Misplaced Pages

Revision as of 22:47, 23 February 2006

• ]

Archives

• Prior to June 2003, requests for adminship were made and discussed on the mailing list.
• For Archives of discussions from June 2003 to the present, please visit Misplaced Pages talk:Requests for adminship/Archives.

Deletion of never accepted RfAs

Since the policy change requiring nominations be accepted before being posted at WP:RFA, nominations that have not been accepted have generally been deleted once the nominee indicates lack of acceptance. There's no policy or guideline about this though. Today, I came across Misplaced Pages:Requests for adminship/Terenceong1992, and it's been added to Misplaced Pages:Unsuccessful adminship candidacies. I'd like to see what people's opinions are on deletion of never accepted nominations. Personally, I don't see the point in keeping them around. They are not posted to WP:RFA anymore, and do not contain really much in the way of useful information since they never received much public scrutiny. Thoughts? --Durin 16:01, 13 February 2006 (UTC)

Like a redirect someone will almost certainly never type in, they're cheap. They don't hurt to be kept around, and in the very very rare instance that they're required, they can be called up again. (Besides, don't people often look at past RfAs if a nominee has failed before?) Johnleemk | Talk 17:14, 13 February 2006 (UTC)

As a general rule I think an RfA would have to be 1) accepted by the candidate, 2) added to WP:RFA, to be counted as legitimate. But there are exceptions (i.e. an RFA added by a troll in bad faith). However, for historical purposes and the appearance of transparency I would like to have them remain undeleted and be able to see the links here of even those RFAs that are not legitimate. Maybe move them to a paragraph at the bottom of the page patterned after this). NoSeptember 17:31, 13 February 2006 (UTC)

I think they could certainly be deleted at the request of the nominee, if not by default. The problem is that they're catalogued at unsuccessful candidacies by the name of the nominee, when if they are useful information they are only useful about the nominator. Chick Bowen 02:33, 14 February 2006 (UTC)

Warn the creator, give him a chance to userfy or copy the page (for, say, a future nom) then delete... unless the prospective nominee suggests that they may be ready at some point in the future. I have an RfA I've been holding open for acceptance for a month, and will hold open for ten more if that's what it takes for the editor in question to feel ready. BDAbramson T 02:59, 14 February 2006 (UTC)

Hmm. Perhaps a distinction should be made between an RFA that hasn't been accepted and one that's explicitly been declined. Chick Bowen 03:06, 14 February 2006 (UTC)

As proposed above, you could make a list at the bottom of Misplaced Pages:Unsuccessful adminship candidacies (or, since that page is alphabetized, on a subpage of it). A good example of this was Misplaced Pages:Requests for adminship/Encephalon. He never accepted it, but eventually someone will nominate him sooner or later. So, the new nom could either overwrite the old one, or it could be on a new subpage. However, it's nice to be consistent about this. Titoxd 03:10, 14 February 2006 (UTC)

Administrator activity for January

Some of you may be interested in having a look at User:Durin/Admin activity. I started this as a trial balloon, and was rather surprised to find that the top twenty from each category were responsible for >50% of the activity. Enjoy. --Durin 19:42, 16 February 2006 (UTC)

A rider: - Care needs to be exercised in interpreting these statistics. For example, some one protecting/unprotecting main page images would score very high as in the case of Petaholmes or myself (I should be on 20th position along with Bishonen, dunno why I don't figure there) but I rarely protect disputed pages or vandalised pages (simply 'coz other admins beat me to it). The usefulness of this statistic lies in our ability to monitor a bulk of admin activities easily. For example, in the case of blocks, we can check if each of those blocks had a corresponding talkpage message or not. Great work, Durin. --Gurubrahma 10:59, 17 February 2006 (UTC)

It is interesting to note that because of User:Curps's automatic blocking bots, he has blocked more users than the next fifteen admins put together. JIP | Talk 09:25, 18 February 2006 (UTC)

Yes, I noticed that. I'm sure it's good faith but it also seems trigger-happy. Marskell 09:35, 18 February 2006 (UTC)

Well, look who he's blocking - hardy good faith editors... User:I love a block from JoanneB in the morning for instance. Or how about User:Fuckyouwhore? --Celestianpower 09:42, 18 February 2006 (UTC)

The population statistics are interesting:

• 2/3 of admins use their ability to delete or restore, but only 1/3 to protect and 1/2 to block;
• the top 10 and top 20 in each case account for over a third and over half (respectively) of the relevant admin actions.

I wonder how many of those 200+ "inactive" admins (like me, I guess) are away, or just writing articles. -- ALoan (Talk) 22:06, 19 February 2006 (UTC)

Who changed everything here to "vote"?

We are not voting here, this is very misleading Kim Bruning 13:29, 17 February 2006 (UTC)

Argh! Since september! I haven't been paying attention here, have I? :-( Kim Bruning 13:42, 17 February 2006 (UTC)

Well, so much for me thinking I've been doing a good job. I mean, drat. Ok, and now I just know I have adminitis. I'm outta here before I do damage. Kim Bruning 13:44, 17 February 2006 (UTC)

Freestylefrappe

For what appears to me to be the first time ever, we managed to promote an admin who was clearly unsuitable for the task, and who was then demoted by the AC mere months later. I am struck by the similarities between the RFA objections and the AC decision findings. I note that this is a clear example that the AC is doing its job, which is a good thing. I also note that this is a rare enough occurrence that it doesn't necessarily indicate a need for process change. All the same, I thought I'd bring it up here so that we could gain whatever insights we might from it regarding the adminship process. The Uninvited Co., Inc. 16:36, 18 February 2006 (UTC)

Could you clarify a bit more...? I don't quite understand your query. -Zero 16:40, 18 February 2006 (UTC)

An immediate point that comes to notice is that FFF was promoted on just 73%, well below the usual thresholding and noticeably outside the region commonly considered to be discretionary. I wonder did this happen because the notvotes after the deadline were discounted? If that's the case, we need to be crystal clear than anyone who expresses an opinion is entitled to have that opinion counted. -Splash 16:40, 18 February 2006 (UTC)

I believe I brought this point up in linuxbeak's rfa refomation project - rfa is not an "vote". The voters should always provide an reason for their stance in an matter. Perhaps votes lacking any explanation or thesis should be discounted. -Zero 16:46, 18 February 2006 (UTC)

Full agreement here, but those oppose votes had rationales, while on the contrary, most of the support votes had none. I say that "votes" after the deadline should be counted; if the 'crats don't close on time, it's their fault, not the "voters'" fault. Johnleemk | Talk 16:50, 18 February 2006 (UTC)

"but those oppose votes had rationales, while on the contrary, most of the support votes had none." - See Raul's 10th law of wikipedia for the reason why this is so. Raul654 06:57, 19 February 2006 (UTC)

Uh, that would leave 'Frappe and most RfAs with unanimous opposition. -Splash 16:52, 18 February 2006 (UTC)

In general, I think it's pointless to require justifications for support votes; we already know the justification for most support votes. If a user has been editing decently for a sufficient period of time and contributed enough to the encyclopedia to understand policy and to give people a chance to decide that they're not gibbering lunatics, then that's enough to support them as an admin. I'm not sure what would be gained by requiring people to essentially say that over and over... Certainly, there are cases where a support voter could cite a user's excellent work at such-and-such an article or in resolving such-and-such a dispute or whatever, or places where they might want to use their support vote to comment on issues that might lead other people to oppose, but those things seem to be the exception and not the rule; in general, we don't need specific reasons to make someone an admin. The basic idea, after all, is to give adminship to as many (sane) people as possible. --Aquillion 19:39, 20 February 2006 (UTC)

I'm not usually one for number crunching, but counting just pre-deadline comments it seems it was 77.5%. It seems kind of strange to say it (since 80% is after all, an arbitrary figure, right?) but this seems to strongly indicate that RfA works if process is closely followed. --W.marsh 16:48, 18 February 2006 (UTC)

His figures are based on 38/52 (support/total). This is correct if the intent is to gauge specifically the level of Support against the total entries. If you remove the three neutrals, it becomes 38/49 which results in the 77% support. - CHAIRBOY (☎) 16:53, 18 February 2006 (UTC)

No they're not. He had 38 support and 14 oppose. As is always the case on any other time-limited process, everyone who gets in before the closure is listened to. -Splash 16:57, 18 February 2006 (UTC)

I know, I was explaining to Wmarsh why you were getting 73% and he was getting 77%. - CHAIRBOY (☎) 17:04, 18 February 2006 (UTC)

Splash, while I understand and respect your views, I'm not sure that what you propose is the best answer. It sounds as if you're suggesting an 'eternal RfA'. At what point does an RfA actually close if post-close opinions are counted? Is it an hour after the close time? A day? Two days? Why isn't seven days long enough? Participation in the RfA process is a responsibility. There are 7 days allocated for the process, and it is the responsibility of those who contribute to, at some point during the week, add their entry. For them to demand to be heard after the period has expired shows a lack of respect for members of the project who are willing to abide by the widely published and openly viewable guidelines already established. Regards, CHAIRBOY (☎) 16:49, 18 February 2006 (UTC)

It closes when the 'crats close it. The RfA is supposed to end at a certain time, but if the 'crats don't close it then, it is their fault for not doing so, not the fault of those who added their opinions while the RfA was still open. Johnleemk | Talk 16:51, 18 February 2006 (UTC)

No, no. I mean that, if an RfA is closed slightly after the 7 day deadline, then anything that was posted until the close is fine (this is a statement of current practise, note). I in no way mean we keep going until there is a result, just that anything said until the closure is fine. -Splash 16:52, 18 February 2006 (UTC)

I agree. However, the bcrats should still take the validness of the vote into consideration after this time period as to make sure no one plays the system. -Zero 16:59, 18 February 2006 (UTC)

How do you measure the validness of a vote? How can expressing an opinion in good-faith ever be gaming the system? If the comment is in bad-faith, it's worthless whenever it is made. -Splash 17:02, 18 February 2006 (UTC)

I'm sorry. I was not clear enough; I was referring to an lack of clarification at all:

• invalid - Oppose/support-User:Bad faith

• valid - Oppose/support, user likes to eat gravy. -User:Good faith

That's what I was trying to say, sorry for the confusion. -Zero 17:06, 18 February 2006 (UTC)

I see. -Splash 17:23, 18 February 2006 (UTC)

Is this current practice? Wikipedia_talk:Requests_for_adminship/Archive_32#Votes_after_closing seems to suggest otherwise; Uninvited Co. provides a couple reasons why votes after the deadline should not be counted. Christopher Parham (talk) 21:04, 18 February 2006 (UTC)

On the more substantive point, I think that the part of RfA reform that calls for a discussion prior to the poll is useful in such situations. I personally would normally oppose an RfA where the problem exhibited in the oppose section are found, but I had supported days before that came to light. Whether I returned to reconsider or not, I do not remember, but if I had, then certianly I'd have had to give the opposition very serious consideration. This kind of change is the exception rather than the rule as Uninvited Company observes, but it's a good example of the exception in action. -Splash 17:23, 18 February 2006 (UTC)

No question about it. Voting is good for only two things: 1. Seeing if any consensus exists (if it's close to unanimous, there is consensus); 2. Determining people's opinions prior to a discussion. Otherwise, discussion trumps it, anytime. Johnleemk | Talk 18:05, 18 February 2006 (UTC)

I have to say, I'm not sure I see a problem here. I see a mistake, absolutely, but such mistakes have historically proven to be very rare, and no matter what solution we employ, mistakes will be made because we are all humans and we all err. I know I've done things on the wiki that I regret (although not many, thank God). The new system doesn't seem bad, but we can't go in pretending that we're never going to annoint a rouge admin if we just change systems, and if we do make a change it can't be for that reason. Lord Bob 18:15, 18 February 2006 (UTC)

Yeah, there is no problem. Might have not even been a mistake in promoting Freestylefrappe; as one can't always guess what kind of admin a person may turn out based on user comments. The current system works just fine, at least as well as it can work in an imperfect world. :) New complicated rules would just be a burden, and I doubt any benefits they may provide. Oleg Alexandrov (talk) 18:33, 18 February 2006 (UTC)

Yes, but DFA will likely reduce errors (just read the concept and you should understand why). We can't be perfect, but we should strive for it. And DFA isn't complicated; just discuss the nominee for a few days, and then after the discussion, "vote". Johnleemk | Talk 18:34, 18 February 2006 (UTC)

Freestylefrappe's promotion was right on the borderline, and I guess that shows why some of these folks get so many oppose votes. However, this and other cases show that perhaps the Arbcom is more likely to desysop a rogue admin. If that's true, I'd say the RfA process doesn't need too much of an overhaul, since part of the reason we were looking into it is because it was nearly impossible to get anyone desysopped. I wouldn't mind people asking additional questions, of course ;-), but I don't think RfA needs too much of an overhaul now that the Arbcom no longr sees desysopping as a taboo remedy. --Deathphoenix 21:20, 18 February 2006 (UTC)

Amen! Oleg Alexandrov (talk) 06:12, 19 February 2006 (UTC)

Well, it might not be that "ArbCom" no longer sees it that way, it may be that the community selected new Arbiters that always saw it that way. New blood type of thing. Changing of the guard. You get the point. --LV 04:19, 19 February 2006 (UTC)

Since the section header of this enthralling conversation is cited as "Freestylefrappe", does this mean that the fellow users here believe FF was an unsuitible admistrator..? -Zero 06:51, 19 February 2006 (UTC)

Just look at the evidence section of the ArbCom case against him. He protected his own talk page. I know we can't know how a user will use the admin tools until they have them, but this behaviour is everything I try to avoid when supporting an admin candidate, and it is indeed comforting to know that ArbCom will deal with bad admins to the same extent they deal with bad users. Raven4x4x 09:46, 19 February 2006 (UTC)

Arbitrators in touch with the current RfA standards?

I think how the Arbitrators think about their requiring a desysopped user to reapply is worth considering. You can get an idea of what they expect by reading this discussion. What they do impacts how we act here. I wonder how aware some of the arbitrators are about the current RfA situation. The standards for candidates expected by "voters" are quite different from those a year or two ago. NoSeptember 11:32, 19 February 2006 (UTC)

I'm not pretending to speak on the arbcom's behalf here, but I think anyone as active and as experienced as they are should be well aware that any election is a popularity contest, and not judging on the merits. Generally people decide whether to support or oppose first, and then find reasons to support their decision. Johnleemk | Talk 11:39, 19 February 2006 (UTC)

I don't know, Misplaced Pages is a big place, you can't stay current in all areas. In the section I linked above, the comments of Morven, Mindspillage and Sam Korn make me wonder how often they read the RfAs that are currently up for review. Do they know how many votes are cast on the issue of edit summaries for example? Btw, your comment in that discussion shows that you do know what is going on here, not surprising since we see you here often anyway. But what is this about a popularity contest - are you saying the ArbCom is ruling that if you are popular you don't get punished but if you are unpopular you do get punished? That would be an odd way to dish out punishments ;-). NoSeptember 11:53, 19 February 2006 (UTC)

Frankly, I don't think this is a problem with the arbitration committee, do you? Kim Bruning 12:06, 19 February 2006 (UTC)

I just want to make sure they are fully aware of the likely low RfA success rate of a desysopped-by-ArbCom user. I have no problem with desysoppings by them if they know the practical effect of it will be semi-permanent (It took Guanaco 4 tries and over a year to become an admin again). Being an admin is a privilege, so a bad admin being desysopped is a good thing. But if they intend a desysop to be only a temporary measure, before the community rushes to re-admin someone, then they may be deceiving themselves. And I do think ArbCom is doing a good job in general. NoSeptember 12:21, 19 February 2006 (UTC)

I think the ArbCom is aware that a desysopped user may not be given the mop back soon. And I don't think they necessarily mean things as a "temporary measure".

I strongly disagree with parts of the comment "any election is a popularity contest, and not judging on the merits. Generally people decide whether to support or oppose first, and then find reasons to support their decision." If you are implying that worthy people fail to get elected, but people who would be bad administrators but are "popular" do get elected, then I will ask for evidence. And implying that voters are naive enough to not reconsider their decisions/biases based existing information about candidates is just naive. Oleg Alexandrov (talk) 16:57, 19 February 2006 (UTC)

Um, there's some pretty silly admins out there at the moment, and there's also some very cool folks who failed. Right now I'm probably being stalked to bits though, so it'd be wiser for me not to name names. Ugh... how's that for a useless comment... maybe mail me or something. Kim Bruning 18:04, 19 February 2006 (UTC)

Being a silly admin is fine, as long as the tools don't get abused. :) As far as cool people failing to get elected, well... one could try again. If plenty of people think a given person is cool, I guess they will vote for him/her... :) Oleg Alexandrov (talk) 18:54, 19 February 2006 (UTC)

I never said this is necessarily a bad thing; most of the time, the set of popular editors coincides with the set of people who should be admins (IMO). There are occasional cases where elections fail, though (note: I was speaking in general about Misplaced Pages elections, not just RfA). This is especially true with elections that have higher standards than RfA, i.e. the arbcom elections (where people like JamesF have never made the cut without a little help from Jimbo). And if I really believed people won't change their minds, why would I support DFA, which is predicated on the presumption (which, from my experience, is correct) that most editors are capable of weighing the material evidence and then deciding? I'm just stating what appears to be a fact, IMO -- people decide whether they like/know X, and only then start weighing the evidence. (I really need to learn precision in the art of making remarks, because my original comment should read "any election is first and foremost a popularity contest".) Johnleemk | Talk 02:07, 20 February 2006 (UTC)

Actually, I'm fully aware of the situation here; I do follow this page. (Note that no one is required to reapply: they are simply allowed to; also note that I expressed some misgivings about the process on the vote page.) And yet—what else can we do when a case is about misuse of sysop powers? The idea of a requests for deadminship page continually gets shot down with the idea that it would only be a troll magnet and that good sysops who get on the bad side of bad users will get unfairly shot down, and where it has been shown that an admin has misused powers the matter can be taken to arbcom. Well, it was. If you don't like that it's too much of a popularity contest (and I don't, myself), make strong arguments against it. Have talks (not accusations, just talks) with people you think are harming the process, especially if they're your friends.

I would strongly urge that people would consider their decisions to accept or reject a re-request without holding the arbcom decision against them. That is, suppose the person asking for adminship again had done exactly the same admin actions, but hadn't been taken to arbcom: what would you think about that person retaining admin status? I'll repeat this as a comment on the RfAs themselves, should they choose to reapply. (Unless someone rebuts it well enough.) Mindspillage (spill yours?) 19:17, 19 February 2006 (UTC)

• Do we actually have evidence that it's happening like this? That is, can someone point to a re-adminship application that is clearly being shot down because of 'bad users' the candidate offended while being a good admin? Granted, de-adminship is so rare that there aren't very many examples to look at, period, but before we have all this discussion over it we should at least check whether it's really a problem or not. --Aquillion 01:06, 20 February 2006 (UTC)

Mathbot

I'd like to see Mathbot stopped from adding the usual bunch of statistical nonsense to the bottom of nominations. For starters, it promotes needless numeric criteria. We're not here to vote for the person with the most interesting edit summaries. We're here to discuss people who we think would find the administrator tools useful, and who wouldn't abuse them. If people want to impose numeric criteria, let them; but let them also go and get their own facts. Having them lurking about promotes bad feeling; RfA is enough of a mudslinging contest at times without having to accept that you missed an edit summary at some point. In addition, it means newcomers to RfA (who would otherwise put forward useful points) resort to using numeric criteria. Rob Church (talk) 18:55, 19 February 2006 (UTC)

Would it be feasible, I wonder, if Mathbot's computations were converted to a tool just like (or integrated with) Interiot's edit counter? That would still make the information accessible for those who feel it's relevant, but it wouldn't make it the first thing you see on a new nomination. (ESkog) 18:58, 19 February 2006 (UTC)

Well, provided it wasn't linked to, I suppose that would be OK. People would still be aware of it if it was useful to them. I just don't like the stigma attached to a lot of RfA's these days, and am working to cut it down; this is part one - beginning to remove the reliance on pure numbers. Rob Church (talk) 19:12, 19 February 2006 (UTC)

Now that there's assistance (in most major browsers) to ensure 100% completion of edit summaries, I think it is significant when an editor chooses not to give summaries, and absent an explanation why not, I'd tend to see that as a factor (a semi minor one, and one of many possible factors, mind you!!!) suggesting oppose. So I appreciate having the stats there in the nom so I don't have to follow links. It gives more time to spend looking at other things relating to the nom, which is goodness. Why make info harder to find? ++Lar: t/c 19:35, 19 February 2006 (UTC)

Just like to point out that I rarely give edit summaries. Probably I should for main article namespace, but for talk pages and user pages? I don't really want to and I don't see the need. - Ta bu shi da yu 02:07, 20 February 2006 (UTC)

Why not? I find edit summaries quite useful. Even on talk pages. In fact, especially there, sometimes, as the summary can give you a clue where to look in a long long section to see what new comment was added. For example THIS comment was added in the middle (to preserve threading) and the summary I used will help readers find it. I think that "because you don't want to do something" is not a reason not to do it. ++Lar: t/c 18:20, 20 February 2006 (UTC)

For the record, the bot looks only at the article and category namespace in calculating the edit summary usage, as those are most important (and I may remove the category, not sure). But I agree with Lar that edit summaries are important in general. Oleg Alexandrov (talk) 18:50, 20 February 2006 (UTC)

Per TBSDY, I'd like to say I almost never use edit summaries on talk pages. I strive to use them as much as I can in my article edits, but on talk pages they seem kind of pointless. If I'm editing Talk:Foo, section ==Bar==, it's plain to see I'm talking about Bar, which relates to Foo in some way. Szyslak ( ) 06:58, 23 February 2006 (UTC)

I had promised to myself to not get involved in this discussion, but can't abstain. :) I don't care a thing about the fate of my silly bot, but I am not happy with Rob's suggestion about his intent in "beginning to remove the reliance on pure numbers". How often a user puts edit summaries is a number, how many edits he/she has is a number, how many of those are in the Misplaced Pages namespace is a number, how many months the admin nominee has been here is a number, how many times a user reverts an article is a number, and so on. Surely the numbers don't tell the whole story, maybe not even a tenth of it, but they are of some importance, albeit each person has his own views on how much weight these are worth. If you remove any quantitative analysis, what are you left with? Oleg Alexandrov (talk) 20:17, 19 February 2006 (UTC)

The Mathbot results were probably the last straw that killed my first RfA, but I still support Oleg. The more objective results are provided, the more informed are the voters. If somebody would write a similar bot that would provide other objective summaries (number of reverts, number of articles ever reverted, number of AfD votes, number of edites per categories, etc. it would also help) abakharev 22:23, 19 February 2006 (UTC)

My own feeling is that the mathbot is useful but can be easily gamed. It gives edit summaries for last 150 major and minor edits - how difficult is it to notch that 100% over a period of 15 days (esp. given the fact that people tend to get active just before getting nommed)? I'd believe that we should have a look at edit summary usage throughout the history of the user - we should probably be ready to accept that we may need to be ready to embrace a lower % as the standard in such a case. --Gurubrahma 20:32, 19 February 2006 (UTC)

All numerical criteria can be gamed. For example, it takes only a couple of weeks to get more than a thousand Misplaced Pages space contributions by just voting with the majority in AfD's that have a couple of votes already. That's why the quality of contributions should always play a role, more than their number. Also, increased edit summary usage immediately before the RfA is only bad if the candidate stops using edit summaries after having been promoted. Kusma (討論) 20:42, 19 February 2006 (UTC)

• Eh, I've observed that a candidate can easilly appease the edit summary crowd just by installing software to force edit summaries. The candidates who've done this have all been promoted, as far as I know. The recent candidates who haven't, and instead have either challenged the opposers or ignored them, have all been unsuccessful. So it's not like someone is doomed to fail if they have a poor edit summary score, they can just pledge to change their ways and most people will be happy with that. Edit summaries are good, after all... --W.marsh 20:44, 19 February 2006 (UTC)

Numbers are important, but they are not critical. There is a problem when a nominee is analyzed completely with numbers. However, there is also a problem when a nominee is analyzed just by nepotism, so eliminating the statistics isn't the best solution. Titoxd 20:45, 19 February 2006 (UTC)

I wonder if it would be possible for Mathbot to provide a history, like Interiot's edit counter does? That would help answer the objection about gaming the system, I should think. Also, I'm curious if Rob Church is proposing eliminating all numerical criteria? I think that might be a good thing. However, as long as edit count is emphasized, usually as part of the nomination, I think that use of edit summaries provides another dimension to those who are inclined to weigh numbers heavily. To respond to ESkog, Mathbot is available in a manner similar to Interiot's edit counter.

Well, we provide a link to the edit count rather than transcluding the number in the RfA itself. It seems to me that a fair compromise would be to link to this report as well, unless server strain and other concerns would outweigh this. (ESkog) 21:15, 19 February 2006 (UTC)

Providing a link to the edit summary usage may give some strain on the server, as my bot now sheepishly goes through the user's contributions fetched with an HTTP request, instead of Interiot's tool which talks directly to the database. I will talk to Interiot about absorbing my tool into his edit count tool. Oleg Alexandrov (talk) 22:31, 19 February 2006 (UTC)

Of the current RfAs, I count 7 where the edit count is quoted in the nomination and 5 where it is not. The edit count for one of the 7 is in bold-face. For those 7, including the summary usage numbers explicitly helps to broaden the numerical picture of the nominee. I think Oleg Alexandrov's work is commendable. Walter Siegmund (talk) 23:25, 19 February 2006 (UTC)

When MathBot began doing the numbers, I noted that there would be significant flack. Well, above you can see I was right. Edit summary usage as an RfA metric has been and will remain controversial. In fact, all numeric metrics for evaluating a candidate have been controversial. As most of you know, I used to do charts for admin nominees with <2000 edits (example). I did a few dozen of them over the span of a few months. I was roundly attacked by a small, but vocal minority who were adamantly opposed to the charts. Eventually, I stopped making them. Interestingly, I later on looked at the success/failure of those RfAs compared with other RfAs in the same category that I did not do charts for. The comparison showed that RfAs with the charts had a ~10% greater chance of succeeding than those that did not. The point in my making the charts was to reduce people's utter reliance on arbitary numbers (999 edits is unacceptable, but 1,001 is...). It seems it was working. Interiot's tool has largely replaced the efforts I was engaging in. I think the general stance "number metrics suck" is flawed; they can and have helped. --Durin 14:57, 20 February 2006 (UTC)

I agree with this. Metrics are like any data, like any tool, they can be misused, or used wisely. Work to change the usage, not the tool availability. (when numbers are outlawed, only outlaws will have numbers!!!) ++Lar: t/c 18:20, 20 February 2006 (UTC)

DFA Discussion

The link got swept away during archiving. Please see WP:DFA do not let discussion die down. — Ilyanep (Talk) 21:48, 19 February 2006 (UTC)

Still seems to me that there's no point in fixing what's not broken. Where's the evidence that we're generally either producing bad admins or rejecting obviously qualified candidates? —Cleared as filed. 21:51, 19 February 2006 (UTC)

I can't name any immediate cases, but Freestylefrappe (see above) comes to mind -- would all the people that supported him have continued supporting had they known the incidents mentioned by the opposers? I also think the "Can't sleep, clown will eat me" RfA shouldn't have failed, but I may be biased, since I supported (I think). Johnleemk | Talk 02:12, 20 February 2006 (UTC)

Clearly in the user community there is some angst about the current set of admins. While there have been accusations that some such users are socks and/or trolls, etc., I think it is irresponsible and naive to say that all such users are. Immediately coming to mind are User:Splash, User:Tony Sidaway, and User:MarkSweep. I'm not really going to comment on their actions, but clearly there is presently a feeling in the community at large that the admin community is broken, and there should be a more stringent or effective process at culling the bad ones (yes, FSF comes to mind) before they get into a position where the harm cannot be reversed. aa v 02:26, 20 February 2006 (UTC)

Sure - but that doesn't indicate a need to change the process of gaining adminship - it's more a call for there to be a way to desysop people. If any process change is necessary, it's the addition of a RfD (Requests for de-adminship). But suggestion of the possibility of that sort of process seems to be taboo. Grutness...wha? 04:10, 20 February 2006 (UTC)

Immediately coming to mind are User:Splash, User:Tony Sidaway, and User:MarkSweep. Calling names is a poor way of making a point. Please abstain from that in the future. Thanks. Oleg Alexandrov (talk) 04:25, 20 February 2006 (UTC)

... aa:talk 18:08, 20 February 2006 (UTC)

In response to Johnleemk's comments above, I'd look at those cases as exceptions; considering the number of editors who go through the RfA process, a few failures once in a while is not unacceptable, and I'm not sure I see how CSCWEM's nomination would have been helped by the new proposed system. As far as User:Avriette's charge that current admins are being too bold with their admin powers, I note that a look back shows that two of the three admins named above breezed through their RfAs with almost no opposition, so this problem (if it exists) isn't a problem for RfA reform, it's a matter of coming up with some way to take care of problem admins. And we already have that in the form of arbitration. So again, what problem are we fixing here? —Cleared as filed. 05:28, 20 February 2006 (UTC)

That's a very valid point that I hadn't considered. I gather this has been brought up before? ... aa:talk 18:11, 20 February 2006 (UTC)

If DFA can help seal all or most exceptions, all the better. Johnleemk | Talk 05:48, 20 February 2006 (UTC)

But why do you expect it to seal those exceptions? What was it about the FSF RfA or the CSCWEM RfA that the new proposal would change? —Cleared as filed. 06:25, 20 February 2006 (UTC)

See above: "Freestylefrappe (see above) comes to mind -- would all the people that supported him have continued supporting had they known the incidents mentioned by the opposers?" If even one rogue sysop is never admined because of DFA, we'll already have saved ourselves the trouble of one RfAr and a bunch of angry posts on WP:AN, IRC, and the mailing list. Johnleemk | Talk 14:04, 20 February 2006 (UTC)

Agreement with Cleared as filed. ANY system we design will result in some people becoming admins that in retrospect probably should not have become admins. This can't be avoided. There is no 'perfect' system. Citing less than five 'problem' admins and asserting a (unsupported) "feeling" by the community that the admin community is broken is not illuminating of the problem. In fact, I think it rather shows that RfA is working quite well, if only a small handful of admins can be identified as being "problem" admins. --Durin 13:51, 20 February 2006 (UTC)

No process is perfect. We promoted 389 admins in 2005 (source), three of those have been desysopped with restoration by RfA required (Freestylefrappe, Carnildo and Karmafist). You won't get much better than less than 1%. NoSeptember 14:56, 20 February 2006 (UTC)

I read that the percentage supporting the above mentioned Admin was lower than usual. If that is the case, you have your problem (and maybe your solution.) Does anyone have the stats on the number of admin taken with lower percentages. It's logical that over time more of these admin would be problematic. Sorry if this offends the discuss don't vote folks. : ) FloNight 14:36, 20 February 2006 (UTC)

Here's a list of RfAs since June 23, 2005 that had less than 80% raw vote support and passed. The numbers are support/(support+oppose). Neutral votes have not been taken into account.

Nightstallion 79.7%

BorgHunter 79.3%

Aaron Brenneman(2nd) 78.6%

Lucky 6.9(4th) 78.3%

Nandesuka 78.2%

The Land(2nd) 77.8%

Alkivar(3rd) 77.5%

Extreme Unction 77.2%

Ramallite 77.1%

EvanProdromou 76.5%

Hedley(2nd) 75.8%

Johann Wolfgang 75.5%

Freestylefrappe(2nd) 74.5%

Luigi30(3rd) 72.4%

--Durin 15:58, 20 February 2006 (UTC)

A great list Durin. I hope you don't mind, I added the bit about 2nd, 3rd and 4th attempts to your list, since that seems to be a factor too. NoSeptember 16:16, 20 February 2006 (UTC)

For comparison, the other two mentioned were Carnildo (RfA, 40 support, 4 oppose, 2 neutral, 90.9%) and Karmafist (RfA, 53 support, 2 oppose, 96.4%). -- Jitse Niesen (talk) 20:10, 20 February 2006 (UTC)

• 2 out of ??? for 80% or more since June 23, 2005
• 1 out of 14 for less 80% since since June 23,2005

Looks like a indicator to me. FloNight 20:33, 20 February 2006 (UTC) What was the number Admin given since June 23 2005? (minus 14 less than 80%)FloNight 20:37, 20 February 2006 (UTC)

• From the same data: 531 RfAs since June 23, 2005 have completed. 342 were successful (64.4%). --Durin 22:29, 20 February 2006 (UTC)

I think that we're also confusing the acute problems (like Splash ;P you know) with the chronic ones, and trying to apply one solution to both of those problems. We're also using one metric to measure both of these.

• We're never going to detect every potential acute problem admin early. That will only be solved by either "trial periods" which add a huge load and wastes time for the >95% that will pass or request for dead-minning/reconfirmation. I see Durin's numbers as supporting this: low rate, poor entry metric. This isn't the problem this is intended to address.
• The chronic problems of "worthy" candidates missing out and "unready" candidates getting promoted is harder to measure. In fact, I don't see any way to mesure the former. For the latter, I've proposed looking at a count of how often admin actions are reversed as a way to get a handle on the second. This is what this is meant to improve.

The current DfA proposal is a very slight tweak on the current system. It amounts to making the nomination a collaborative effort, while retaining every other aspect: the voting, the way it's closed, everything. The potential for harm is very low, and even if it does nothing but make people feel that the system is better it will have provided someincremental improvement.
brenneman 00:19, 21 February 2006 (UTC)

I disagree that we ought to change a system that is, by all evidence, working very well, just because it might make some people feel that we've improved even though we haven't. Sure, the risk is low, but why take even a low risk just for its own sake? Why do we have this solution in search of a problem if so many people supposedly think RfA is broken? Why can't anyone pinpoint exactly what it is that we need to fix? —Cleared as filed. 01:13, 21 February 2006 (UTC)

Ah, here's one case I've had in mind for quite some time. Look at Misplaced Pages:Requests for adminship/Lulu of the Lotus-Eaters. Right until the RfA closed, there was a lot of discussion on the RfA about Lulu, and quite a bit of vote-changing here and there because of this discussion. Most RfA voters (well, at least me) don't look back at RfAs we've voted on, so for all we know, Lulu could have gained (or lost) more support votes if DfA (where all that discussion occurs before voting) was in use. Lulu even lost a few votes because he "campaigned" on people's user talk pages when he was just trying to point out new developments in the discussion or show why the reasoning for opposing could be faulty. All this could have been averted under DfA. Johnleemk | Talk 04:18, 21 February 2006 (UTC)

Well, that's one. ...of >500. Single point failures is not enough to condemn RfA in my opinion. Also keep in mind that RfA is not a one time shot; people are welcome to reapply at a later date. Lulu can reapply. --Durin 14:02, 21 February 2006 (UTC)

Nobody is condemning RfA. DfA is an extension of RfA, and is largely based on it. It only splits it into two sections; one for discussion, and one for voting. You just discuss for a few days, and then go on to vote. Just like RfA, except you're a lot less likely to need to change your vote. And this isn't a single point failure -- this is just one case I can think of. My failed RfB might also be a good example. And I've already brought up Freestylefrappe. If I had the time to dig through the archives, I could find more. Johnleemk | Talk 14:25, 21 February 2006 (UTC)

• My arguments against the DfA proposal as it now stands are on the discussion page for it. DfA might be a great idea, and it might not. The main problem I have is the manner of the genesis of the idea. It's a blind shot into the dark, and just as likely to cause harm as not. --Durin 14:33, 21 February 2006 (UTC)

I stated my own opinion on DfA sometime ago on that page also (concur with Durin). This discussion can't go forver, seems quite a bit of people are already familiar with what is going on and made up their minds. There's got to be some kind of poll to gauge how community feels about adopting or not this new DfA thing. Oleg Alexandrov (talk) 17:06, 21 February 2006 (UTC)

We could do a poll, but the choice of trying DfA out for a trial run should be an option. We can continue this theoretical debate forever without conclusion. If we had a dozen real DfAs, then we could compare the two methods and decide on the results. NoSeptember 17:30, 21 February 2006 (UTC)

de-indentingJohnleemk has used an example, which, imo, does not still establish the need for DFA. Also, it is a dangerous idea. As the adage goes, "A man who doesn't change his mind has no mind at all." What DFA assumes is that after discussion, voting would take place. What if, after the discussion period, I find something in the candidate's history that is detrimental for his becoming an admin. The current rfa is much better in the sense that some rfas can swing from one end to another with availability of new evidence. DFA does not provide for that and unless discussion takes place, in several situations, it is difficult to even establish that a particular behavior is egregious. Also, a community gets the leaders it deserves. People would and should change their votes as it happens now, with the availability of new evidence and discussion of that evidence. If the concern is that people change votes, I am sorry, it is healthy for the process. If the concern is that people do not change votes with availability of new evidence, I cannot see how DFA would solve the problem, because new evidence can be available even after the discussion period is over. There may be some flaws and fads with the rfa system but we can easily live with these. --Gurubrahma 17:41, 21 February 2006 (UTC)

We use common sense. The point is RfA will reduce such vote-switching because many people who would have switched their vote won't need to — by having seen most/all of the evidence, their mind can change several times without requiring constant flip-flopping and constantly keeping up with the discussion. If suddenly some new point of evidence to oppose comes up, then it should be perfectly fine for someone ('crats? They don't have enough work as it is ;-)) to notify every voter on their talk that something has come up. Since most discussion (and thus most of the muckraking) should have already been completed, such incidents should be uncommon. Your argument is predicated on the misconception that DfA is about reducing how much people change their minds. It's not. It's meant to encourage people to change their minds by having all the evidence neatly compiled prior to voting, so they don't need to flip back and forth between support/oppose/neutral. Johnleemk | Talk 18:01, 21 February 2006 (UTC)

This process *really* needs to be changed.

The entire thing is too subjective and there exists too large a faction that will simply vote oppose no matter what (e.g. user:Radiant). I think that the entire process to become an admin should be changed so that it's more objective and more than .8% of users can become admins. Here's how I think it should work:

The user wanting to become an admin presents what he/she feels is his best work on wikipedia and writes about that and why he/she would make a good admin. He/she then has to get a certain number of "endorsements" from users he or she has come across in the encyclopedia (say 10-15). After the required number of endorsements is met, other users can sign the comments section regarding their opinion on the candidate. Finally, the bureaucrats, taking all "evidence" into consideration, have the support/oppose vote themselves.

Infinitely more fair, no? Juppiter 08:34, 22 February 2006 (UTC)

What you propose is more similar to how it actually works than you think. Bureaucrats already have discretion, and there are no hard-and-fast rules for how large a majority is required for them to promote someone—so if a user is being blocked for what seems not to be a good reason, they'll be promoted anyway. There are already descriptions of what administrative jobs a user would do, and "endorsements" in the form of Support votes. The only things that are different from what you say are that a) only one Bureaucrat makes the decision, because it would be a waste of time to require several to vote against each other, and b) we really don't need a large number of new admins. -- SCZenz 08:48, 22 February 2006 (UTC)

If we were to call these folks "trusted users", which is accurate (and wikinews is thinking of switching to that name) , well, do you really want large numbers of *untrusted* users let loose on the site? Indeed. Kim Bruning 10:06, 22 February 2006 (UTC)

I'd really oppose that term "trusted users". If anything the term should be made somewhat unflattering, like "bureaucrat" already is. I think the term "bureaucrat" helps dissuade people from trying to become one for the wrong reasons, and is thus a good thing. Maybe "admin" could become "janitor" or something. Phr 14:21, 22 February 2006 (UTC)

I disagree that we have a large faction who oppose no matter what. Radiant, who you mention, had pretty strict criteria for supporting, stricter than mine, but he supported several candidates, some of them contentious like Aaron Brenneman and Hermione1980. Sjakkalle (Check!) 11:14, 22 February 2006 (UTC)

Is there some sense that there's currently an admin shortage? If there's enough admins now, why look for ways to speed up the process of making more? Why not slow it down, even? Is there really .8% now? That's something like 8000 admins, a pretty big number IMO. Phr 12:31, 22 February 2006 (UTC)

No we are not at 8000 admins, according to Misplaced Pages:List of administrators we have 824, 683 of which are "active", 93 "semi-active" and 48 "inactive". I believe that the bulk of admin work is carried out by a hundred or so admins. Take a look at RC patrol some time, nonsense does slip through the cracks there, so there at least I think that we are understaffed with admins as it is. Sjakkalle (Check!) 12:36, 22 February 2006 (UTC)

I heard there was about a million users, so .8% would be 8000. I guess it should say .08%. RC patrol can be done by anyone (I've done it a little) and doesn't need admins. Non-admin RC patrollers can put notices at WP:AIV and they get handled pretty fast. So stuff slipping through RC patrol doesn't indicate lack of admins. Does anything else? Phr 14:14, 22 February 2006 (UTC)

Here I am butting in again with my "2 cents worth". Normal editors should be able to apply for rollback to help with vandalism. But admins are often looked to to settle disputes, more and more often. The criteria should be a bit tighter there. Perhaps Admin level 1 and 2, who knows. But there seems to be a lot of wheel warring lately, and people who are trusted shouldn't engage in this. Perhaps wheel warring should carry a stronger penalty. Also: Admins are often not being held accountable for what they do, and at articles like Brian Peppers, have created masses of conflict by arbitrarily bypassing procedure. I think this is going to have to be altered, as there should be another level between "trused users" and Bureaucrats. Trusted can be immature to some extent, but administrators need to be a little more firm in the maturity department. So we have more complicated, beaureaucratic policy on one hand and chaos on the other. We'll likely have to choose the lesser of two evils. --DanielCD 16:16, 22 February 2006 (UTC)

IMO, we should just create more 'crats and give them the authority to butt into wheel wars, etc. Johnleemk | Talk 16:22, 22 February 2006 (UTC)

Below I didn't mean that 'crats shouldn't get involved. Just that admin have the best chance to stop conflicts early. FloNight 19:08, 22 February 2006 (UTC)

Peer to peer is much better. Calling in 'crats may increase the tension in admin to admin disputes. Far better if they can work it out between each other with the least intervention possible. Two or 3 Joe Average's giving *quality* feedback is better than one 'crat. After new admin are comfortable with the nuts and bolts, they might actively practice dispute resolution. It is a learned skill. FloNight 17:06, 22 February 2006 (UTC)

There needs to be editors, then basic trustee level, then admins. Amin should be strengthened a bit, and reduced in number slightly, or at least have entry tightened. Self-nominations should just go. --DanielCD 19:31, 22 February 2006 (UTC)

If we were to change the way groups of permissions are granted, the wisest thing we could do is limit the blocking of logged-in users. Blocking and unblocking of logged-in users (not IPs) has been a major part of nearly all serious conflicts involving admins, particularly considering that its effective and proper use is fairly rare compared to deletion, protection, and blocks of IPs, all of which are done routinely. I'd like to see us grant that priviledge fairly routinely but only after someone has already been an admin for a considerable time. The Uninvited Co., Inc. 20:01, 22 February 2006 (UTC)

I disagree. While most of our block-worthy vandalism comes from IP addresses, there is a sizeable amount that comes from logged-in users who create accounts just to vandalize. I don't think the community would be served by having fewer people able to perform that essential part of fighting vandalism. (ESkog) 20:50, 22 February 2006 (UTC)

Totally with ESkog on this. Some vandals have gamed the WP systems so well that they register to indulge in page-move vandalism and image vandalism. Also, if a registered user's account is hacked and the hacker is vandalising, a block is immediately called for. Also, it is a good idea to dig the archives before saying things like "self-noms should go" as it has been discussed in the recent past and the consensus was against dis-allowing self-noms. --Gurubrahma 11:19, 23 February 2006 (UTC)

It might also be a good idea to take a good look at admins who were selfnoms to see if there's anything whatever less trustworthy or trusted about them. What's the point of having a system where self-nomination is obligatory for bureaucrats and outlawed for admins, other than the pure pleasure of legalism? We should try to trim the rule system down to a tight package instead of encouraging it to sprout into such luxuriance. Extra bishpoints for selfnoms! | talk 15:12, 23 February 2006 (UTC).

I'm a self-nom, and aside from applying the mischief rule instead of the literal rule, I don't think I'm a rogue admin. Johnleemk | Talk 15:18, 23 February 2006 (UTC)

Self noms are fine when done by a user who has been here long enough to understand the system. Its the newbies who nom themselves at 150 edits that makes self-noms look bad. So we should strongly urge getting yourself nominated to reduce the later, but ignore self-nom concerns about the former. NoSeptember 15:26, 23 February 2006 (UTC)

user:swatjester's "no big deal" supports

I am concerned at the macro-esque voting by this user. We recently wholesale discounted votes by a user during the arbcom elections who was making the same vote over and over again without explanation. Is this okay to do? It's been going on for weeks now. ... aa:talk 08:38, 23 February 2006 (UTC)

Weeks eh? Nice hyperbole Avriette. Still holding a grudge against me? Does my opinion not count here? Are you telling me that I can't vote for half a dozen or so admin candidate the way I want? Welcome to censorship at it's finest here. Administratorship is SUPPOSED to be no big deal. I point you to this . Finally, Lets look at a grand total of how many times I've edited the RfA page shall we? Oh look...I made 10 edits yesterday to the page. 10 votes is hardly "macro voting" for "Weeks now" considering I've made a grand total of 20 administrator votes my entire time here, and the last one was "2006/02/07 17:02:27" and the last one before that (was only 1) was "2006/01/11 07:58:06". You know, I'm really quite a bit offended by this Avriette....⇒ SWATJester Aim _Fire! 22:39, 23 February 2006 (UTC)

Ironically Avriette, (or should I say pot) your kettle looks in need of a paint job.....Nearly 70 votes on the Jan 2006 arbcom elections and you question MY voting here? ⇒ SWATJester Aim _Fire! 22:47, 23 February 2006 (UTC)