is only to disrupt the process, may be ignored.
2. Sockpuppet opposes. These may be expressed more diplomatically than the example above, but their purpose is still disruptive, and so they too should be ignored. I think everyone also agrees on that.
3. Bad-faith opposes. These are cases where an editor opposes a candidate for a reason that that editor does not view to be relevant to adminship. The defining feature of this category is the commenting editor's thought process. For instance, if I think that tea-drinkers cannot make good admins, and I oppose a tea-drinking candidate, that does not qualify as a bad-faith oppose, no matter how ridiculous my standard may appear to be.
4. Good-faith opposes. These are cases where an editor opposes a candidate for a reason that that editor does view to be relevant for adminship. If I sincerely believe (and I don't, by the way) that an editor cannot be a good admin until he or she has acquired 4336 edits, and I oppose an editor with less than that number of edits, my oppose is made in good-faith.

Here's the problem when it comes to ignoring "oppose"s. The first two classes of "oppose" comments can be relatively objectively identified and there is near-unanimous agreement that they should be ignored as disruptive. I believe bad-faith opposes should also be ignored as disruptive, but realise that it's difficult or impossible to identify whether an oppose was made in good or bad faith. W should always assume the former in the absence of evidence to the contrary.

Now, the only argument that can be offered to justify ignoring good-faith opposes is that the standards by which editors judge are somehow "wrong" (i.e., irrelevant to adminship). "Wrong" is a hopelessly subjective word. RfA is mostly about trust; we can't just say "if an editor meets criterion X, you will trust him, damn it!". So, the only way I see that ignoring a good-faith oppose can be justified is if the editor making the argument is stupid and should not be allowed to participate. I know that this is not the intention of the various editors who've suggested ignoring good-faith opposes which they deem to be "irrelevant", but I view that to be the logical conclusion of any proposal to ignore certain good-faith opposes. Either we must allow them, no matter how much we disagree with them, or we must simply ban most editors from participating in RfA. -- Black Falcon 20:21, 17 April 2007 (UTC)

I'm a racist. A Jewish editor requests adminship. I sincerely believe that Jews abuse power, given the chance, and therefore make bad administrators. When I write "oppose, Jewish", I am certainly meeting your standard of good faith. How do you believe the community should react to this (obviously hyperbolic example)? I simply ask because the correct answer is plain, but it is appears to me that your system would arrive at a different one. Christopher Parham (talk) 20:36, 17 April 2007 (UTC)

That's actually not hyperbole. They tend to draw swift criticism and accusations of all sorts of violations, but I don't know if they've ever been officially discounted by a bureaucrat. Kafziel ^Talk 20:40, 17 April 2007 (UTC)

It's hyperbolic in its directness, since usually the comment is more of the form "pushes POV on balkans article and vandalizes serb pages". Usually this requires some sorting out of the extent to which the claims are true. HolyRomanEmperor's third RfA featured some attempts to discount such votes although it nonetheless turned into a disaster. But dealing with opinions that are obviously motivated at least in part by nationalist fervor is a real problem, and a good example of one that a simple test of good faith can't really overcome. Christopher Parham (talk) 20:51, 17 April 2007 (UTC)

Sorry for the late reply - I've been busy. ;) I've actually seen people just say "Oppose: there are too many Indian admins already" and the like. Yes, the vote is immediately derided as racist/nationalist/what have you, and notes telling the closing crat (in boldface, of course) to disregard them, but I don't remember a bureaucrat ever showing up to confirm that it wouldn't be counted. I don't know if there's precedent for a comment like that actually being thrown out. That's all I mean. Kafziel ^Talk 16:43, 18 April 2007 (UTC)

• Agreed with Christopher. And where do we say an opposition vote is absurd for saying "Oppose: less than <X>000 edits"? If a person opposes for having less than 6000 edits, is that unreasonable? What about 8000? 10? 20? Where, in your system, do we begin to say "this is absurd and not a reason to oppose"? --Durin 20:42, 17 April 2007 (UTC)
  • Nowhere. Nowhere is it said that anything is not a valid reason to oppose. If 10,000 edits isn't enough to earn your trust, that's up to you. We don't codify things beyond that, and we shouldn't start. Kafziel ^Talk 20:50, 17 April 2007 (UTC)
    • So let's say we've got an RfA that's teetering on 80%. A user comes along, and posts a vote saying "Oppose: less than 1,000,000 edits. Can't trust". We should allow that vote to sink the rfa? How about an editor who opposes every RfA in sight for no reason? We had that a long time ago (those of you who remember Boothy...) Bureaucrats are expected to reject ridiculous reasons for opposition. If we don't allow the bureaucrats to do that, then we might as well have a straight up/down vote and let a bot do the promotions. Minus the bot, this is precisely what the German Wikipedia does. --Durin 20:55, 17 April 2007 (UTC)
      • To reply to your examples (and this is just for me; Kafziel may disagree). Example 1. The 1 million edits example you gave falls under clear vandalism as no one has a million edits. If the number is 1000, then it's probably a good-faith comment and should be considered. An RfA at 80% is almost sure to succeed. If one comment changes it back to 79%, I'd say that's well within bureaucrat discretion. Example 2. An editor who opposes every RfA in sight for no reason falls under the "obvious vandalism" category and can be ignored. Ridiculous reasons can be rejected, but who's to say that "has less than 3000 edits", which is shorthand for "does not seem to have enough experience" is a ridiculous reason? What if I claim that opposing based on "the candidate posted death threats on multiple user pages" is a ridiculous reason to oppose? Who's to say I'm wrong? As with anything on Wikipedia, every process should be handled with a certain degree of common sense. Editors whose sole purpose is to disrupt the encyclopedia should be ignored. -- Black Falcon 21:08, 17 April 2007 (UTC)
  • (edit conflict) I'll reply to both you and Christopher in this post. Here's how I envision a "good" RfA process: if we perceive an argument to have flaws (the racist one or the one based solely on edits), we identify those flaws, state that we disagree with the reasoning, try to convince the opposer to change her mind, and ... leave it at that. Let other people judge the oppose vote and your counterarguments on their merits. I realise that this system has flaws, but I still think it's the better alternative. Under my proposed system, the worst that can happen is that a few racist opposes will make their way in (and let's be realistic ... most of the regulars at RfA are intelligent, committed to the project, and not outright racists). If we start ignoring "oppose" votes (e.g., based on edit counts), then what we're effectively doing is trying to forcibly create the impression of "consensus" when there isn't any. In any group where there is dissent, "consensus" can be achieved by getting rid of the dissenters. I doubt that that is what we want. -- Black Falcon 21:01, 17 April 2007 (UTC)

A few racist opposes will make their way in is I suspect an understatement. I believe a problem we've had getting good administrators with expertise in the Balkans, for instance, is that whichever nationality they happen to be, people of the other tribe show up and oppose. No, we should not be accepting bad opposes at all. There is no need to do so, and doing so devalues the honest and well thought-out opinions of the rest of us. --Tony Sidaway 22:12, 17 April 2007 (UTC)

See right below? --Kim Bruning 22:17, 17 April 2007 (UTC)

And how exactly do we determine what is a "bad oppose" and what is a "honest and well thought-out opinion"? The simple criterion of "is the comment relevant to the candidate's merits as a potential admin" is not enough as people will have different "honest" and good-faith interpretations of what is and is not relevant. -- Black Falcon 22:22, 17 April 2007 (UTC)

O RLY? I wonder what the section right below this one is about? --Kim Bruning 22:26, 17 April 2007 (UTC)

I commend you on the subtlety of your nudges. =) Your suggestion could possible yield positive results, but it's certain that it will result in a lot of pissed-off people. Lest we forget, that's how the French Revolution started. ... Hmm ... maybe you're onto something ... In order to implement your suggestion, you'd first need to find a willing victim candidate. -- Black Falcon 22:40, 17 April 2007 (UTC)

I believe ignoring oppose votes is a bad idea. I read Christopher Parham's comment above where he gives example of a voter who is racist, and he states that this is a vote to be ignored. Sounds reasonable.

However, if you think about it, such silly votes would be extremely rare and this is a hypothetical situation. Now let me give you a real example. In Danny's RfA the vast majority of oppose votes were well reasoned and in good faith. Some of them mentioned that they did not like Danny's WP:OFFICE activity, yet each of them as far as I saw also mentioned other reasons for opposing. And here is Christopher Parham's own vote:

Support, I don't think any of the issues brought up below are particularly serious, and the large majority are idiotic even by the low standards of RfA.

So, here's my belief. Ignoring oppose votes is, in general, a bad idea. The bigger danger in my view is not a few clueless voters, but rather voters who are firmly convinced other people's votes are worthless. If we were to start an education of the electorate, the latter class of people need to be educated first. Oleg Alexandrov (talk) 01:41, 18 April 2007 (UTC)

• I tend to agree with Oleg on this one. People are not giving enough credit to RfA participants. First of all, nobody (or maybe I should be careful and say almost nobody) is stupid enough to write "Oppose, Jewish" in an RfA. Now of course, a smart antisemite would say "Oppose, don't trust this user for reason blah". That's not something we can fix, regardless of the system we use and the best and perhaps only way to avoid giving too much weight to these is to ensure greater participation in RfA so that these become negligible. It's also no secret that IRC lobbying for RfAs does happen occasionally both in favor and against candidates and it seems to me that, once again, a simpler RfA process that draws more participants reduces the overall importance of such distortions. First and foremost I think people are arguing about hypothetical extreme cases and forget to assume that users who say "I trust this editor" or "I don't" have put as much thought into it as we each do. Pascal.Tesson 14:42, 18 April 2007 (UTC)

Contested opinions

Would it be ok to ignore "contested opinions?"

I think people are worried that bureaucrats might start ignoring random opinions.

How about keeping that under community control to an extent?

Sometimes people leave opinions that other people would like to discuss with them. (This is often -but not always- one of those valuable oppose opinions).

It can be very frustrating when people just do "hit and run voting" as opposed to participating in a meaningful discussion.

So I'm sure lots of people have already thought of this themselves, but we could just ignore opinions where someone has asked a question, and that question has gone unanswered.

People should at least have the decency to keep an RFA on their watchlists while it's still running. If someone really cares about their opinion, and it gets contested, they will reply. (and thus prove their sincerity, and their opinion gets counted :-)).

Does this seem fair? Does this cover all the issues mentioned in some of the sections above? If not, where are the flaws?

--Kim Bruning 20:45, 17 April 2007 (UTC)

Unfortunately this process sounds rather mechanical. Practically, it might get better results. But, it has the same flaws as simple vote counting since it uses basically the same heuristic: the existence of a block of text in a specific place is the primary input it considers. The content of that block of text is devalued. Christopher Parham (talk) 20:51, 17 April 2007 (UTC)

I agree that this is mechanical, but then again the objective is rather limited. It just forces people to actually respond when people speak with them, no more.

Despite being purely mechanical, perhaps it's already enough to get people past some certain invisible threshold. Or perhaps not. The only way to find out is to try it. :-) --Kim Bruning 20:55, 17 April 2007 (UTC)

Umm ... I can think of one question and one flaw. First, the flaw: someone can easily manipulate the process by writing "Could you clarify your position?" after even the most thought-out and developed arguments. And now, the question: wouldn't we also have to apply this to the "support" comments? -- Black Falcon 21:12, 17 April 2007 (UTC)

Dunno if this applies, but if you're asking why oppose !votes get more attention than support !votes, it's because an oppose is worth about 5x as much as a support, so folks want to see a reason why. - CHAIRBOY (☎) 21:17, 17 April 2007 (UTC)

That's a good point, although I would say the correct number is about 2.5 (70-30 usually passes). I'm not opposed to "oppose" comments being subjected to more scrutiny; but determining whether "someone really cares about their opinion" applies both ways. Hit-and-run supports are no more desirable than their counterparts. -- Black Falcon 21:25, 17 April 2007 (UTC)

Actually I'm hoping that people will put "can you clarify your position" after every single support opinion at least once. ;-) the answers might be quite educational

The correct answer, of course, is "Which aspect would you like clarified?".  ;-)

--Kim Bruning 21:35, 17 April 2007 (UTC)

Could we stop calling opinions "!votes"? That alone would be a huge improvement. I'll always do my best to justify my opinions and explain why I think they're relevant to the question of adminship. I expect others to do the same. I don't see a problem with a bureaucrat ignoring my opinion if he disagrees with its relevance to candidacy. We expect bureaucrats to be honest and good judges, so there isn't a problem there. --Tony Sidaway 22:33, 17 April 2007 (UTC)

So do you agree we should try a Contested Opinions rule in some next experiment? --Kim Bruning 22:37, 17 April 2007 (UTC)

No, I think it's too mechanical. The bureaucrat should use his common sense. --Tony Sidaway 22:44, 17 April 2007 (UTC)

Which has different results. Sometimes a hard rule can have more interesting consequences than a soft rule --Kim Bruning 01:54, 18 April 2007 (UTC)

(edit conflict) A second flaw is timing. Someone could mass add the "clairfy" (or equivalent) questions a few minutes before the RfA ends. In principle this is solvable via instruction creep. (Maybe a 2448 hour clause in the rule?)

A more significant flaw is that sometimes the person leaving the comment that was questioned has seen the question and doesn't think it merits a response. Times when I've done it is when I've said effecitively candidate C should not be an admin for reasons X, Y, and Z. Additionally, they might want to improve on areas Φ, Ψ, and Ω but these are not reasons to deny the tools, and then someone asks a question about Ω, which I've already said was an area for improvement but not a reason why I believe they shouldn't have the tools. GRBerry 22:38, 17 April 2007 (UTC)

This might not entirely be a flaw. Apparently someone misunderstood your opinion, right?

If we do this experiment, and such a situation occurs, you could state that Ω was merely an additional optional area for improvement, but that you might change your opinion if something can be done about X, Y, and Z.

That would seem to be a reasonable statement.

Since the current criterium would be strictly mechanical, you could also get away with "you misunderstand" or some such very short answer. But be aware that I'm hoping that people would respond to such an answer in interesting and useful ways.

--Kim Bruning 22:45, 17 April 2007 (UTC)

This would seriously increase the commitment required to vote. I know I don't pick up every response people make to my edits even though I watchlist every page I edit. Simply, watchlists don't work like that for busy pages. All it takes is another edit, or an edit without an edit summary, or 50 edits to other pages that bury the XFX edit so I don't see it. Regards, Ben Aveling 23:19, 17 April 2007 (UTC)

Yes, the objective is to seriously increase the commitment required to vote to a level needed to hold a proper consensus forming discussion.

It's a deceptively simple measure that would force people to continue to follow the whole discussion until the conclusion. :-)

So by that (perhaps counter-intuitive) path, it suddenly means that people's opinions will have to be listened to, and possibly might influence other people. (whereas right now, peoples opinions get averaged out and lost.) --Kim Bruning 23:27, 17 April 2007 (UTC)

I think that the influence of opinions on other people is one of the great flaws of the "discussion" part of RfA. If people could simply "oppose" for any reason they wouldn't have to invent stupid "criteria" which then spread like a disease. I have seen more harm come from "oppose, not enough experience with images" which was copied and used by others than from a simple "oppose" with no reasons disclosed. The "reasons" do not answer the question "should this candidate be an administrator?", they answer the completely different question "what should this candidate do differently so I will support him?" Kusma (talk) 05:16, 18 April 2007 (UTC)

It would mean that people who have the most time to troll rfas would have their opinions count the most. If someone systematically asks nitpicky questions from every opposer/supporter, it shouldn't be counted against them. This change would make it more like a shouting match and less like a debate. - Bobet 10:48, 18 April 2007 (UTC)

Another idea (less substantial change) - thread mode.

Have each RFA (in otherwise substantially the current format) preceded by a few days or so of freeform discussion leading up to it, so that people can all get on the same page rather than what one issue seemed to be with the previous format, which is tendency of people to vote and then their concern is later addressed but they don't come back to change it even though they might. This way any concerns that people have either way can hopefully be brought up and addressed before 'voting' opens. We keep trying to add various more or less structured formats, but i really think that adding a period of simple thread-mode discussion would be refreshing --Random832 03:33, 18 April 2007 (UTC)

This idea isn't new, but I can find one potential weakness already: people have to come back to the RfA page twice. —²¹⁰physicq (c) 03:35, 18 April 2007 (UTC)

People are free to only participate in one or the other phase. This simply attempts to ensure that a thorough discussion/sanity check has occured before voting opens. --Random832 03:37, 18 April 2007 (UTC)

No point. Block voters have already had their minds made up for them made up their minds even before the RfA is posted. The only thing this will do is perhaps make it easier for waterfall voters (voters who try to vote with the consensus) to vote -- they will presumably be able to tell from the discussion whether or not to vote for or against. Kelly Martin (talk) 03:44, 18 April 2007 (UTC)

Isn't that important? Being able to look at the discussion and use that as a factor in whether or not to vote for or against? Say a discussion starts with 50 supports and 0 opposes, and ends at 60-10. It's a passing nomination, I suppose, but someone raised an issue that convinced half of the subsequent voters that the nomination shouldn't pass. Was it voter #49 who "waterfalled", or the tenth oppose? Yet it's "consensus" because of the overall tally. An exploration period would encourage a deeper evaluation of the nominee's background before anyone piles on. Dekimasuよ! 04:11, 18 April 2007 (UTC)

I will say something heretical here. The current support/oppose format works better for discussion. When I look at an RfA, I glance quickly at the support votes, read carefully the oppose votes, perhaps take a look at the questions section, and then decide what to do. And if you wanna say something to the stinky opposer of your favorite candidate, just start a thread under his vote (be polite though :) Oleg Alexandrov (talk) 04:07, 18 April 2007 (UTC)

I support trying out a discussion period. Its implementation can't possibly prove as controversial or confusing as the current experiments. Dekimasuよ! 04:13, 18 April 2007 (UTC)

Want to throw in contested opinions too? (fits with practically anything. :-) ) --Kim Bruning 04:27, 18 April 2007 (UTC) now to find a vict volunteer

What should an RfA change do?

I think we need to

• increase the number of candidates processed every week
• make sure we promote more candidates than now

To keep RfA readable and participation easy, we need a system that does not take more space than our old system (which is confusing enough in cases with more than 100 participants). If we make it hard for people to participate, we move further away from the idea of community consensus: only the people who can be bothered to go through the difficult system will participate, a probably poorly self-selecting sample. If we discuss points like whether something is or is not a valid oppose, we will spend a lot of time on the adminship process, making it even more of a big deal than it is now. The motivation for these changes seems to come from the ideology that "voting is evil", not addressing anything about the problem with RfA's output.

The main thing that is wrong with RfA from the view of the goals above is that opposes are given so much weight, making it very hard to pass. Fortunately, there is an easy remedy for this: lower the promotion threshold. Yes, RfA will still be a vote (or mostly a vote; see somewhere above where I argue for bureaucrats to try to use their discretion consistently instead of creating exceptions for a special class of people), but voting is a useful tool to find out whether the part of the community that opposes a certain action is small enough to have their concerns overruled. Kusma (talk) 05:31, 18 April 2007 (UTC)

• Once more, the easy solution is to lower the bar from 75% to 67%. >Radiant< 08:34, 18 April 2007 (UTC)

  Not that this is a straw poll, or anything (grin), but I would support dropping it to 2/3 (67%). (And honestly, I believe that the bureaucrats could make that choice by their own consensus at any time, since the 75% was created by a previous bureaucrat, as far as I can tell...) - jc37 08:40, 18 April 2007 (UTC)

• I agree that 2/3 is a pretty reasonable numerical threshold. Kusma (talk) 08:50, 18 April 2007 (UTC)
• I am also in favor of 2/3, but only if it were applied evenly. It would be bad to have bureaucrats citing that number to pass some candidates while still failing others who get 70%+. Everyking 13:43, 18 April 2007 (UTC)
• I'm also in favor of 2/3 but I think you still need to have some kind of a discretionary range (say 60%-70%). Above 70%, a bureaucrat would need to explain why a candidate was not getting promoted. Below 60%, a bureaucrat would need to explain why a candidate was getting promoted. Ideally, bureaucrats would just document the rationale for every promotion/failure so that the community would understand how the process was working or not working. --Richard 16:26, 18 April 2007 (UTC)
• I'm not entirely in favour of a 67% threshold, if we decide to go by pure vote or part-vote/part-bureaucrat discretion. I feel a lower limit of 70%, with discretion in the 70-80% range is better. You may ask ... what difference does 3% make? Well, it requires that the support ratio be slightly greater than 2:1. Then again, 75% is a safer percentage that's bound to result in less controversy, less quibbling over results, and less bad blood between editors. To increase the # of candidates and successful candidates, the easiest solution is: nominate more editors. No change in format is really needed for that. -- Black Falcon 17:04, 18 April 2007 (UTC)

Is there anything "wrong" with simple voting?

Is there anything wrong with simple voting? Majorly (hot!) 09:39, 18 April 2007 (UTC)

Yes, a lot of things. But all the other options are even worse. Haukur 10:49, 18 April 2007 (UTC)

It has not evolved properly from when it was a vote. If it had, it would have lost the sections, the tally, the voting. We still have them, yet it is supposed to be a discussion. People don't want to move away from old ideas which simply do not work any more. Majorly (hot!) 11:02, 18 April 2007 (UTC)

RfA isn't supposed to be discussion. It is supposed to be an easy way to determine who gets the admin bit. Kusma (talk) 11:05, 18 April 2007 (UTC)

An easy way to do this would be to randomly assign the bit to users. No, RFA is at heart a discussion of the merits of granting adminship to a candidate. That it has sometimes been mistaken for a vote is a bit of a shame, and it's what we're trying to fix now. --Tony Sidaway 11:11, 18 April 2007 (UTC)

That's what you think. Perhaps people have been treating RfA as a vote because they are wise enough to see that voting is far superior to using an RfC style method to determine an answer to a yes/no question when many people participate. Consensus does not scale, and discussion easily turns RfA into a monumental time-sink, all just due to a misunderstanding that makes people think that voting processes are inherently bad. Kusma (talk) 11:22, 18 April 2007 (UTC)

It's not so much that voting processes are inherently bad, it's that they're not very useful for reaching consensus. If you want to overturn our Consensus policy, that's fine. But for now we're looking at ways to get consensus on whether someone should be an administrator. --Tony Sidaway 11:32, 18 April 2007 (UTC)

Well, I'm not. I am looking for a good method to determine who should be an admin. I do not care at all whether that method involves consensus or magical wiki fairy dust. It should be workable, fair, not too much effort for the participants, and produce many good admins. Consensus is a tool, we are not forced to use it if other tools do a better job. Kusma (talk) 11:38, 18 April 2007 (UTC)

Voting would not be a useful method for determining who should be an administrator. Our problems of factionalism around RFA have arisen from the tendency to treat the process as a vote. The hostility of the process means that we're currently promoting slightly fewer administrators than at the same time last year, despite the growth of the community and the growth of the task of maintaining the encyclopedia. --Tony Sidaway 11:49, 18 April 2007 (UTC)

The hostility seems to be based in the discussion aspects (the voting rationales), not the votes, though. Kusma (talk) 11:52, 18 April 2007 (UTC)

I think I'll leave it there. You and I seem to have irreconcilably different views of what Wikipedia decision-making is about. --Tony Sidaway 11:55, 18 April 2007 (UTC)

Per my comments earlier on this page ("RFA is not a vote...yeah, right"), the question to focus on is "Who is really responsible for making the decision? The community or the bureaucrats?"

If it is the community that is making the decision, then some variation of voting with supermajorities and vote discounting is the way to go. It's not a straight majority vote or even a straight supermajority vote but it is still a vote at its core.

If it is the bureaucrat that is making the decision, then the "!vote" is just input to the decision and the bureaucrat could make the decision based solely on the merits of the support/oppose reasons without counting the number of people with the same opinion. Now, a bureaucrat might wish to consider the opinion of the community in making his decision but, in this view, the bureaucrat is not "determining the will of the community" but only giving weight to "the opinion of the community".

Consider this scenario... what if 30% of RFA !voters said "Oppose, failure to use edit summaries more than 60% of the time"? Should this candidate pass or fail? What if that 30% was 30% of 200 votes? 60 users opposing! I know this is an extremely contrived example but the point is... What is important to you? The quality of the opinion, the percentage of opposes or the actual number of opposes?

--Richard 16:26, 18 April 2007 (UTC)

I'll say it again: one comment, giving valid and serious doubts about a person's suitability to be an administrator, may be enough to make the determination. Forget voting. Voting cannot do what we need. --Tony Sidaway 17:07, 18 April 2007 (UTC)

(edit conflict) I reject the notion that we can predetermine what opinions are relevant, or even determine what opinions are currently relevant for a specific candidate by any means other than the support/oppose division in a specific RfA. The hypothetical example of something extremely great or terrible being discovered by the last opiner is a reason for relisting, not a reason for deciding on the basis of that opinion. Any other decision algorithm makes the bureaucrats the master of the community, not the servant of the community. Since Wikipedia is not a tyranny, the crats should be the servants of the the community.

As to how the support/oppose division should be measured, it should be measured in the percentage ration of the unique supporters/opposers among members of the wikipedia community. (Sockpuppets and banned users are excluded, everyone else has the right to an opinion, and their opinion is valid.) If the decision is 3/1, 30/10, or 300/100 is not relevant; all of them are a 3:1 ratio. If the ratio is 1:1, 2:1, 3:1, 4:1, ... is relevant. GRBerry 17:09, 18 April 2007 (UTC)

In essence, Tony trusts the b'crats to recognize the significance of a minority opinion whereas GRBerry wants the b'crats to go back to the community to have them reconsider what looks, at that point in time, like a "wrong" decision. There is probably no resolution to this difference of philosophy. The current compromise is that b'crats will extend RFA's when it appears that new information may change the opinion of the community and they will explain their rationale when the support/oppose ratio is in the "discretionary range". Since it seems impossible to form a consensus to do anything else, we should just document the current compromise and move on. --Richard 17:44, 18 April 2007 (UTC)

Just treating RfA as a simple vote is clearly almost meaningless. I recently succeeded with a !vote of 108/1/1, an approval rate of about 99.5%. User:Danny succeeded after me (in a controversial RfA) with an approval rate of less than 70%. But I am totally certain that Danny is of far greater value to the project than I am. I suggested, some months ago (Village Pump), that in an RfA the candidate should be asked to make a lengthy and detailed presentation explaining why he should be made an admin, and should be required to answer any and all questions put to him by editors-at-large, and the closing bureaucrat should decide solely on his statement, his answers, and his wiki contribution record. Is that really such a bad idea? You can tell me - I can take it. --Anthony.bradbury 17:24, 18 April 2007 (UTC)

• There are too many people addicted to voting to make this, or any other non-voting proposal, become acceptable. --Durin 17:29, 18 April 2007 (UTC)

It's real bad. :-) For one thing, and despite the great respect I have for Danny, I have no reason to think you're of lesser value than he is and RfA is not really about editor value anyways. My own two cents on this is that RfA seeks to give admin duties only to people that the community trusts to do that responsibly and in an appropriate manner. It seems as though there were quite a few who were uneasy about Danny's ability to do that consistently and while I believe they're wrong, the fact is these editors exist and should be heard. If we start promoting admins on the basis you propose, we'll get a lot more resentment and outraged cries of "Cabal!". We'll also get a system that favors eloquence above competence and we'll completely lose the transparency which I believe to be crucial. As far as Durin's last comment goes: I know you mean no harm but really, there's no need to toss off people who find value in the current system as being "addicted to voting". Many have made perfectly reasonable arguments in favor of the current system or some tweak of it and among them you find experienced Wikipedians who are simply trying to find a solution to the perceived problems of RfA. Pascal.Tesson 17:36, 18 April 2007 (UTC)

• Ok, how about I am addicted to non-voting. Hope that helps :) --Durin 17:38, 18 April 2007 (UTC)

Much better. You little junkie you. :-) Pascal.Tesson 17:48, 18 April 2007 (UTC)

• <quiver> <shake> I need my fix... --Durin 17:53, 18 April 2007 (UTC)

Quick, somebody get Durin a guinea pig! :-) Pascal.Tesson 17:55, 18 April 2007 (UTC)

• hahahahaahha! Thank you Pascal, you just made my day :) --Durin 17:58, 18 April 2007 (UTC)

I'm probably wasting my time here, but here goes. I, for one, like voting because I just don't see any reason to try to reinvent the wheel. For thousands of years, the whole world has known that the only way to get usable opinion data from large groups of people in a finite space of time is to vote. (Lots of governments don't use votes, but it's not because they favor consensus; it's because they don't care what most of the public thinks.) Discussion is great if there are just a few people, but it gets off track when you have 100. The only consensus-based system I can think of that works consistently is the College of Cardinals, and even there they still use ballots. (And some of our RfAs have more participants than they have.)

That doesn't mean I want to turn Wikipedia into a democracy. The U.S. uses votes, and it's not a democracy. Millions of people vote, and then a smaller number of people look at those votes and make a decision. Usually they base their decision strictly according to the data from the masses, but in certain cases they make exceptions. If they do go against the popular vote, the public will expect them to account for that. I've always thought of the bureaucrats as the electoral college of RfA. It strikes a good balance of power between democracy and autocracy. The votes are suggestions—perhaps strong suggestions—but they are not the end-all. On the other hand, having vote tallies to look back on makes it easy to see when a bureaucrat should explain his reasoning a bit better. No one group is in charge.

As I said, this isn't some crazy scheme we cooked up. This is how large societies work. Like it or not (I don't), we are a large society. Wikipedia is a heck of a lot bigger than it was back when adminship was first said to be no big deal, back when an RfA might not even be seen by twenty differnt people. I don't have a problem with discussing votes, or adding comments to votes, or asking for clarification on votes, but taking away all semblence of voting puts too much decision-making power in the hands of too few people and can make it too complicated to have one's opinion heard in the first place. Kafziel ^Talk 18:09, 18 April 2007 (UTC)

To put that last statement another way: In a vote system, all you are is a tick mark. But in a vote system at least you are a tick mark. If there's no tally at all, you will never know whether your input was dismissed, disqualified, skimmed over, or completely overlooked. I'd rather turn my opinion into a tick mark that counts for something than turn it into a thousand words that nobody ever reads. Kafziel ^Talk 18:24, 18 April 2007 (UTC)

WP:POINT violation on Matt Britt RfA

RfAs exist for one purpose; the review of a candidate towards adminship. They are not a forum for discussing the validity or lack thereof of a particular RfA format. There is a reason talk pages exist. When I created the nomination for Matt Britt, I specifically instructed contributors to that RfA to take meta discussions to here, at WT:RFA. This is precisely the appropriate place regarding meta discussions on the format of RfAs. Then...

• Subsequent to this, User:AKAF added a view section titled "This method of RFA is so confusing that I am unable to participate." [6].
• Some hours later, I removed the section [7] and placed it where it belonged, here on WT:RFA [8], noting that it is a meta discussion and inappropriate for the RfA.
• Recognizing that there were editors who felt incapable of contributing (even if I disagreed) I placed a notice on the bureaucrat's noticeboard informing them that some contributors felt incapable of using the format, pointing the bureaucrats to the centralized discussion [9].
• About 10 hours later, User:Xaosflux copied the then current contents of this section of this talk page, creating a situation where there were duplicate sections and disarray in centralizing discussions [10].
• Two hours later, we still had the situation with duplicated sections and disarray in centralized discussion, with both talk pages still having copied content from each other. So, I removed it from the Matt Britt RfA talk page again, placing a marker for the location of centralized discussion [11].
• I pondered for about an hour, and recognizing the validity of the contention that it would perhaps be important for there to be an historical reference to this information, placed a comment on the Matt Britt RfA talk indicating that I would put a marker to the centralized discussion on the talk page of the RfA after it closed [12].
• Three hours later, User:Xaosflux reverted this and re-inserted it again into the Matt Britt RfA talk page [13].
• At this point, Xaosflux created a subpage of the RfA at Wikipedia talk:Requests for adminship/Matt Britt/Confusing, and transcluded it to this talk page [14], attempting to create another place for centralized discussion.
• Xaosflux then transcluded it to the talk page of the RfA [15] and this talk page [16]
• Four hours later, User:Radiant! moved a majority of the contents of the /confusing subpage [17] to the RfA itself [18].

Now we have a situation where this discussion is scattered all over the place. There's comments on the RfA itself, where meta discussions are inappropriate. There's comments on the RfA's talk page. There's comments here at WT:RFA. There's comments on the /confusing page that Xaosflux created. It's scattershot, all over the place. This makes it extremely difficult to engage in discussion regarding this format. Where is somebody supposed to comment? Who knows? In the least, it keeps moving, and keeping track of the bouncing ball is a trick best left to a bored cat.

I consider pushing the discussion regarding the format of the RfA onto the RfA itself to be a WP:POINT violation because:

• Meta discussions are inappropriate for a given page/article in dispute. We do not, for instance place discussion regarding the content of an article on the article itself. It goes on the article talk page. This is why we have talk pages!
• If a person moved content from this talk regarding the atrocious state of the current 'normal' format of RfAs onto an RfA talk page they would be immediately reverted. If they continued to do so, they would be warned to stop doing it as it causes a disruption to the RfA. This RfA is a perfectly valid and live RfA. It is highly inappropriate to have such discussion on the page itself. It deserves the same respect we give to 'normal' formatted RfAs.
• The people who are insisting on putting this content on the RfA have a strong dislike of the form of the RfA. By pushing this discussion onto the RfA, they are working to disrupt the RfA and prevent it from continuing in an appropriate manner.

This is not about me being in disagreement with the people who's opinions are against this format. Frankly, I have significant concerns about the format myself. This is about these users attempting to force discussion regarding the format of this RfA onto the RfA itself in violation of WP:POINT and going against the ideals espoused at Wikipedia:Centralized discussion.

It is blatantly obvious to me that the editors who insist on pushing this content back onto the RfA page itself are quite willing to revert war in order to get their way. I refuse to do so. I would appreciate it if others would step in and re-re-re-re-centralize the discussion to an appropriate place. The current situation is inexcusable as meta discussions are inappropriate for the RfA itself, and the meta discussion is scattered across at least four different pages now. --Durin 12:47, 18 April 2007 (UTC)

Does it really matter? The experiment has failed already. Who did what is irrelevant here at WT:RFA; if you find the behaviour of individual users disruptive, that should be taken care of via WP:DR. Kusma (talk) 12:51, 18 April 2007 (UTC)

• The experiment is not complete, thus asserting it has "failed" is premature. Further, per my comments elsewhere, there are more possible outcomes of the experiment than simply worked/didn't work. The people disrupting this RfA to make their point that it doesn't work are salting the experiment and forcing their view in a disruptive manner. --Durin 12:53, 18 April 2007 (UTC)

File:MrT.jpg

Durin is a one-man wrecking machine! Kafziel said so!

**This could all have been avoided if you had left the section in the RfA in the first place, or if you had returned it after others asked you to. You've been a one-man wrecking ball around here the last couple of days, so I don't think it's fair for you to start accusing others of wrongdoing the first time you don't get your way. Kafziel ^Talk 13:12, 18 April 2007 (UTC)

• • • I'm a one-man wrecking ball now? Wow. I mean, wow. --Durin 13:22, 18 April 2007 (UTC)

Reply to Durin's accusation of WP:POINT

For those who are reading this for the first time, I should make it clear that the addition of the section to Matt Britt's RFA was not a WP:POINT. I added it, after trying for 10 minutes to understand how to add a new section to the RFA. It occurred to me that I am not too inexperienced as an Editor (Although clearly not as experienced as Durin). The extreme difficulty of adding new sections to the RFA, compared with the relative ease of adding a single comment to an existing section made me think that it was appropriate to add a section to canvass how many users felt disenfrancised by the new format. I would have accepted Durin's move (I made an aborted revert after realising that Durin had moved, not deleted the section) but the revert was finally made by others. AKAF 13:30, 18 April 2007 (UTC)

---

I should add that this is not related to my personal feelings about running experiments under "Live" RfAs. I find the deliberate malforming of RfAs unhelpful and disruptive. Additionally I find Durin (and other's) combined attacks of anyone voicing this opinion to be extremely uncivil when taken as a whole. Durin's repeated assertion that "meta discussions are inappropriate for the RfA" is rather difficult to justify, given that the whole RFA is a POV push of Durin's personal opinion. I find that this is exemplified by the following:

• Almost all !votes of oppose are attacked, while all !votes of support are accepted without question.
• Any suggestion that deliberate malforming of an RfA can lead to an invalid RfA is immediately attacked and discounted.
• Any suggestion that the judgement of a candidate in accepting a malformed nomination is poor leads to further attacks.

While I appreciate that Durin has taken the time to have new ideas, I do not accept that there is any kind of consensus as to what the problems in RfA are, and thus testing anything on a live RfA is problematic at best and deliberately disruptive at worst. Durin has indicated that he wants to have a large number of these "live tests", and so I see this as a potentially escalating problem. AKAF 13:30, 18 April 2007 (UTC)

The trend that people opposing candidates have their opinions commented on more than people supporting candidates is not confined to this RfA; look at any other and you'll see that trend appearing. --bainer (talk) 13:56, 18 April 2007 (UTC)

• AKAF, first whether I am more experienced or not is irrelevant. I have held and continue to hold to the ideal that all good faith editors whether making their first edit or their 100,000th edit are equals. My comments regarding WP:POINT violations are considerably less directed at you.
• Second, you're now the third person who has made the accusation that I am "attacking" all or nearly all opposes in these experiments. This is flatly and provably false. When the first such accusation was made, I counted on the Moralis RfA. 46 opposes at the time; I'd responded to only 10. Less than 25%. Further, I've barely responded to anything on the Matt Britt RfA. I've made very few comments there since the RfA opened. I'm sorry but this accusation is simply false.
• Third, I am far from being the only one who thinks that attacking a candidate for being a guinea pig is a wrong idea. Please see Wikipedia:Requests_for_adminship/Matt_Britt#Nominee_may_cause_undue_disruption.
• Lastly, I'll restate what I stated above. Meta discussions do not belong on the main page of whatever the meta discussion is about. This applies to articles, policy pages, images,...you name it. This is how we do things around here. I recognize people feel concerned they are "disenfranchised" (hard to understand when a) it's not a vote and b) plenty of other people felt well capable of contributing), and placed a pointed on WP:BN to alert the bureaucrats to this possibility, even trying to word it in favor of that viewpoint as much as I could [19]. We don't need to keep forcing meta discussion onto the RfA in order for the bureaucrats to be aware of this. --Durin 13:40, 18 April 2007 (UTC)

I phrased that poorly, sorry. You have personally replied to a lot of oppose comments (I find 10% a lot). But you will note that an environment has been created in which a group of other editors who agree with the reformats are replying to oppose comments with considerably less civility than you are.

As for the meta-discussion, I just don't think that you're right. The RFA is an discussion of the User's judgement and record, more than anything else. If it is the decision of other editors that the acceptance of an uncommonly formed RFA is an indication of poor judgement, how is that meta-discussion? I agree that there's a certain amount of POINT voting, but it's hard to tell since there is no consensus whether a user agreeing to the reformatting is poor judgement or not.

I can see that you feel strongly about it, but in my opinion having a section on all these "live tests" for those who felt unable to use the format would defuse a big problem, namely that there will always be a subset of users who feel that these are invalid RFAs, and that there should be no promotions from these tests. AKAF 14:04, 18 April 2007 (UTC)

• Would you then be opposed to having such a section on all other RfAs using the "normal" format? A number of people find the current format quite objectionable. If it is reasonable to have similar objections on Matt Britt's RfA, it's entirely reasonable to have such objections on any other RfA where people feel the format is objectionable. --Durin 14:10, 18 April 2007 (UTC)
• Please note that this is one of the points I noted above; if someone added such a section to a "normal" RfA they would be immediately reverted. If they re-added it, and then re-added it again they would be rightfully accused of violating WP:POINT. That's why it's a WP:POINT violation here, on this RfA. --Durin 14:12, 18 April 2007 (UTC)

• You should really just stop now. There is no basis in policy for removing others' comments from a discussion and relegating them to a talk page or subpage. You were absolutely wrong to move them. The policy doesn't make exceptions for metadata, and sarcastic comments like this (and this) aren't going to change that. Kafziel ^Talk 14:14, 18 April 2007 (UTC)

• I will continue voicing my opinions as I see fit, as long as they are in line with Wikipedia policy and you have absolutely no right to insist that I stop. Further, your comments [20][21] are not exactly in line with your ideals expressed in this latest comment from you. If you think metadata belongs on main pages, then please tell me why we have talk pages? Why not just push all the commentary from WT:RFA to WP:RFA? Afterall, it's just metadata...deserve to be on the main page! --Durin

• I don't have a problem with sarcastic comments per se; my feelings aren't hurt. I'm just saying those comments aren't helping you prove your case. There are plenty of places that say it's bad to touch anyone else's comments. There's nowhere I know of that says it's okay to move people's comments around however you want, amidst the protests of others, as long as you deem them irrelevant. Kafziel ^Talk 14:57, 18 April 2007 (UTC)

• I didn't deem them irrelevant. I gave them considerable weight. Moving them doesn't mean they are irrelevant. Would you please show me a main article page where we have discussions about the article on the article itself? Also, your attack on me for being sarcastic is...well, you should look in the mirror. :) --Durin 14:59, 18 April 2007 (UTC)

• Stop saying I'm attacking you. I just said I don't mind the sarcasm. I love sarcasm. Perhaps irrelevant was the wrong word. You (and only you) deemed them "inappropriate". Semantics aside, there's still no basis for moving them. Sarcasm won't change that, whining about my comments won't change that, and trying to compare RfA to an article won't change that. Kafziel ^Talk 15:07, 18 April 2007 (UTC)

• You're not attacking me but call me the boss and a one man wrecking machine. I'll try to keep that in mind. --Durin 15:10, 18 April 2007 (UTC)

• Really? That's it? That's your response? Still no policy to back up moving others' comments? Just a picture of Mr. T and more lamentations about your sensitive feelings? All I'm asking for is a link to where it says what you did is okay. If you can't come up with one, just say so. Why are you trying so hard to make this personal? I'm not the one who moved the section back, or moved it to a different page, or anything. I never touched it. As far as I'm concerned, we're just talking here. Kafziel ^Talk 15:21, 18 April 2007 (UTC)

• de-indent And now my feelings are too sensitive. Funny :) Have a nice day Kafziel. --Durin 15:30, 18 April 2007 (UTC)

• Yeah, you said that yesterday, too, but you didn't leave me alone. Still no luck on finding that policy, huh? Okie dokie. Kafziel ^Talk 15:34, 18 April 2007 (UTC)

• Silence on a point does not mean I concede it. Do not stuff words in my mouth. I simply recognize that you and I are totally incapable of interacting with each other in a rational manner. Thus, continuing discussion between us is an exercise in utter futility. --Durin 15:38, 18 April 2007 (UTC)

• It's funny, though: I was asking for diffs or links when you decided to stop yesterday's conversation, too. You can't come up with any, so you'll "be the bigger man" and stop the conversation on your terms without admitting that you were wrong, only to show up a little later saying the same things I'm asking you to prove now (or worse, levelling accusations WP:POINT at other editors for daring to defy you). You've tried to change the subject six ways from Sunday, but I'm not having it. You can make this about me, but it still won't make you right. Kafziel ^Talk 15:54, 18 April 2007 (UTC)

• You are welcome to your opinions and (wrong) insights into my behavior. Have a nice day. --Durin 15:55, 18 April 2007 (UTC)

---

I would actively encourage the addition of extra discussion sections. Perhaps two discussion sections:

• I couldn't take part because of problems caused by the format.
• I dislike this format (but voted above), and it's being a problem in this RFA because....

In my opinion, this would be a better method of centralising the discussion, and hopefully would take a number of POINT opposes to a central area, leaving only those who feel that a candidate has showed poor judgement.AKAF 14:20, 18 April 2007 (UTC)

Everybody needs to take a deep breath here. It's alright to have a little experiment and it's also ok for people to point out that the experiment is giving everyone a headache and should be aborted. There's absolutely no reason to turn this into such a personal affair and anyone who has trouble with that part should just walk away from this debate and cool down for a while. Pascal.Tesson 14:28, 18 April 2007 (UTC)

Has anyone noted that the RfA is a lot easier to comment on now, after all the <includeonly>s have been stripped? Those who want it aborted are also those who lack the desire to help improve it. –Pomte 14:56, 18 April 2007 (UTC)

• I was hoping one or more people would help improve the format. If it was you, THANK YOU! :) --Durin 14:57, 18 April 2007 (UTC)

• User:Freakofnurture gets the credit.[22] –Pomte 16:00, 18 April 2007 (UTC)

• If people have legitimate concerns against this novel process, it follows they have legitimate concerns against promoting candidates via this process, and hence that should be noted. This has nothing to do with disrupting anything, nor with making a point. >Radiant< 08:21, 19 April 2007 (UTC)

A Simple Suggestion

Allow the crats to exercise discretion early - Explicitly state what !votes are to be discounted by crats, i.e. what they consider to be poor arguments, sockpuppets etc. These !votes are to be struck out (e.g. Oppose, not enough blah counts!) by any crat anytime during the RfA along with an explanation. These struck !votes may be overriden by any other crat which would then be discussed at the noticeboard, and they be discussed/critiqued by any other editor if they feel that there are lapses in discretion. Everything else be left the same as it is.

Also consider : possible extension to support !votes as well? - Mailer Diablo 14:46, 18 April 2007 (UTC)

I'd always agree simply writing Support 15:07, 18 April 2007 (UTC) isn't helpful, that's just voting (although I'm guilty of it when I'm being lazy). Since it is the 'crats discretion to discount such opinions, this could be a good idea. Majorly (hot!) 15:07, 18 April 2007 (UTC)

This would alleviate the problems encountered with Carnildo's, Danny's and Ryulong's RfAs, where the numbers indicated that the candidates hadn't passed but the Bureaucrats decided that consensus had been achieved. Explanatory notes and a summary of the 'crats findings would have been useful in determining how the decision had been made. (aeropagitica) 15:20, 18 April 2007 (UTC)

Indeed, I'm looking out for the crat's thought process during the course of the RfA !voting, rather than at towards the closing time. - Mailer Diablo 15:23, 18 April 2007 (UTC)

I like this idea but I'm concerned as to what the criteria would be for discounting votes. There has been some dispute as to which !votes should be ignored by bureaucrats.

Here's a list of criteria that are discounted by general agreement...

1. Sock puppet accounts
2. !Votes which are based on a personal grudge
3. Very new accounts
4. !votes by anonymous IPs

Here's a list of criteria that some people would argue should be discounted...

1. No need for tools
2. Failure to use edit summaries
3. "Me, too" or "Support/oppose per User:X"
4. Not enough edits/experience
5. Not enough edits in a particular namespace (e.g. the Wikipedia and Wikipedia talk namespaces)
6. Not enough edits in xFD discussions
7. !Votes which are based on Wikipedia philosophy such as inclusionist/deletionist

These lists are "off the top of my head" and I readily grant that the lists could be expanded. I also grant that not all of these criteria have the same weight with regards to "discountability".

So... which of these criteria are you proposing that bureaucrats discount?

--Richard 16:28, 18 April 2007 (UTC)

There was one RfA I remember (no idea whose) where the first two Oppose votes both attracted heavy criticism from users (the nominee wasn't involved, but other users were complaining that the votes weren't based in policy, etc.) As a result, the third vote simply said Oppose without a reason. (By the way, see the talk page of WP:QAV to see why you have to be very careful about disallowing votes.) --ais523 16:32, 18 April 2007 (UTC)

I think that any of these could be discounted - each RfA can be equally measured against these criteria and the only those that they fall down upon need be highlighted in the striking of votes and summation at the end of the process. (aeropagitica) 17:31, 18 April 2007 (UTC)

Strong Oppose. If bureaucrats have this much discretion in closing RfAs, we may as well let them do on-sight promotions. See a user, promote a user; no community input required. -- Black Falcon 17:11, 18 April 2007 (UTC)

• That would be ok with me. The reasons people are being opposed these days are often absurd. I trust the judgement of a bureaucrat considerably more than I trust the judgment of a person who says "Oppose: Only have 897 mainspace edits, and I demand 1000". --Durin 17:17, 18 April 2007 (UTC)

This would then raise the accusation of the existence of a cabal of like-minded admins who support a particular line of the promoting Bureaucrat. A transparent process would be better for all involved and passive observers too. A meritocracy is more beneficial to this collaborative project than a series of cliques. This is a worst case scenario but I can all-too easily imagine the accusation being raised as soon as a promotion goes ahead when someone opposed to the new admin's way of thinking decides to speak out. (aeropagitica) 17:31, 18 April 2007 (UTC)

I'm all with aeropagitica: transparency is important and supporters or opposers whose arguments will be discounted by the closing b'crat should get some sort of heads-up if possible. I strongly advise against making precise rules about what's a good reason and what isn't: unless the motivations are obviously bad-faith, disruptive junk then let people speak their mind. "Ridiculous" is in the eye of the beholder: for instance I find Walton Monarchist's "support unless awful" approach to RfA careless but I'm positive he's being honest and thorough in his assessments and at least he's consistent. We just disagree on what the criteria should be. I find "lack of experience with images" a bad reason to oppose but I can live with it as long as people who oppose on that basis are consistent. Pascal.Tesson 18:09, 18 April 2007 (UTC)

If there was something that could give information about what not to vote in RfAs besides Wikipedia:Arguments to avoid in adminship discussions (which is only ideas and theories anyway), we would have fewer arguments about which reasons are valid reasons to support and oppose. Captain panda 21:31, 18 April 2007 (UTC)

• In response to Aeropagitica, "where the numbers indicated that the candidates hadn't passed but the Bureaucrats decided that consensus had been achieved" - as far as I can tell the 'crats decided that consensus had not been achieved but they would promote anyway. There is a difference there. >Radiant< 08:23, 19 April 2007 (UTC)

The noinclude tags on Matt Britt's RfA, and support/include/neutral sections using the current RfA method

This isn't yet another post about the correct method to use for RfAs, but instead a merely technical suggestion. At the moment, RfAs either have to use Support in bold as a section header (with no section-edit link), use ===Support=== as a section header (lengthening the RfA TOC to beyond a usable length), or use noinclude trickery (acceptable on current-style RfAs, but leading to much confusion in the case of Matt Britt's). If a proposal I've made on MediaWiki talk:common.css to make it possible to limit TOCs to a certain depth doesn't receive objections (and I don't see why it would), a fourth method will become possible: using ===Support=== as a section header (complete with [edit] link), and limiting WP:RFA's table of contents so that the 'Support' entry doesn't appear there. What do people think of this possibility? --ais523 15:16, 18 April 2007 (UTC)

If there is support to use ====Support==== and ====Oppose==== and ====Neutral==== (level 4 currently), I don't think there's a need to supress them from the table of contents because it's only 3 subheadings per RfA. With 10 or so RfAs at any given time, the TOC at WP:RfA won't override into the first nomination. –Pomte 15:53, 18 April 2007 (UTC)

Section editing on RfA. Yes, please. SchmuckyTheCat 19:01, 18 April 2007 (UTC)

A record has been set

I got curious, and ran the numbers totaling contributions to WT:RFA per rolling 7 days through the extant history of this talk page. The prior record for the number of contributions per rolling 7 days was during the Ryulong promotion affair with 567 comments, back in January of this year. In a close second place was the Carnildo promotion affair back in September of '06 with 538. We've just blown that record out of the water, with 748 contributions in the last 7 days.

Correct me if I'm wrong, but this doesn't seem to indicate the Moralis and Matt Britt RfA formats are controversial. ;) --Durin 18:59, 18 April 2007 (UTC)

I would say that you added an incorrect negative sign in your mental math. More discussion implies more controversy. GRBerry 19:05, 18 April 2007 (UTC)

• I would say you missed the ";)" on the end of my comment. :) --Durin 19:08, 18 April 2007 (UTC)

• I wouldn't say that having so many comments is necessarily a good thing. Every comment made here (to fix something that isn't broken, in my opinion) is one less minor improvement to an article. -- Black Falcon 19:15, 18 April 2007 (UTC)

• Getting RfA to work properly will significantly help the project. These aren't wasted edits. --Durin 19:31, 18 April 2007 (UTC)

• That assumes that it currently doesn't work well enough. But, as you noted yesterday, I suppose this is something about which we'll have to agree to disagree. -- Black Falcon 19:36, 18 April 2007 (UTC)
• What he said. Kafziel ^Talk 19:44, 18 April 2007 (UTC)

• What, it's somehow detrimental to participate in policy-forming activities? There are no deadlines on Wikipedia. Every article improvement that can be done will be done - whether sooner or later. You seem to be implying that a comment made here will actually subtract from the total number of article edits a contributor will make in their career here. That's a totally false implication. -- Earle Martin [^t/_c] 21:05, 18 April 2007 (UTC)

• No, I'm not implying that. I'm just pointing to the opportunity costs of every comment made here. Every edit made here consumes time that could be spent elsewhere. Now, of course, I'm not lobbying for an end to policy discussion as such discussion is necessary to establish and maintain frameworks in which article editing can take place. The intent of my comment above was simply to express the opinion that the commenting in the past few days (in which I admit I heavily participated) may not have been ... and please don't lynch me for this ... the most productive. I would love to be proven wrong by seeing some sort of positive result, but I can see only two: (1) we adopt one of the two experimental formats as the standard, or (2) we keep the existing standard. I view the first to be a negative result and the second to be neutral (no change). Of course, there is the fact that we now know of 2 formats that won't receive consensus support. If you think the current format of RfA is broken and needs significant reform, then that's probably consoling. If you don't, well ... -- Black Falcon 22:51, 18 April 2007 (UTC)

A record of another sort?

An admin has just gone rogue. Has this ever happened before? The rogue admin got in three edits before being blocked and then emergency desysopped, and got in 25 blocking, unblocking, unprotecting, and deleting actions. See WP:ANI#Robdurbar. Carcharoth 10:45, 19 April 2007 (UTC)

Thirteen minutes from when he deleted the Main Page until he got desysopped... Carcharoth 10:47, 19 April 2007 (UTC)

That's shocking. I would say we need either (a) more active stewards or (b) to give crats the ability to desysop. Higher standards at RfB, whoopee. – Riana ^ऋ 10:48, 19 April 2007 (UTC)

I agree with you completely, except for the bit about raising the standards. B-crat candidates are already pretty much doomed because of all the "we don't need anymore" voters; hence, the standards are already high. — Deckiller 10:53, 19 April 2007 (UTC)

Heh, my sarcasm didn't translate - I don't want higher standards. RfB is ridiculous. We're going to need to get people canonised before they can become crats. – Riana ^ऋ 10:59, 19 April 2007 (UTC)

Undoubtedly this will raise new questions about whether admins claiming "right to leave" should be desysopped. – Chacor (RIP 32@VT) 10:49, 19 April 2007 (UTC)

Looks like I was beaten to it. – Chacor (RIP 32@VT) 10:52, 19 April 2007 (UTC)

Is there a good reason why the crats can't desysop? Moreschi ^Talk 10:53, 19 April 2007 (UTC)

Probably some separation of powers thingy. Carcharoth 10:55, 19 April 2007 (UTC)

I can't think of one, especially with the already amazingly high standards. Plus, my immediate response to this incident was "we need mroe crats!" before someone reminded me that crats can't desyssop. That incident alone means I shouldn't run for RfB for six months :) — Deckiller 10:56, 19 April 2007 (UTC)

Was there actually a bureaucrat around and more accessible than jhs was? —Cryptic 10:58, 19 April 2007 (UTC)

Not so relevant: standards at RfB are stupidly high anyway. Moreschi ^Talk 11:02, 19 April 2007 (UTC)

I think more to the point is: should admins be able to unblock themselves? That would be the short-term answer before the question of de-sysopping is even required. – B.hotep ^u/_t• 10:59, 19 April 2007 (UTC)

Yeah that has always seemed bizarre to me. Will (aka Wimt) 11:01, 19 April 2007 (UTC)

Admins can and do block themselves by mistake, and in this case admins who were blocked found it pretty useful to unblock themselves. Moreschi ^Talk 11:02, 19 April 2007 (UTC)

True but you could make it possible to remove your own blocks but not someone else's. Or just wait for someone else to unblock you! Will (aka Wimt) 11:05, 19 April 2007 (UTC)

Allowing to self-unblock only if you self-blocked seems like a reasonable idea. To control a rogue admin with a blockbot who blocks every available sysop you'll still need a steward, though. Kusma (talk) 11:08, 19 April 2007 (UTC)

What is needed is to set the system up so a self-unblock only restores editing privileges, and not admin privileges. To regain the admin privileges, you need to get someone else to complete the process. Then you hope that the number of good admins outweighs the bad ones and they get their blocks in first... Hmm. Maybe that wouldn't work! Carcharoth 11:11, 19 April 2007 (UTC)

Ouch, that's the terrifying scenario. The problem with allowing self-unblocking only if you've blocked yourself is exactly that: in that scenario, you'd need the stewards to unblock maybe hundreds of admins, desysop the rouge admin, block him, and then clean up any other mess he's left behind. That's probably why we need admins to be able to self-unblock. Moreschi ^Talk 11:15, 19 April 2007 (UTC)

I'm not sure whether it would be possible to introduce some kind of protection that prevented a sysop blocking more than three sysops in 24 hours or something. That would prevent that kind of blockbot attack. Will (aka Wimt) 11:13, 19 April 2007 (UTC)

That's a good question too, I thought of that when I unblocked myself. Blocking should mean blocking from all editing privileges, including administrative privs. – Riana ^ऋ 11:03, 19 April 2007 (UTC)

So, does anyone actually know whether anything like this has happened before? Is it that common? Carcharoth 11:00, 19 April 2007 (UTC)

• Whoa! What on earth?! Still, 13 minutes is a pretty good response time. >Radiant< 11:15, 19 April 2007 (UTC)

The only case I know of where an admin was emergency desysopped for what seemed at the time to be rogue behaviour was in May 2005 - see Wikipedia:Administrators' noticeboard/IncidentArchive23#Ævar Arnfjörð Bjarmason. It used to be possible for admins to use rollback, protect/unprotect a page ETC. while blocked - that was fixed as bug 3801. Graham87 12:35, 19 April 2007 (UTC)

Ah, one other: User:Husnock was rapidly desysopped after it emerged he'd given away his password to help someone evade a block. Moreschi ^Talk 12:55, 19 April 2007 (UTC)

• There will undoubtedly be calls for reform (and already are) regarding things we could change to have avoided this admin account going rogue. Reality; three admins emergency desysopped out of >1100.    .3%    (I'm apparently not allowed to use forms of emphasis, so I'm trying out adding spaces around something I want to emphasize in an effort not to annoy certain people). --Durin 13:12, 19 April 2007 (UTC)