Wikipedia talk:Requests for adminship/Archive 152

Proposal: Dedicated Purpose Administrators
Technically, I don't think it's possible to restrict administrator rights in the software.

However, we can do it by policy and policy enforcement without changing the software.

I'd like to propose that we allow dedicated purpose administrators where an RFA candidate will specify a very limited list of what he needs the tools for and a limited time period in which to use them - probably 1-12 months - as part of his RFA. Part of the RFA would include a responsible party or parties who could block the account and recommend desysopoing should the DPA exceed the RFA. The responsible party or parties should be established administrators with a habit of logging in almost daily. Of course, any administrator could do the blocking and recommend desysopping, but having a responsible party makes it clear to both the candidate and the world that the rights restrictions really are being enforced.

The RFA would be convertible to a regular RFA or renewable upon request, although those needing long-term use would probably be expected to just run for full a RFA eventually.

With such a mechanism would make Requests for adminship/lustiger seth almost a no-brainer, or at least it would greatly accelerate his ability to edit the en-wiki spam-blacklist. davidwr/ (talk)/(contribs)/(e-mail)  20:54, 18 December 2008 (UTC)0


 * What prompted this proposal? Without an explanation I fail to see the need for this. Basically we open ourselves to a nightmare in enforcing such limitations. ≈ jossi ≈ (talk) 20:57, 18 December 2008 (UTC)
 * I would assume to suggest it is the current de. admin whose RfA is currently running. Caulde  21:01, 18 December 2008 (UTC)
 * (ec)This is an interesting idea, but I think the nightmare, as jossi puts it, of enforcing "no, you stepped over the line" is really not worth creating. "You said you wanted to only delete stuff at AfD, but you deleted my userpage in MfD! You should be desysopped!" Basically while this is an interesting concept, as there's no way to segregate admin tools by subject area, this is likely to produce more heat than light. For admins to say "I want the tools to do X" is good; to try and enforce that would be drama of epic proportions. // roux    21:03, 18 December 2008 (UTC)
 * I agree that if it can't be coded, it's probably not worth the hassle. Breaking up Administrator duties into individual authorities (as is done with Rollback) might be interesting, where a new Admin only receives the tools they specifically request in their RfA.  "Administrators extraordinary and plenipotentiary" (the current role) would be a separate tier and limited admins could be "autoconfirmed" to that status after, say, three months unless they were specifically authorized as a limited admin.  Would require code changes, and I'm not sure that it fixes anything, but it's really a question of how difficult it is to have finer distinctions in the access permissions.  If it would take 8 hours to program, it'd be worth doing.  If it would take substantially longer, it's probably not worth it.  SDY (talk) 21:15, 18 December 2008 (UTC)


 * Requests for adminship/Herbythyme. It's quite possible for someone to pass RfA under the banner of "I only need to do a little bit of sysop work here." One current candidate not going a specific way does not warrant a knee-jerk reaction. EVula // talk // &#9775;  // 06:47, 21 December 2008 (UTC)


 * See Perennial proposals. Prodego  talk  17:20, 22 December 2008 (UTC)


 * It's technically possible, but I doubt Bryon's gonna break open MediaWiki for this.   CWii ( Talk  01:36, 23 December 2008 (UTC)

Proposal: Split adminiship rights
Is there a general feeling pro- or con- for splitting up some of the rights held by administrators, so you could, for example, block or unblock a user without having the rights related to protecting, overprotecting, and editing fully-protected pages? This came out of the Requests for adminship/lustiger seth discussion as well: His RFA would probably go smoother if his future rights did not include user-blocking. davidwr/ (talk)/(contribs)/(e-mail)  21:20, 18 December 2008 (UTC) Update: This would of course require a change in software and in procedure. davidwr/ (talk)/(contribs)/(e-mail)  21:23, 18 December 2008 (UTC)
 * WP:PEREN--- Balloonman  PoppaBalloon 21:21, 18 December 2008 (UTC)
 * Yeh.. we just did the discussion on that about two months ago, and while debundling is an obviously logical and intelligent idea, inertia is against it. // roux    21:22, 18 December 2008 (UTC)
 * Is that a "yes, it's time to get it off the perennial list and just do it" or "no, it's been discussed to death?" Seriously, just because it's brought up time and time again doesn't mean it's not timely now. davidwr/  (talk)/(contribs)/(e-mail)  21:23, 18 December 2008 (UTC)
 * Very much WP:PEREN per 1) technical issues 2) no agreement on what rights should be split 3) view deleted edits a bit of a deal breaker for many and 4) Either you're trusted for the whole lot or you're not. Don't get me wrong, I support breaking out the bits, but I've seen this discussion maybe 5/6 times in full on this baord alone (and 15/18 times in part) over the last two years. I wish it were otherwise but consensus on this seems unlikely (ahhh..... the safe haven of futility!) Pedro : Chat  21:43, 18 December 2008 (UTC)
 * Assuming the technical issues could be dealt with, "edit fully protected page" or even "edit fully protected page(pagename)" or "edit fully protected page(namespace)" seems like a relatively safe thing to give to trusted users who need it, especially when the page is outside of article-space. This would be particularly useful in Template: space.  Other than Village_pump_(technical) where's a good place to start a discussion on this?  davidwr/  (talk)/(contribs)/(e-mail)  21:49, 18 December 2008 (UTC)
 * I wholeheartedly agree. Move-over-redir, too, especially for cleaning up nastiness from He Who Shall Not Be Named. Not sure if time is ripe for reproposing.. the last version of this discussion happened here and at VP, within the last couple of months. And sort of just petered out in a deadlock. Perhaps a nicely drafted RfC with links from the relevant policy pages? Suggest each adminright to be debundled (not delete, undelete, viewdeleted; Godwin has ruled that those can only br granted to admins after a full RfA) with endorse/oppose sections? //  roux    21:56, 18 December 2008 (UTC)
 * Wait, what? What's Godwin got to do with this? Also, where did he make this "ruling"? Glass  Cobra  09:05, 19 December 2008 (UTC)
 * He stated that for legal reasons, undelete and viewdeleted must be granted by an RFA process. Somewhere on VP I believe. // roux    16:27, 19 December 2008 (UTC)
 * See here. -- Amalthea 17:39, 19 December 2008 (UTC)
 * There is no legal reason why "an RfA process" and only "an RfA process" is necessary, it would be perfectly possible to have a separate process just for granting those permissions; Godwin's concern was that it would be given out to many more users, not a problem if the process is just as restrictive as the current adminship request process, which it inevitably would be -- Gurch (talk) 02:06, 20 December 2008 (UTC)


 * Splitting rights and dedicated/partial rights - Isn't it the same thing other than the social and technical functions? Why two sections? - Mailer Diablo 21:57, 18 December 2008 (UTC)


 * Given our ever-increasing size and the proportionally large backlogs of stuff, unbundling the tools increasingly sounds like a good idea. — sephiroth bcr  ( converse ) 09:26, 19 December 2008 (UTC)
 * Our largest backlogs (such as All articles lacking sources at 156,186, or All articles with unsourced statements at 131,275, or the smaller All pages needing cleanup, weighing in at a paltry 41,474) have nothing to do with adminship or its variety of userrights. EVula // talk // &#9775;  // 17:47, 19 December 2008 (UTC)

Possible problem with Interiot edit counts
Based on looking into this Help_desk, I think Interiot's wannabekate tool may consistently overstate Mainspace edits for the data used on candidate RFA talk pages. This of course is no big deal (unless someone gets opposed for not having enough Portal: edits :), just letting it be known. It may be helpful to run the real Kate, which is deadly fast btw, as a sanity check. Further technical details are at the help page thread. Franamax (talk) 20:30, 21 December 2008 (UTC)


 * I think it is looking for edits in the "Image:" namespace, finding zero, and counting the unrecognized "File:" namespace as article edits. — CharlotteWebb 20:58, 21 December 2008 (UTC)
 * That's a fair point, I guess wannabekate is now officially broken no matter what. But it doesn't explain the discrepancy of 2400 edits for Aitias, whereas the miscounting of User talk: does. So not only is the space-naming wrong, I think there's also a problem with the regex's that parse the article name, however, I don't speak that dialect of regex, so I can't be sure. Franamax (talk) 21:19, 21 December 2008 (UTC)


 * Usercheck-short hat been updated with X!'s counter, which I think is a pretty good replacement. I think instead of real kate we should use that one as it allows the same per month overview as interiot's. Regards  So Why  22:08, 21 December 2008 (UTC)
 * With the added benefit of having an active maintainer  X  clamation point  00:52, 22 December 2008 (UTC)
 * I only use wannabekate because it does those green/red bars with summary/no summary color-coding, is there another tool that does something similar? Foxy Loxy  Pounce! 00:47, 22 December 2008 (UTC)
 * Unfortunately, no. I'll see if I can add that to my edit summary calculator though. X clamation point  00:53, 22 December 2008 (UTC)
 * I'd agree with X!'s being the tool of choice, it's slick, fast and colour-y :) wannabekate OTOH is much more Christmas-y with the red and green. Based on the apparent deficiencies with wannabekate though, it looks like it should be widely deprecated. I'll raise it at VPT, and even though I think last time I asked Interiot about it they said "leave me alone", I'll approach them to ask for a disclaimer on the tool. Franamax (talk) 01:28, 22 December 2008 (UTC)
 * Needs to be re-written to use the API anyway. It would be much faster and not rely on the current name of each namespace, see . — CharlotteWebb 01:46, 22 December 2008 (UTC)
 * Yes, it needs to use the API for speed, and it could also use it to get the interwiki map, which would help it locate the servers and also help it find |namespaces the namespaces and their names, as I'm doing for the WMF-wide version 0.5 of uContribs. ;) Franamax (talk) 02:05, 22 December 2008 (UTC)
 * Yeah the names of the namespaces would only be needed for a pretty presentation (and could be cached in the source code itself, as rarely as they change). Easier to identify and tally edits by ns number anyway. By that I mean who cares if it still says "Image" as long as the numbers are correct. — CharlotteWebb 03:43, 22 December 2008 (UTC)
 * If you look at the source code of my edit counter, you will see that it does use the API to get the namespaces. X clamation point  04:05, 22 December 2008 (UTC)

Since Mathbot uses wannabekate for it's talk page edit count, I have notified Oleg Alexandrov here. Foxy Loxy Pounce! 03:11, 22 December 2008 (UTC)


 * I can make Mathbot dump the real Kate's output instead of the wannabe Kate output. The real Kate's output is just a summary though, does not have the details. Whatever you guys decide. I'll visit later this page and modify the bot based on the comments here. Oleg Alexandrov (talk) 04:45, 22 December 2008 (UTC)


 * I think that if a user creates a User Talk page, such as to add a vandalism warning for an anonymous editor, it will be added to the Mainspace edits for Interiot's edit counter. Gary King  ( talk ) 16:29, 23 December 2008 (UTC)

I can't speak for anyone else, but I just ran them both and got the exact same number. So no problems for me, AFAIK. 16:41, 23 December 2008 (UTC)
 * I ran them and got the same thing, too. – Alex43223T 11:26, 27 December 2008 (UTC)

A CSD Survey
Well, I've gone through a number of CSD nominations from the past month and found about 40 that I thought might pose interesting questions on how people perform CSD's. Since CSD is such an important part of RfA, I figure I should announce it here as well. Basically, I'm asking people to review the article in question and answering the question, "how would you handle this" with one of four options:

1. Agree with criteria for deletion.

2. Disagree with criteria for deletion, but would delete the article under another criteria.

3. Disagree with the criteria for deletion, but this is a situation where IAR applies.

4. Disagree with speedily deleting the article.

To see the surveys, go to this page. I'm hoping to get a good mix of people to participate in the surveys---people who agree with my interpretation of CSD and people who have different views. I'll post the results in a couple of weeks after getting a decent return.--- Balloonman  PoppaBalloon 01:28, 28 December 2008 (UTC)
 * Is it just me, or I seem to find almost all of them CSDable but tagged with the wrong criterias? - Mailer Diablo 01:41, 28 December 2008 (UTC)
 * Well, I guess we'll see when BM releases the data. I think I remember having 3/8 wrong criteria speedied deleted, 1/8 or so correct, and the rest speedy merged. NuclearWarfare  contact me My work  01:50, 28 December 2008 (UTC)
 * Just done mine, and realized that my answers are "blind". Looked over the other pages after the survey and realized that Balloonman has written his own assessment over the four pages. :o) - Best regards, Mailer Diablo 01:57, 28 December 2008 (UTC)
 * Actually, only survey 3 had anything from my various 'assessments.' They were kind of the inspiration behind the full fledged survey.  The first, second, and fourth surveys are completely new.--- Balloonman  PoppaBalloon 05:38, 28 December 2008 (UTC)

The Full Faith and Credit Clause
Ok, I think we are witnessing an interesting scenario with Lustiger Seth. Seth is an admin elsewhere, which indicates that he is a trusted user on two other projects. He has earned and maintained that trust, which is ultimately the core of what it means to be an admin. He might not have many edits here, which makes it difficult for us to judge. But, he has shown elsewhere that he can be an admin.

Which leads to an interesting proposal. Why don't we open the doors to admins from our sister projects? I'm not suggesting that we open them to all projects, but have three possible proposals:
 * We will recognize the adminship of anybody who is an admin on a sister projected with over 500K articles. (This would eliminate the small wiki's with virtually no standards.)
 * We will recognize the adminship of anybody who is an admin on Meta or the Commons.
 * We will recognize, as an admin, a 'crat from our sister projects.

OR
 * We will recognize the adminship of anybody who is on a sister project with over 500K articles AND an admin on either Meta or Commons.
 * We will recognize, as admin, anybody who is a 'crat on a sister project with over 500K articles or Meta or Commons.

Yes, this does require an assumption of good faith on our part and allows for backdoor entry into adminship here, but this concept shouldn't be unfamiliar to most people.--- Balloonman  PoppaBalloon 14:59, 18 December 2008 (UTC)


 * I believe the infamous Poetguy was a crat and checkuser on Wikiquote while being banned from en.wiki, that several admins who have resigned under a cloud at Wikipedia are administrators at commons, and that several administrators at meta and commons have failed RFA at en.wiki. How would your system account for that?  MBisanz  talk 15:03, 18 December 2008 (UTC)
 * I'm a bureaucrat on Meta, and was until recently an admin on Commons when I resigned due to inactivity (but I could get it back just by asking). I very much doubt I'd pass any kind of adminship here :-) I think we should just take candidates as they come, on their own merits.  Majorly  talk  15:06, 18 December 2008 (UTC)
 * You probably wouldn't, but believe it or not, I would support you if you ran.--- Balloonman  PoppaBalloon 15:10, 18 December 2008 (UTC)
 * You would also have 'crat discretion. Make it parallel the return of an admin to the project who has voluntarily given up their bit.  As a general rule, somebody who voluntarily gives up the bit will get it back, but it does require crat discretion.  Or we can have them run through a cursory RfA, where the expectation is "Does anybody have any reason to oppose."--- Balloonman  PoppaBalloon 15:09, 18 December 2008 (UTC)
 * I'm not seeing it, to be honest. For one thing such a wide-ranging proposal would be next to impossible to get consensus on. If we're going to go with it I'd suggest we still need to do some checks; what about admins one step away from being de-sysoped, or those with a controvertial record? There's also a problem of comparing en-wiki and the "sister projects"; some of them have very different policies and standards to us (although I'll counter my own point here by saying an admin should be able to read up on ours before doing anything too controvertial). Since you're quoting law at us (and fyi, it may be unfamiliar to most people outside the US, of which there are many on en-wiki) maybe you could take the current RfA as something to set Precedent; see how it goes and use that to evaluate the consensus of the RfA crowd re: admining sister-project admins. Ironholds (talk) 15:11, 18 December 2008 (UTC)


 * I am impressed with the input at Seth's RFA, and this of course helps nail the "mandatory minimum edit count" thing (assuming it passes which looks probable at the moment). I sincerely hope it does pass just to show why we don't have infelxible rules. On the other hand, I've long been of the opinion that rights on another wiki do not by necessity add value to a request for rights here. Yes, I personally would look more kindly on a commons admin requesting the bit here, but I would be loath to see any form of automatic rights been given just because of access levels on another wiki. Seth's RFA is a welcome exception to an unwritten rule, but that's really all it is. Pedro : Chat  15:26, 18 December 2008 (UTC)
 * Giggy is a crat on Commons, and he failed 5 times. Fairly certain people in this particular RfA are just trying to make a point.-- Koji †  15:34, 18 December 2008 (UTC)
 * Yep, I voted support to make a point - the point that I wouldn't want to touch the spam blacklist with a 10 foot pole, and if Seth wants to do that (and he's said that is all he will do) then that's good news. What other point is there? Pedro : Chat  15:41, 18 December 2008 (UTC)
 * Same here, as I explain on the RFA - I did once, on meta, and I caused all sites starting with t to be blacklisted :P Hence why I don't touch it, and most others don't. It's too risky for me at least.  Majorly  talk  15:44, 18 December 2008 (UTC)
 * More accurately, he was only a bcrat during his last request. I'm fairly certain he failed for completely different reasons each time. I'm positive he will be able to pass eventually.  Majorly  talk  15:36, 18 December 2008 (UTC)
 * He should have been an admin LONG ago... and would have passed his last time if not for his dumbass nominator... but this is besides the point.--- Balloonman  PoppaBalloon 15:39, 18 December 2008 (UTC)

Majorly's point really is the key one - if he can be trusted to block sites globally on meta, what's the problem here? I'm glad to see the community are actually seeing this rather than just blind edit counting, etc. GTD 15:40, 18 December 2008 (UTC)


 * Having particular rights on sister projects and the like should not be a free pass. However, good (and bad) deeds done on sister projects should be considered during requests on .en. Kingturtle (talk) 16:01, 18 December 2008 (UTC)


 * It might be useful in these cases to ask the candidate "What are your best contributions to, and why?" and "Have you been in any conflicts on over editing in the past or have other users caused you stress? How have you dealt with it and how will you deal with it in the future? " Kingturtle (talk) 16:09, 18 December 2008 (UTC)
 * I could go along with this, but I think we need to also shift the emphasis when dealing with admins/crats from sister projects (especially the established ones.) Change the concept away from "granting them the bit" but rather to confer the bit from another project.  It's a subtle difference, but a big one.  Perhaps the analogy might be to a doctor.  If I were a doctor in Texas, I could move to California and start a practice in California.  Now, I would still have to get licensed and jump through some hoops, but I won't have to go through medical school again.  The Vetting process is different.--- Balloonman  PoppaBalloon 16:24, 18 December 2008 (UTC)
 * The problem is there isn't cross-wiki standards like there are across state line. The analogy is more like you're a doctor in Britain, and move to California - in which case you would have to take the boards, jump through many more hoops, and ultimately prove yourself to the (American) community - people won't take your license at face value, which is what's happening in this RfA. Tan   &#124;   39  16:33, 18 December 2008 (UTC)


 * Current cases rarely make good law; I suspect Seth would have had a much easier ride, and not have started with a NotNow close if he had been nommed by someone like Majorly who could, as an active editor here and on a site where Seth is active have explained why this is a sensible exception to make. Whilst I don't think that Edit count is very important for those who pass, I think that over a hundred RFAs fail per year with speedy closes per Snow or Notnow, and that if we had some code to prevent users self nomming if they had fewer than 1500 edits we would avert a lot of angst. In reality we do have edit count thresholds below which few !voters will seriously consider a candidate, and IMHO not to formalise this is just leaving a mantrap set in the nursery.     Ϣere  Spiel  Chequers  16:18, 18 December 2008 (UTC)
 * I actually wonder what would have happened if I hadn't closed and reopened this case. I think my actions actually have helped Seth's RfA along.  If I had closed it, there MIGHT have been some discussion, but I think most would have accepted it.  By reopening it, it forced others to think about it and not act on first impulse.  (And based upon the current status, I think reopening it was the right thing to do.)--- Balloonman  PoppaBalloon 16:24, 18 December 2008 (UTC)

Treat it as a case by case basis for now, if we wind up doing the same thing over and over again then it will become the consensus position. My personal feelings are your edits and administrative actions on sister projects count for better or for worse here, and odds are if you got the bit on a project not known for handing out the bit like candy, then the same contributions that led your peers on that project to give you the mop will be reason enough for me to give you my support. On the other hand, if the project's standards for adminship were significantly lower than the standards here, then you'll still have to show you meet my personal standards.

There is an issue with sister projects that are not in English: It's impossible for me to research your work on those projects. In this case, I'll want either a nominiator, co-nominator, or early supporter to say "I've reviewed his work on the other projects, here's the good and bad that I've found...." davidwr/ (talk)/(contribs)/(e-mail)  18:03, 18 December 2008 (UTC)

I have mentioned my enhanced flags (sysop or 'crat, depending on when it was) when applying for adminship on other projects (for those curious, I'm at six). In those requests, participants considered my previous sysop/'crat appointments, as well as my local contributions, before making their choice. That is exactly what we should be doing. Someone is a sysop on another project? Fantastic. But their edits here are what should be the final determining factor in whether they pass or fail; I don't want a Spanish Wikipedia sysop coming here, getting the flag, and then applying their Fair Use policy on our articles (for those that don't know, they don't have any Fair Use). A cross-wiki sysop needs to demonstrate here that they understand our policies before being granted the flag; it shouldn't be an automatic matter. EVula // talk // &#9775;  // 07:01, 21 December 2008 (UTC)
 * I think a case by case basis is the best way to go. Someone who wants to be an admin is asking for the trust of this community, and this community should be allowed to decide.  The community (and, in discretion cases, the closing 'crat) can take all these things into account.  I think there is an undertone that someone who has earned the trust of another community should not have to undergo auto-da-RfA here, but I think that this community should look at the candidate and decide.--Wehwalt (talk) 06:34, 23 December 2008 (UTC)
 * I have to agree - en.wiki has to make an assessment but as the current case shows take other projects into consideration for good or bad. Agathoclea (talk) 08:43, 23 December 2008 (UTC)

Considering the variable (and often inferior) standards on almost all sibling projects (commons being an exception), I strongly oppose this proposal. Sandy Georgia (Talk) 05:51, 24 December 2008 (UTC)


 * Activity on other wikis may, and should, be brought up during the RfA, but everyone needs to go through the EnWiki RfA process to become an EnWiki sysop. -- Avi (talk) 06:00, 24 December 2008 (UTC)
 * The currently ongoing RfA and discussion surrounding it has convinced me this is a very bad idea. Despite the fact that this RfA appears to be passing.--- Balloonman  PoppaBalloon 06:29, 24 December 2008 (UTC)


 * I concur with Avi. Admins from other wikis should not receive full admin rights on en.wiki without community consensus via the RfA process. Majoreditor (talk) 06:32, 24 December 2008 (UTC)

I think that now that the user has passed under the existing rules, we can mark this proposal down as unnecessary (and interesting, as well). The system works! Lankiveil (speak to me) 08:49, 30 December 2008 (UTC).

Speedy-pass RFAs
OK, this won't go anywhere now but I'm planting the seed for the future:

A successful RFA can be snow-closed early if in the last 24 hours of the RFA, there are no opposes, neutrals, removed supports, or any allegations of !voter fraud, and during the entire 24 hours supports outnumbered oppose+neutral 19-to-1 and there were at least 100 supports. But in any case, the RFA should run 50% of its scheduled duration.

As I said, this has a WP:SNOWball's chance of passing now but if I mention it now, maybe in a year or two it will have a chance.

Why bother? What's the sense in waiting a week if the outcome is obvious after 3-4 days? davidwr/ (talk)/(contribs)/(e-mail)  23:01, 27 December 2008 (UTC)
 * See WP:PEREN, RFA section.  — Rlevse • Talk  • 23:04, 27 December 2008 (UTC)


 * Total oppose. We keep them open for a week for a good reason; because contrary to popular opinion, not all our editors are schoolchildren, and many people can only check in once a week. It's entirely possible that someone will come along towards the end of an RFA with new evidence that's thus far been overlooked; conversely, nothing is gained by passing an RFA early. –  iride scent  23:05, 27 December 2008 (UTC)
 * The reasoning is to allow for the possibility of new diffs and whatnot turning up. Lets say X has 100 supports, 0 opposes and 0 neutrals after 3 days, and you speedy approve it. A couple of days later a RfA bloke (lets say balloonman, he always vets candidates very well) is looking through his contributions and finds a polemical, racist essay people missed that would have sent all 100 of those supports rushing to the oppose button. What are you meant to do? You cannot justify simply removing his adminship unless you want to go through the whole rigmarole of an official RfC and so on, and re-doing the RfA to take this info into account renders the entire "speedy pass" idea moot. Ironholds (talk) 23:06, 27 December 2008 (UTC)
 * Oppose, with respect for the intent. Why? Because maybe someone who really, really knows the editor well, including some shady behaviour, may not see it [the nomination] because he or she [that is, the editor "in the know"] only edits on, say, weekends.  I've always felt that we are lucky to get 2% of the voters in an RfA to actually have some personal interactive history with the nominee.  And sometimes you may have to wait for that.   Un  sch  ool  23:06, 27 December 2008 (UTC)
 * Oppose per Iridescent. I'd rather have WP lose the services of a qualified admin for at most 84 hours than risk unnecessarily passing an unqualified admin.--Wehwalt (talk) 23:09, 27 December 2008 (UTC)
 * Oppose per the graph at the right which shows that consensus really only solidifies after seven days. Some candidates with almost 100% support after five days have gone on to fail so that extra time is being put to good use. Icewedge (talk) 23:14, 27 December 2008 (UTC)
 * Oppose, mainly the once-a-week thing, and also the fact that something could surface in the last day which causes a lot of people to rethink.  Garden . 23:22, 27 December 2008 (UTC)
 * Oppose. I understand your sentiments, as I personally know an administrator whose RFA was not in question after three days, but as Iridescent said, the once-a-week thing is far too dangerous, IMHO. J.delanoy gabs adds  23:59, 27 December 2008 (UTC)
 * Oppose Horrible, horrible path for us to go down with RfAs. I've seen plenty that looked grand in their first 24 hours, and then started going downhill. EVula // talk // &#9775;  // 00:03, 28 December 2008 (UTC)
 * Pile On Oppose (does that equal POO?) Ling.Nut (talk&mdash;WP:3IAR) 00:52, 28 December 2008 (UTC)
 * Oppose per the reasons stated above. Simply not a good idea. –Juliancolton <sup style="color:#666660;">Tropical <sup style="color:#666660;">Cyclone  01:01, 28 December 2008 (UTC)
 * Oppose now and in the future There is no advantage to the proposal other than efficiency for efficiency's sake. A week is a good balance between expediency and making sure the broad community has a chance to participate. Townlake (talk) 01:06, 28 December 2008 (UTC)
 * Oppose with moral support - I see where you're aiming at here, and it's not entirely bad. Unfortunately it's one of those ideas that would only ever work well in theory, and only under really specific circumstances that can only be defined in hindsight. // roux  <span style="border:1px solid #801818;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;">  01:11, 28 December 2008 (UTC)
 * Oppose. A Recipe for disaster. - Mailer Diablo 01:23, 28 December 2008 (UTC)
 * I'm going to withdraw this tomorrow based on the graph and the weekend-only editors, neither of which I gave as much thought to as I should have. That graph is a good one, it nicely answers the counter-question of "extend RFA to x days" where x > 7.  I'm not archiving this discussion yet in case anyone has a novel reason that will be of interest to the next person who brings this up.  The more independent reasons why this is a bad idea, or for that matter, any reasons not mentioned why it's a good idea, the better before I close this.  To everyone who helped cure me of my ignorance, I say thank you.  You know what they say about ignorance and stupidity:  Ignorance can be cured.  davidwr/  (talk)/(contribs)/(e-mail)  01:27, 28 December 2008 (UTC)
 * Yeah, sorry about misreading and sillily archiving it. I wonder if I'm stupid or merely ignorant. Looks like the former at the moment :( NuclearWarfare  <sup style="color:green;">contact me <sub style="color:purple;">My work  02:14, 28 December 2008 (UTC)
 * Withdrawing per chart showing convergence on day 5 and weekend-only editors. Lessons learned.  Lesson number zero, zero because I already knew it:  Talk is good, especially when proposing significant changes.  davidwr/  (talk)/(contribs)/(e-mail)  04:02, 29 December 2008 (UTC)

Bot idea
'''I was originally going to close this thing at 1200 UTC today, but I found that was insufficient time. Instead, opinions will close when the ball drops. -- Dylan 620  Contribs Sign! 12:20, 29 December 2008 (UTC)'''

I have an idea for a bot: DylanBot. DylanBot would update the tallies for Requests for Adminship once in a while. How frequently should this be done? Once every 10 minutes? 30? Once an hour? -- Dylan 620  Contribs Sign! 17:44, 28 December 2008 (UTC)
 * User:SQLBot already does that. Ironholds (talk) 17:47, 28 December 2008 (UTC)
 * No it doesn't. He means on the RFA itself.  Majorly  talk  17:49, 28 December 2008 (UTC)
 * Ahh, right. Might be a bit difficult to do if there are wiki-code errors, though. Example: where someone buggers the code and the support (for arguments sake) numbering then goes 1, 2, 3, 4, 5, 6, 7, 8, 1, 2. How would it deal with such things? Ironholds (talk) 17:52, 28 December 2008 (UTC)
 * Support Also I think that if we had a bot doing the routine tally work we could rely on editors to fix such code errors.  Ϣere Spiel  Chequers  18:00, 28 December 2008 (UTC)
 * Agreed. — <span style="border:1px solid #20406F;padding:1px 3px;font-family:Verdana,sans-serif;"> Aitias  // discussion 18:09, 28 December 2008 (UTC)
 * @Dylan620: Very good idea. :) Every 10 minutes would be okay, I think. — <span style="border:1px solid #20406F;padding:1px 3px;font-family:Verdana,sans-serif;"> Aitias  // discussion 17:51, 28 December 2008 (UTC)
 * Thank you, Aitias. You too, WSC. If the idea has at least 60% support by 1200 UTC tomorrow, then I guess my first bot is in the works! -- Dylan 620  Contribs Sign! 18:05, 28 December 2008 (UTC)
 * Oh, and the bot would do this on Requests for Bureaucratship, too. -- Dylan 620  Contribs Sign! 18:06, 28 December 2008 (UTC)
 * Great. :) — <span style="border:1px solid #20406F;padding:1px 3px;font-family:Verdana,sans-serif;"> Aitias  // discussion 18:09, 28 December 2008 (UTC)
 * I agree as well.--Wehwalt (talk) 18:19, 28 December 2008 (UTC)
 * Support. - Dan Dank55 (send/receive) 18:39, 28 December 2008 (UTC)
 * FYI, a bot like this was already denied at Bots/Requests for approval/WenliBot. And I don't disagree with it, either. <span style="font-family:Verdana,Arial,Helvetica;color:steelblue;">X clamation point  18:21, 28 December 2008 (UTC)
 * Just found a third area where the bot could do this: Requests for BAG membership. -- Dylan 620  Contribs Sign! 18:42, 28 December 2008 (UTC)

(<-) FWIW, the bot would run on Auto-Wiki Browser. -- Dylan 620  Contribs Sign! 18:46, 28 December 2008 (UTC) Please note that I'm not making this statement as an RfA participant, but is my official stance on the tally as it pertains to my role as bureaucrat. EVula // talk // &#9775;  // 19:30, 28 December 2008 (UTC)
 * I'm sure this won't make the slightest difference and it'll go ahead regardless, but I oppose this – I don't think we should have tallies at all, let alone make them an integral part of the process like this. If we're going down this route, we're essentially saying "yes, it is a vote". (If we do need to go ahead with this, might I point out that we do already have a perfectly good vote-counting bot written by Gurch for the Arbcom elections?) –  iride scent  19:01, 28 December 2008 (UTC)
 * I agree with this, it confirms that RfA is a vote.  aye matthew  @ 19:21, 28 December 2008 (UTC)
 * As X! puts it, has been suggested and denied, I think the concerns are still valid. I think we should just remove the tally completely and just add it once it was closed. There is no reason to have it there before...or at all. Regards  So Why  19:26, 28 December 2008 (UTC)
 * [ec]For what it's worth, I consider the numbers a simple tally of how many supporters and how many opposers have participated in an RfA; the numbers merely provide a rough barometer for the tone of the RfA, but aren't, in and of themselves, the end-all/be-all. (for example, see RfA/Tadakuni)


 * Support with caution I support this with caution. Can you do this as a test:  Every 10 minutes, copy all existing RFAs to your user space and run the bot against them?  While this might require BAG approval I think "speedy approval for user:-space testing" would be granted.  Of course the usual caveats apply - publicly accessible code, stop button, etc., before approval to edit in Wikipedia: space.  I'd want at least 50 test edits over at least 2 RFAs, including tests with +support, oppose, and neutral, undoing edits/minus support/oppose/neutral, and changes of !votes.  In particular, if someone botches a change to a !vote and causes the numbering to start over in a given section, will your bot account for this?  Quitting without saving the changed total is an acceptable way of handling such unexpected behavior.  To those who say "this confirms the tally is a vote" it doesn't change the existing situation.  If we want to remove the running tallies, go ahead.  I find them useful: RFAs which are not a landslide support or which have late +opposes and +neutrals get my attention more than landslide supports without late +opposes and +neutrals.  Landslide for me generally means 50+ net support within the first day or two and 90% support. davidwr/  (talk)/(contribs)/(e-mail)  19:32, 28 December 2008 (UTC)
 * The reason the tally is there is to provide a rough view of how much suppose vs. oppose there is. However, most (if not all) bureaucrats discount the tally. Why would this be necessary if the crats discount it anyway? <span style="font-family:Verdana,Arial,Helvetica;color:steelblue;">X clamation point  22:57, 28 December 2008 (UTC)
 * Everyone wants to keep track of the tally, in my view. We might as well have it accurate with as little effort as possible.  Why else would people have the tallies of current RfA's transcluded onto their talk pages?--Wehwalt (talk) 23:00, 28 December 2008 (UTC)

When I update the tally, to see what the numbers are, I just use the TOC to go to the oppose section and scroll up one line (to get the supports), hit home to get to the top, TOC to the neutral section and scroll up one line (to get the opposes), and then hit End to zip straight to the bottom to get the neutrals. It takes me all of three seconds. Even if the tally is removed from the RfA (which I think is a silly thing to do), the people who are more interested in following the crowd than in actually asserting their opinions will be able to do so. Again, the tally is merely a convenience. EVula // talk // &#9775;  // 23:48, 28 December 2008 (UTC)
 * Is there a specific need for the tallies on individual RfA subpages to be continuously updated? The SQLbot summary transcluded at the top of this page and on WP:BN already provides the 'scores' for those who care about the horse race.  (And the SQLbot summary is available for transclusion to interested editors' user pages.)  Enough RfA participants already update the on-RfA tallies that they're seldom far wrong, and frankly there's no great utility to having up-to-the-minute accurate-to-the-vote counts.  The 'crats don't – and shouldn't – care about the precise numbers, and a rough eyeballing of the RfA should give any voter an idea of whether or not an RfA is seriously contested.
 * If anything, we should be be removing the tallies from the RfAs altogether &mdash; it would compel readers to scroll through the RfA contents, encouraging them to pick up on important discussions. TenOfAllTrades(talk) 23:12, 28 December 2008 (UTC)
 * Sorry, but I think that's a rather naïve idea (that no tallys will suddenly end any !vote-tracking). The people that are too lazy to actually read the arguments aren't suddenly going to start reading more just because the tally is gone. The tally is a convenience, nothing more, nothing less.
 * I think it's just as naive to think that the tally is just a convenience and nothing more, that it doesn't affect how people think. I agree that removing it isn't going to magically solve RfA's problems and make people consider the arguments and the evidence more rationally, but anything that moves it away from a numbers game and towards that goal is worth it and a good thing in my opinion. RfA at its best can be a consensus gathering exercise where people take into account other's arguments. The overemphasis on tallies and keeping them updated takes a bit of the focus away from that. And your point that it is very easy to get the equivalent information in the tallies if you need them is more evidence that the tallies aren't needed. But if they are going to stay, they may as well be updated by a bot to save people wasting their time. - Taxman Talk 19:39, 30 December 2008 (UTC)


 * I think it's good if the bot can detect formatting that is commonly broken, and some general fixes. I think tallying is just an aesthetic issue and would not tip the status quo of whether it is a discussion/vote/!vote between editors. It's just like how the bots actually calculate the percentage and list it here/BN, and yet we still get the humans to do their own assessment in promoting the candidate. - Mailer Diablo 01:01, 29 December 2008 (UTC)
 * Support This will allow editors to spend their time on other parts of the wiki. The  Helpful  One  01:14, 29 December 2008 (UTC)
 * *WHACK* for your horrible misspelling of "their". <span style="font-family:Verdana,Arial,Helvetica;color:steelblue;">X clamation point  07:32, 29 December 2008 (UTC)
 * *WHACK* for your hideous pedantry ;) —Cyclonenim (talk · contribs · email) 11:02, 29 December 2008 (UTC)
 * *WHACK* for your hideous joke :P. Ironholds (talk) 11:04, 29 December 2008 (UTC)
 * *WHACK* just to keep *WHACK* going. :P -- Dylan 620  Contribs Sign! 12:22, 29 December 2008 (UTC)
 * *WHACK* for making edits just to prove a point :P. Ironholds (talk) 12:35, 29 December 2008 (UTC)
 * *WHACK* because everyone else is doing it. ~ mazca  t 12:43, 29 December 2008 (UTC)
 * *WHACK* for collapsing in the face of peer pressure. Ironholds (talk) 12:56, 29 December 2008 (UTC)
 * *WHACK* Ironholds for WHACKing others 3 times! C'mon, Ironholds! Some people haven't even WHACKed once! :P -- Dylan 620  Contribs Sign! 13:00, 29 December 2008 (UTC)
 * *WHACK* for resorting to WP:EDITCOUNTITIS :P. Ironholds (talk) 13:53, 29 December 2008 (UTC)
 * No idea why you are all whacking me! Refresh your page you silly people :P The  Helpful  One  13:47, 29 December 2008 (UTC)
 * Hey! I was whacking Ironholds that last time! :P -- Dylan 620  Contribs Sign! 13:51, 29 December 2008 (UTC)

I don't think a bot is necessary, I think we do a decent enough job as it is, and it doesn't take much effort. That being said, if a bot were introduced, I wouldn't have a problem with it IF it included the count in the edit summary. EG If it doesn't give the count in the edit summary, it is worthless. Also, if it updates the count every 10/30 minutes then it needs to check to see if the count has changed. If the count hasn't changed, then there is no reason to update.--- Balloonman  PoppaBalloon Take the CSD Survey 14:55, 29 December 2008 (UTC)
 * I think this (the idea of introducing the bot) is a good idea. The tally is a controversial feature of the RfA page, but I agree with EVula it is there for "convenience", and its existence alone I do not think has a huge impact on how RfA operates. Having a bot to update the tally and fix any other formatting errors will also be a "convenience", and one that is worth having. Camaron | Chris (talk) 13:13, 29 December 2008 (UTC)
 * Indeedy. We do it quite well ourselves. This is not to say I don't see the point, but rather that I feel it is an awful lot of effort for something rather small. If Dylan620 is willing to code the thing and it passes through the approval process without any issues, fine. It isn't what I'd feel is an economical way to spend time, but if someone else wants to do it it is no skin off my nose. Ironholds (talk) 15:01, 29 December 2008 (UTC)
 * Dylan has a legitimate proposal here, but I'm afraid we won't gain much benefit from such a bot. Even if the bot updates every 5 minutes, I highly suspect people are still going to update the tally with every new vote. –Juliancolton <sup style="color:#666660;">Tropical <sup style="color:#666660;">Cyclone  16:12, 29 December 2008 (UTC)
 * Exactly... if he is looking for a bot idea to write, he might want to check out Dweller's talkpage. I think there is a discussion there, that would be prime for a bot and beneficial.--- Balloonman  PoppaBalloon Take the CSD Survey 17:41, 29 December 2008 (UTC)EDIT: See WP:BN for discussion of said bot proposal if you are interested.--- Balloonman  PoppaBalloon Take the CSD Survey 21:40, 29 December 2008 (UTC)
 * Making my rare vacation-time appearance...I don't think this bot is necessary. The tallies are pretty much always updated anyway; I think most people update the tally immediately after they post their support/oppose. Useight (talk) 22:11, 29 December 2008 (UTC)
 * I've checked a few diffs, and no most people don't update the tally themselves, on Requests for adminship/Suntag usually its been done by user:Wehwalt. I suspect most people just edit the section they are voting in, so yes the tallybot would be useful.  Ϣere Spiel  Chequers  22:47, 29 December 2008 (UTC)
 * It's fairly hit-or-miss; sometimes, the tally is updated constantly, while others, not so much. If you look thru my old contribs, you'll see me updating the tally a lot (enough so that I got named TallyBot at one point). It's a convenience, but not crucial. EVula // talk // &#9775;  // 22:49, 29 December 2008 (UTC)
 * I see. So you've given me a bot case of botulism?--Wehwalt (talk) 22:52, 29 December 2008 (UTC)
 * *groan* I'd try to respond to that, but I don't want to botch the joke. EVula // talk // &#9775;  // 22:56, 29 December 2008 (UTC)
 * Don't bot-tle it up.--Wehwalt (talk) 22:59, 29 December 2008 (UTC)
 * Yes, that's strictly verboten. Tan   &#124;   39  23:00, 29 December 2008 (UTC)
 * You're driving me bot-ty!--Wehwalt (talk) 23:01, 29 December 2008 (UTC)--Wehwalt (talk) 23:01, 29 December 2008 (UTC)
 * If I was feeling jovial in my post Christmas input to WT:RFA, I might mention that this bot will remove the opportunity for "aspiring admins" to add some much needed Wikipedia space edits to their overal count in certainty that they will be totally non-confrontational.... :) Pedro : Chat  22:55, 29 December 2008 (UTC)
 * Good point, and yes I've seen an RFA oppose for lack of Wikipedia space edits. But there are plenty of non-contentious areas in the Wikispace where you can do useful things that clock up edits without any risk of confrontation wp:fac and Disambiguation pages with links for starters, so why not automate something that can be automated?  Ϣere Spell  Checkers 12:06, 30 December 2008 (UTC)
 * Did you mistype that? As it stands, it looks like you're suggesting that there's no risk of confronation at WP:FAC ... --Malleus Fatuorum 14:17, 30 December 2008 (UTC)
 * Nope I didn't mistype that, I do believe one can do useful and non-controversial work at wp:fac. But for the avoidance of doubt I do regard the more controversial work one can do there as also being useful.  Ϣere Spiel  Chequers  19:31, 30 December 2008 (UTC)

About closing this
Above it is mentioned that this discussion would be closed at New Years Eve (tomorrow). I'd recommend it stay open longer. The discussion seems split, and I don't see a clear consensus for yes or no.  aye matthew  @ 12:40, 30 December 2008 (UTC)
 * I don't like the idea of having a deadline for a talk page discussion. Polls are evil. –Juliancolton <sup style="color:#666660;">Tropical <sup style="color:#666660;">Cyclone  17:02, 30 December 2008 (UTC)
 * Oppose something. Tan   &#124;   39  17:03, 30 December 2008 (UTC)
 * Support something // roux  <span style="border:1px solid #6D351A;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;">  19:12, 30 December 2008 (UTC)
 * Just let the discussion die on its own; there's no need to establish an arbitrary limit. We can consider the discussion closed once the bot archives it. :) EVula // talk // &#9775;  // 17:07, 30 December 2008 (UTC)

Splitting up admin rights for bot use only
I think you guys might be interested in a new proposal I thought of here. Foxy Loxy Pounce! 01:14, 31 December 2008 (UTC)

8 failed RFAs so far this month
Dear Pedro, whether or not we regard it as "one of the reasons why RFA is broken" we all know that RFAs from editors with less than 2,000 edits will be snowed. The recent amendments to Guide to requests for adminship may be dissuading some, and in my view will be a successful change to the RFA process even if they only prevent one snow per annum; but is there any reason why we can't add a rule that no-one can go for RFA without 2,000 edits?  Ϣere Spiel  Chequers  15:02, 5 December 2008 (UTC)
 * Only the one that says we've had the debate a dozen plus times and it's never got anywhere. Imagine the (unlikely) secnario of the perfect candidate; 900 mainspace edits which created 10 DYK's and 5 GA's and 2 FA's. 100 AIV Reports. 400 other project space edits. 300 Talk page edits. 250 User talk edits. 49 user edits. Total 1999. This perfect candidate (assuming civility, accuracy et. al) should clearly get the mop but would be barred from trying. The reason for not putting in an arbitary figure is two fold;
 * that there will allways be the exception to the rule, and
 * no-one can ever agree on the arbitary figure.
 * Don't get me wrong - I personally agree with the idea you present (tweaking the numbers maybe but in general) but the community has, until now, never supported such an idea. Pedro : Chat  15:18, 5 December 2008 (UTC)
 * Pedro, you forgot, he has to accept the nomination, that pushes him to the 2000 point!--- Balloonman  PoppaBalloon 15:26, 5 December 2008 (UTC)
 * Nah, that was one of the 400 project space edits. 399 at AFD / RFPP and 1 to RFA. :) Pedro : Chat  15:29, 5 December 2008 (UTC)
 * We've had a few cases recently where people who have been around for a while encouraged someone to run who was either likely not to pass, or who needed more instruction than they got in how to pass an RfA. I'm reminded of the advice that it's a bad idea to "help" a small child on a playground to try to do things that they're not comfortable doing; they have a built-in sense of what they can't handle yet, but this sense can be overridden by an adult saying "Go ahead!  You can do it!" - Dan Dank55 (send/receive) 15:30, 5 December 2008 (UTC)
 * Pedro, surely if the user is that perfect and has made 1999 exceptional edits, and reads that he needs 2000 to apply, he'll simply go and make that one more edit? The truth is, if they are that committed to the project, they can wait until they pass the tiny border of 2000 edits. —Cyclonenim (talk · contribs · email) 15:35, 5 December 2008 (UTC)
 * Which is why setting a hard and fast number of edits will not work. If we say "you cannot apply for adminship with less than xxxx edits" then editors will simply HUGGLE or WP:WELCOME their way to that number. We don't need to set a basis in numbers - we set a basis in quality - and the (alleged) purpose of RFA discussion is to agree on the quality of an editors contributions in respect of their ability to use admin tools. Pedro : Chat  15:41, 5 December 2008 (UTC)


 * Auto passing someone at RFA based on edit count wouldn't work, but stopping people standing if they didn't have two thousand edits would reduce our snows and not nows. OK maybe this would again tank, and maybe 2,000 isn't the right level - but I don't think anyone recently has succeeded on less than 3,000. Howabout holding a poll on this here, if anyone thinks 2,000 is too high they can vote for a lower limit. - Please !vote below:  Ϣere Spiel  Chequers  15:46, 5 December 2008 (UTC)

Vote
Do you agree that there should be a requirement for RFA candidates to have at least 2,000 edits?

Agree in Principle:


 * 1) Support any limit up to 2,500 1,500  Ϣere  Spiel  Chequers  15:46, 5 December 2008 (UTC)
 * 2) Although I disagree that it's the best way to review a candidate, we all know a candidate with only 1000 edits will not pass, regardless of how good they are in theory. Until people's mindsets about edit count change, I thinks setting a limit will help to reduce the number of SNOWed and NOTNOWed RfAs we see here. If people decide to HUGGLE and TWINKLE their way up to the limit, do you not think people will see through it? —Cyclonenim (talk · contribs · email) 15:56, 5 December 2008 (UTC)
 * 3) I would support some limit (say 500) below which any admin may delete the RFA and untransclude with an appropriate non-bitey note to the candidate. (Yes I'm aware of the debate about deletion of premature failed RFA's). Pedro :  Chat  15:59, 5 December 2008 (UTC)
 * I too believe in deleting true RfA's that are premature, but this was rejected a few months ago at VP.--- Balloonman  PoppaBalloon 16:10, 5 December 2008 (UTC)
 * 1) I think 1k is a sufficiently round number to make the minimum requirement (something I've been a vocal fan of for quite some time). mazca is spot-on about the minimum requirement having the side-effect of pushing up the standards all the more; someone with barely over minimum, whatever that is, will always be looked at suspiciously for being just barely over. No need to set it high; just high enough that your average newbie can't submit an RfA. EVula // talk // &#9775;  // 16:39, 5 December 2008 (UTC)
 * 2) I would be in favor of reasonable minimums for self-noms and nominations by people who have not done any successful nominations in the last year. If someone with successful nomination experience is willing to nominate you, I'm willing to at least look at you, on the assumption they will do their research first. I'm not sure if 2000 or 1000 or some other number is the best, a dig-through of recent successful and failed noms would help.  I would also favor reasonable time-in-service minimums for people who aren't nominated by experienced nominators.  12 months, or 6 months + 1 featured content, significant participation in 1 WikiProject, or 5 participations in good-article creation/review and/or DYN would substitute for 12 months.  I'm sure I left out some other things that could allow someone to come in without waiting 12 months or getting an experienced nominatior-nomination.  davidwr/  (talk)/(contribs)/(e-mail)  22:44, 5 December 2008 (UTC)
 * 3) Really, in today's wiki, how many people with <2000 are going to pass? It saves alot of people from having a bitey experience and being officially stamped NOOB on their wiki-record.-- Koji †  23:52, 5 December 2008 (UTC)
 * There will always be special cases, such as experienced editors or even administrators on non-English Wikis, an experienced editor returning after abandoning another account for non-negative reasons, and that very rare bird, the perfect editor whose first 500 main-space edits are all major contributions to featured content and whose non-mainspace edits are all thoughtful and helpful. How often will this happen?  Not as often as I'd like, but we need to be able to override any guidelines should the need arise.  davidwr/  (talk)/(contribs)/(e-mail)  03:42, 6 December 2008 (UTC)
 * !voting, but instead of raw edit count, how about a requirement for some other activity? For example, the editor must have been involved in promotion of an article to GA or FA status, either as "primary" editor or as a reviewer?  This would demonstrate a strong grasp of Wikipedia's content policies, an ability to work with other users, and other attributes which I believe are critical to a successful admin.  As a pleasant side effect it will help improve the quality of articles on the project rather than further the fever of editcountitis.  SDY (talk) 17:43, 9 December 2008 (UTC)
 * Content may be king, but not all editors who would make good admins have the wherewithal to get an article to GA or FA. Many gnomes--who would be otherwise excellent with the mop--would be excluded on this basis. // roux  <span style="border:1px solid #4B0082;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;color:#4B0082;">  18:26, 9 December 2008 (UTC)
 * There's always a backlog of articles on the good article nominations page to work with as reviewer, and being able to review articles for GA status again shows knowledge of content policy and an ability to work with people. Gnomes may be great people, but admins have to be able to work with other editors in complex situations where judgment and discretion will very likely be challenged.  It may have been intended as WP:NBD, but admin intervention is a lot more than just a mop.  SDY (talk) 18:37, 9 December 2008 (UTC)
 * Yes, but some people just don't have the talent for reviewing articles. Nor should they have to demonstrate competence in that area if, for example, their admin interests lie in (e.g.) WP:UAA and blocking vandals. // roux  <span style="border:1px solid #36454F;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;color:#36454F;">  18:40, 9 December 2008 (UTC)
 * Hmm. If being an Admin is treated like a job application, it still seems to make sense that there should be some default barrier to entry, a "You Must Be This Tall To Ride" sign to cut back on the snowballed RfA's.  Length of tenure would be another alternative, though some people (like me) go long stretches without much activity.  The rules can always be set aside for exceptional cases, but having a "don't bother to apply yet" cutoff seems appropriate.  To spin one of the proposals below, maybe limit self-nomination to people who have met some arbitrary criterion? I don't think raw edit count is the best choice (quality is much more important than quantity), but the idea at heart isn't a bad one for minimizing rejected RfAs, which have some bitiness to them that's considered harmful.  SDY (talk) 20:46, 9 December 2008 (UTC)
 * Aye, which is why I suggested that the cutoff be 2K edits for selfnoms--that is to say, if the user has fewer than 2K edits, they need to have someone (IMHO, preferably someone involved in the admin coaching process as they tend to be seasoned in terms of looking at someone, evaluating their chances, and suggesting avenues for improvement) nominate them. And note that this would need to be phrased as exclusionary; "you must have 2K to selfnom, success dependent on other criteria." Think of it as getting a driver's licence: you have to be 16 (or whatever), but simply being of age doesn't guarantee a pass. // roux  <span style="border:1px solid #465945;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;color:#465945;">  21:02, 9 December 2008 (UTC)
 * I'd shy away from edit counts alone as a mark of performance, they're too easy to inflate and don't reflect knowledge of policy or ability to play well with others and other elements considered in the RfA, but a "one of the above" system where if an editor must have either 1) 2000 edits or 2) 3 months of tenure or 3) participated in some substantial fashion (GA/FA/etc...) or 4) has been nominated by another established editor (i.e. someone who is clearly not a sock or meatpuppet), et cetera. The nomination could follow a format where the applicant indicates what criterion they met (and be deleted by a bot with a friendly message on their talk page if they clearly failed to read the directions).  SDY (talk) 21:16, 9 December 2008 (UTC)
 * 1) Agree in principle, although I think 2K is too high. Who are we fooling?  Yes, editcountitis is bad, but we do it now anyway.  With very few (no?) exceptions, there is not one candidate with, say, 100 edits who is going to pass.  Sure, you'll get people who say "I reserve the right to vote for someone with 50 edits", but everyone here knows that candidate is not going to pass.  Even if every single edit is exceptional.  And yes, edit count is a bad way to determine suitability.  Which is why we're not going to have an auto-pass at x edits.  Regardless of how many edits one has, we still do the research to see if they would be a good admin.  Many will still fail.  Nobody is saying that is going to change.  If we're not biting newbies, what is better for them to see?  1) Sorry, you don't have x edits, try again later.  2) You don't have a snowball's chance in hell, try again later.  3) The community looks for certain basic levels of contributions, you didn't meet these, try again what do you mean, what are the basic levels, everyone has their own, some written down, some not, look, just come back later.  That all said, as italicized, this is only agreement in principle.  In reality, I don't think we could ever get it to work properly.  --Kbdank71 17:21, 18 December 2008 (UTC)

Disagree to any limit:
 * 1) Edit count simply is a bad way of determining someone's suitability. – How do you turn this on (talk) 15:49, 5 December 2008 (UTC)
 * 2) It's not a bad idea, but I'd prefer two other approaches.  Either encourage people who sometimes make the wrong call when they nom to talk with people who usually don't make the wrong call, or encourage all applicants to talk with one or more of the people who usually make the right call before they run. - Dan Dank55 (send/receive) 15:56, 5 December 2008 (UTC)
 * 3) individual choice (but I am in favor, as I proposed before of putting verbiage that candidates without 2K are unlikely to pass.)--- Balloonman  PoppaBalloon 15:57, 5 December 2008 (UTC)
 * Verbiage already exists at Guide to requests for adminship, but I suspect not all candidates read the guide.  Ϣere Spiel  Chequers  17:06, 5 December 2008 (UTC)
 * 1) If we set an absolute minimum it is more likely to double on a yearly basis than not. — CharlotteWebb 16:03, 5 December 2008 (UTC)
 * 2) Better would be something like "at least X months of stable activity in a variety of both mainspace and project areas." X= 6 <> 3 // roux  <span style="border:1px solid #614051;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;color:#614051;">  16:04, 5 December 2008 (UTC)
 * 2,000 is far too high. It will have the side effect of psychologically pushing up the minimum edit count acceptable: If 2,000 is a hard limit, then someone with 2,500 suddenly looks even less likely to pass. In a year we'd end up discussing raising the limit to 3,000. You can make a decent case for adminship with 1,500 edits, and while I doubt many 1.5k candidates would pass, they're likely to know what they're getting in to, and the choice should be allowed. ~ mazca  t 16:13, 5 December 2008 (UTC)
 * As stated by WSC above, the limit doesn't have to be 2000. It could be as low as what Pedro is suggesting such as 500. Any limit would help to reduce the number of immediately failing RfAs. No RfA with 500 edits will pass, ever. —Cyclonenim (talk · contribs · email) 16:21, 5 December 2008 (UTC)
 * The problem with setting limits is that somebody will come along with 2001 edits, and a number of people will say, "You don't have enough experience." The person then gets upset because "they met the minimum requirements."--- Balloonman  PoppaBalloon 16:37, 5 December 2008 (UTC)
 * Then you explain to them kindly that there is not one requirement? Edit count is a minor factor, I've only agreed to this theory because I think it'll cut down on unnecessary nominations. If someone gets over 2000 or so edits then they still need to be nice people, good overall experience etc. —Cyclonenim (talk · contribs · email) 16:46, 5 December 2008 (UTC)
 * Sorry if I didn't elaborate my opinion as much as I should have; I basically think any hard limit is unlikely to help in any meaningful way. The person with 83 edits who doesn't really understand what RfA entails is obviously going to fail, but we have WP:NOTNOW for that reason. Having a hard limit will not discourage people from being rude to snow-able candidates: if anything, I suspect it would result in more incivility, as you tend to feel better about yelling at someone that broke a specific rule, rather than someone who broke a vague precedent. ~ mazca  t 16:56, 5 December 2008 (UTC)
 * Plus raising the bar—to exclude all those "unnecessary" (read: need not apply) candidates with 2001–2499 edits—is so much easier than considering less trivial factors! — CharlotteWebb 17:06, 5 December 2008 (UTC)
 * I'm not implying all 2001-2499 edit candidates are lost causes, but you can't be implying that a user with 500 edits has any chance of succeeding? —Cyclonenim (talk · contribs · email) 17:13, 5 December 2008 (UTC)
 * If something like this were to be adopted, it should be worded as an exclusionary criterion. Not "You must have 2000 edits to submit an RfA," but "If you have less than 2000 edits, you may not submit." Seems like a minor semantic distinction, but the latter wording makes it clear that there are lots of other criteria, and that the edit count is simply a minimum bar that must be reached before the other criteria will even be considered. // roux  <span style="border:1px solid #4B0082;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;color:#4B0082;">  17:17, 5 December 2008 (UTC)
 * 1) Absolutely not. There has been some progress made at RfA recently towards examining the substance rather than number of candidates' questions; this would be a huge step backwards. I'm not averse to a solution akin to Balloonman's above, where we quote actual statistics from past events as a guide. Not "2000 edits are the minimum requirement for administratorship", but rather "Note In 2007, a large majority of candidates with fewer than X edits were unsuccessful." The former pushes a poisonous prescriptive requirement, while the latter is a descriptive statement beyond empirical reproach which might help valued contributors avoid getting their fingers burned.  Skomorokh  16:39, 5 December 2008 (UTC)
 * So you expect a user with 500 or so perfect edits to pass? I find that highly unlikely. This system is only intended to help make some people realise they don't need to go through all the stress just to get rejected anyway. Statistics won't stop new editors from applying, for the most part, a 'requirement' will. —Cyclonenim (talk · contribs · email) 16:46, 5 December 2008 (UTC)
 * Did you mean to address this to my comment? It doesn't seem relevant to anything I wrote, sorry. Skomorokh  16:59, 5 December 2008 (UTC)
 * You spoke of including statistics instead of a requirement, so yes, I do feel it was relevent. —Cyclonenim (talk · contribs · email) 17:13, 5 December 2008 (UTC)
 * 1) Unnecessary. Simply encourages edit count inflation with Huggle and Twinkle, rather than actual learning or competence.  May inadvertently screw us over in the event we find a worthy exception.  We already focus too much on edit count and not enough on demonstrated good judgement.  And so forth. TenOfAllTrades(talk) 16:40, 5 December 2008 (UTC)
 * Did you read my comment above? I mean, if a candidate tries to boost their count with just Huggle and Twinkle edits, then we'll see right through it. We already do. —Cyclonenim (talk · contribs · email) 16:46, 5 December 2008 (UTC)
 * Yes. Have you read mine, or SheffieldSteel's?  I fear you're missing my point.  TenOfAllTrades(talk) 17:00, 5 December 2008 (UTC)
 * Sorry if you found that patronising, it was a genuine question to whether you'd seen it. Nevermind. —Cyclonenim (talk · contribs · email) 17:13, 5 December 2008 (UTC)
 * 1) Editcountitis shouldn't be encouraged, much less enshrined in policy. I reserve the right to support a candidate with just 500 edits spread between article building, dispute resolution and policy discussion, and I would not feel happy supporting a candidate with 10,000 automated edits. S HEFFIELD S TEEL TALK 16:54, 5 December 2008 (UTC)
 * 2) Edit count is a poor indicator of experience. –Juliancolton <sup style="color:#666660;">Tropical <sup style="color:#666660;">Cyclone  17:29, 5 December 2008 (UTC)
 * 3) I thought we are way past the era of Editcountitis. - Mailer Diablo 20:33, 5 December 2008 (UTC)
 * 4) Silly. Editcountitis=bad and should not be enforced. It also assumes that all edits are of equal merit, and I assume the only way to enforce it would be to have someone delist RfAs who's candidates fall below that level; if the candidate is obviously not going to pass that'd be done anyway per WP:SNOW and WP:NOTNOW. Ironholds (talk) 20:36, 5 December 2008 (UTC)
 * 5) Pure editcount is an exceptionally poor metric to determine the value and suitability of candidates for adminship.  Lankiveil (speak to me) 04:32, 6 December 2008 (UTC).
 * 6) Institutionalizing editcountitis? No thank you. Glass  Cobra  19:31, 6 December 2008 (UTC)
 * 7) Holy crap the editcountitis. To those who think 2000 edits is trivially easy: not all Wikipedians edit like you. In fact, the vast, vast majority of them don't. Go look for some ordinary good Wikipedians who aren't on the self-selecting "admin track", and marvel at their low number of edits.  r speer  /  ɹəəds ɹ  08:28, 7 December 2008 (UTC)
 * 8) Each editor may decide for himself what bars to set for RfA success. Kingturtle (talk) 16:08, 8 December 2008 (UTC)
 * 9) We should maintain the concept of quality and evaluation of the overall candidate and their circumstances, rather than setting hard limits. Pedro's 1999 point and others above are well taken. (I never thought I would support a candidate with less than 100. Lusty Seth from Germany with his SBL work needs the tools, and I see no need for a pro forma oppose based on edit count or territoriality.)   Dloh  cierekim  16:36, 18 December 2008 (UTC)
 * 10) I remember some years ago (and it was really not so long ago) when we had one candidate passing RFA with less then 1000 edits. The presence of Twinkle, Huggle, and non-admin rollback has inflated the typical number of edits of an average RFA candidate, but I would be more than willing to support an experienced and qualified candidate who uses the "Show preview" button when writing good articles. Sjakkalle  (Check!)  06:59, 23 December 2008 (UTC)
 * 11) For many reasons, already listed, and per lustiger seth, for a recent counterexample. Everyone is free to apply their own "hard" criteria upon prima facie glance (how many edits, for how long a time, who nominated, etc), but I don't wish to see any of them become official requirements. ---Sluzzelin  talk  01:04, 24 December 2008 (UTC)
 * 12) Absolutely not (and I'm sympathetic with the third option too). &mdash; Dan | talk 02:00, 24 December 2008 (UTC)
 * 13) Absolutely not, such a hard limit is a terrible idea. While it is true that most candidates with very few edits would fail, because we simply won't know them well enough to be comfortable, I reserve the right to support an exception to that general statement. Seraphimblade Talk to me 03:32, 24 December 2008 (UTC)
 * 14) No. Why!?, I hear you cry! Well, A) JulianColtan hit it well, edit count does not reflect the quality or experience of the editor. There are many editors with 2500 edits that probably will not pass (as they are inexperienced) so in principle it probably will not work. B) Enforcing a limit on edit count may put some great candidates (with FA's and the admin bits and bobs) with 1500 edits off. Andy (talk) 16:03, 31 December 2008 (UTC)
 * 15) I can't see any benefit other than avoiding having to SNOW.   Garden . 16:06, 31 December 2008 (UTC)

Voting is evil:
 * 1) Per heading.  This is a proposal that I personally find useless; however, it should definitely not be voted on.  Straw polls can be useful on individual articles, but should not be used to determine policy.  Ral315 (talk) 17:54, 5 December 2008 (UTC)
 * 2) I find such a measure thoroughly useless now that I think about it, and voting even more so.  bibliomaniac  1  5  23:30, 5 December 2008 (UTC)

Why are we voting on this? No such resolution on this page would be binding on the wider community anyway. If you want to make a proposal that the policy be changed, you should start with the policy page and publicize it on WP:VPP and elsewhere. <strong style="color:#228B22;">Avruch <strong style="color:#228B22;"> T 17:06, 5 December 2008 (UTC)
 * Informal straw polls are useful at gauging consensus. I imagine that if there was support for this proposed change,a discussion would follow as to what the best ways of implementing it would be, and then it would be posted to cent or WP:VP for broader consensus. Skomorokh  17:09, 5 December 2008 (UTC)
 * What he said. It's not supposed to be a formal vote but rather a poll to get an idea of where the RfA community stands on the issue. —Cyclonenim (talk · contribs · email) 17:13, 5 December 2008 (UTC)
 * This looks to me more like 'sorting' rather than 'voting'. // roux  <span style="border:1px solid #00009C;-moz-border-radius-topright:10px;-moz-border-radius-bottomleft:10px;padding:0px 7px;font-size:30%;color:#00009C;">  18:01, 5 December 2008 (UTC)


 * Straw polls are usually good at demonstrating a consensus - a discussion is the place to start, and if the discussion (and by that I mean more than four or five comments by fewer people) shows that there may be broad support for something, then a straw poll might make sense. In this case its a perennial proposal that any substantial discussion would show does not enjoy a consensus now, nor is it likely to in the near future. <strong style="color:#228B22;">Avruch <strong style="color:#228B22;"> T 17:15, 5 December 2008 (UTC)


 * (Proof I'm not against polls in general can be found here). <strong style="color:#228B22;">Avruch <strong style="color:#228B22;"> T 17:25, 5 December 2008 (UTC)


 * Good point Avruch, I've put a note on wp:VPP  Ϣere Spiel  Chequers  17:18, 5 December 2008 (UTC)
 * Think that was way premature. Would be better to come up with a proposal here, THEN take said proposal to VP.--- Balloonman  PoppaBalloon 17:28, 5 December 2008 (UTC)
 * (ecx2)What you mean like this proposal that was accepted here at WT:RFA a few months ago, but shot down at VP? One wherein we weren't trying to establish a minimum, but rather add some simple words to the template?--- Balloonman  PoppaBalloon 17:15, 5 December 2008 (UTC)

Alternate proposal with the same intent
I think we may be going about this the wrong way. I know that when I was thinking about applying for adminship earlier in my career (when I would surely have failed, looking at how close my RFA was when I did pass), I saw this line in the page header: "Please be familiar with the administrators' reading list, how-to guide, and guide to requests for adminship before submitting your request." Prompted by that, I went and read all of those, which pretty clearly gave me the idea that I wasn't ready yet. Instead of having another no-consensus debate on a minimum level of edits/months for requesting RFA, we should present these links to RFA hopefuls in a manner that says "Do NOT file this RFA unless you've thoroughly read these!" (Note: I'm not saying that should be the exact wording, I'm just saying that should be the effect of reading it.) While certainly some people may not read them anyway, the people who didn't read them would probably not pay attention to any minimum set of standards we put there, either. Either that, or maybe the first question in the RFA template should be: "Have you read XX pages thoroughly?" I think this could recieve more consensus than the perennial proposal to establish a formal minimum standard.--Aervanath lives in <b style="color:green;">the Orphanage</b> 03:46, 6 December 2008 (UTC)
 * I absolutely endorse the last suggestion here. Obviously, a part of the problem is overly enthusiastic newbies diving right in without reading the associated literature.  We'll never know how many read the guide to RFA, look at their edits, and say "never mind", but plainly more than a couple of people have skipped it.  By putting it in the questions, we can encourage these editors who might have missed th requirements to have a read through (because seriously, who would answer "No" to such a question?).  I doubt it will stop such candidates from nominating altogether, but hopefully it will stop good contributors from getting their fingers burned.  Lankiveil (speak to me) 04:31, 6 December 2008 (UTC).
 * This might be a keeper?--- Balloonman  PoppaBalloon 04:34, 6 December 2008 (UTC)
 * I prefer this idea. We could set a minimum limit, but it's likely at least some of the users being SNOWed will not read it, or will ignore it (since it seems most don't read the associated docs anyway). Unless the software is going to check their number of edits, and technically prevent them from creating an RFA if they're under the limit, I don't think there's much point. Either way, people will create early RFAs, be declined, and be disappointed. Somno (talk) 05:10, 6 December 2008 (UTC)
 * I like this proposal, though I see it as complementary rather than alternate as neither this proposal nor the preceding one will stop all obvious NotNows from starting premature RFAs. Reducing the hundreds of premature RFAs that crash every year is a worthy aim and this could help.  Ϣere Spiel  Chequers  08:09, 6 December 2008 (UTC)
 * Could we put one of those edit page header dealies (I'm not sure what they're really called, like the message above the edit box when you edit a disambig page) on the RFA page saying, "If you are transcluding an RFA onto this page, have you read these documents?"?--Danaman5 (talk) 04:24, 15 December 2008 (UTC)

WP:PEREN
I don't know about the rest of you, but I think this falls under perennial proposals. I have added a new entry to that page: WP:PEREN. szyslak ( t ) 09:11, 6 December 2008 (UTC)
 * I love WP:PEREN. <span style="font-family:Verdana,sans-serif"> — <b style="color:#060;">Rlevse</b> • Talk  • 11:01, 15 December 2008 (UTC)
 * Peren was my favorite character as well.--- Balloonman  PoppaBalloon 04:11, 16 December 2008 (UTC)