Wikipedia talk:RfA reform 2011/Coordination

This project is divided into sub pages in order to keep discussions focused and on track.
Read the main page first, then join a discussion on the respective page.
Please do not start a new thread or a new sub page on something that is already under discussion - see the TOC on each talk page.

Coordinators

Kudpung (talk) •

Pyfan (talk) •

Swarm (talk) •

Worm That Turned (talk)

A kind of newsletter

I'm thinking of sending this (see below) out now that we're getting somewhere. The distribution list includes task force members and everyone who has contributed to the project's talk pages:

Looks good to me. We should keep as many people as we can informed that we're still here and just as dedicated as ever. Swarm ^{u | t} 02:56, 13 September 2011 (UTC)

I concur. People need to know that this isn't just another one of those perennial discussions that never reach any meaningful conclusions. They need to know we are completely dedicated to improving the system. — Oli ^OR Pyfan! 06:51, 13 September 2011 (UTC)

I read it earlier and thought "Well, it seems awfully like we're chastising people and not reporting progress" but the fact is everything in there needs to be said. It makes sense to me. Worm^TT · (talk) 16:43, 13 September 2011 (UTC)

I know what you're saying, but Kudpung's toned it down a good bit so that it's not quite so harsh (although there are some people who are more than deserving of a good chastising :P ). Swarm ^{u / t} 20:07, 13 September 2011 (UTC)

Sent. May take a day or two for the message bot. --Kudpung กุดผึ้ง (talk) 00:07, 14 September 2011 (UTC)

RFA2011 news

RfA reform: Progress, and what you can do now.

(You are receiving this message because you are either a task force member, or you have contributed to recent discussions on any of these pages.) The number of nominations continues to nosedive seriously, according to these monthly figures. We know why this is, and if the trend continues our reserve of active admins will soon be underwater. Wikipedia now needs suitable editors to come forward. This can only be achieved either through changes to the current system, a radical alternative, or by fiat from elsewhere. A lot of work is constantly being done behind the scenes by the coordinators and task force members, such as monitoring the talk pages, discussing new ideas, organising the project pages, researching statistics and keeping them up to date. You'll also see for example that we have recently made tables to compare how other Wikipedias choose their sysops, and some tools have been developed to more closely examine !voters' habits. The purpose of WP:RFA2011 is to focus attention on specific issues of our admin selection process and to develop RfC proposals for solutions to improve them. For this, we have organised the project into dedicated sections each with their own discussion pages. It is important to understand that all Wikipedia policy changes take a long time to implement whether or not the discussions appear to be active - getting the proposals right before offering them for discussion by the broader community is crucial to the success of any RfC. Consider keeping the pages and their talk pages on your watchlist; do check out older threads before starting a new one on topics that have been discussed already, and if you start a new thread, please revisit it regularly to follow up on new comments. The object of WP:RFA2011 is not to make it either easier or harder to become an admin - those criteria are set by those who !vote at each RfA. By providing a unique venue for developing ideas for change independent of the general discussion at WT:RFA, the project has two clearly defined goals: Improving the environment that surrounds RfA in order to encourage mature, experienced editors of the right calibre to come forward, pass the interview, and dedicate some of their time to admin tasks. Discouraging, in the nicest way possible of course, those whose RfA will be obvious NOTNOW or SNOW, and to guide them towards the advice pages. The fastest way is through improvement to the current system. Workspace is however also available within the project pages to suggest and discuss ideas that are not strictly within the remit of this project. Users are invited to make use of these pages where they will offer maximum exposure to the broader community, rather than individual projects in user space. We already know what's wrong with RfA - let's not clutter the project with perennial chat. RFA2011 is now ready to propose some of the elements of reform, and all the task force needs to do now is to pre-draft those proposals in the project's workspace, agree on the wording, and then offer them for central discussion where the entire Wikipedia community will be more than welcome to express their opinions in order to build consensus. New tool Check your RfA !voting history! Since the editors' RfA !vote counter at X!-Tools has been down for a long while, we now have a new RfA Vote Counter to replace it. A significant improvement on the former tool, it provides a a complete breakdown of an editor's RfA votes, together with an analysis of the participant's voting pattern. Are you ready to help? Although the main engine of RFA2011 is its task force, constructive comments from any editors are always welcome on the project's various talk pages. The main reasons why WT:RfA was never successful in getting anything done are that threads on different aspects of RfA are all mixed together, and are then archived where nobody remembers them and where they are hard to find - the same is true of ad hoc threads on the founder's talk page.

Kudpung กุดผึ้ง (talk) 01:46, 13 September 2011 (UTC)

Messagebot & message

Message bot is a bit slow. It's probaby an exception - maybe the bot handler was on vacation, but it's worth bearing in mind. The message has produced some trolling which has been deleted, and a characteristic comment by one user who does exactly what the message asks editors not to. The message has also produced an interesting new task force member. Things seem to be slow again at the moment, but RfA is extremely slow at the moment, in fact it's the worst it's ever been and it's almost ground to a standstill., and so has WT:RfA. I was probably wrong a few weeks ago when I suggested that drama mongering on RfA had taken a break though - there's a new kid on the block who has been warned by many editors that his question and voting are inappropriate. There is probably nothing to be gained by releasing an RfC for the minimum qualification at the moment - the best time to get max participation is when RfA is very active, and if possible, highly controversial. --Kudpung กุดผึ้ง (talk) 05:06, 26 September 2011 (UTC)

Agreed. People can have very short memories on Wikipedia, and many will only be interested in reforming RfA if there is a visible and controversial RfA fresh in their minds. — Oli ^OR Pyfan! 06:44, 26 September 2011 (UTC)

Absolutely agree. Right now, we've had 4 (hopefully 5 soon) successful and 0 unsuccessful this month - hardly the best time to say that RfA is broken Worm^TT · (talk) 08:44, 26 September 2011 (UTC)

Right, but those figures do prove that RfA is boken! --Kudpung กุดผึ้ง (talk) 09:30, 26 September 2011 (UTC)

Try explaining that to someone who hasn't looked at the candidates or the system. From the outside it appears that the system is working. Worm^TT · (talk) 09:41, 26 September 2011 (UTC)

From the point of view that that the candidates who should pass usually do, and those that shouldn't usually don't, yes, it works. Even close calls are a fairly rare occurrence. From the aspect that its reputation as the biggest drama theatre on Wikipedia is putting potential candidates of running for office, it is well and truly broken. Kudpung กุดผึ้ง (talk) 21:01, 26 September 2011 (UTC)

Trolling

I think it's probably time to reopen our discussion about the value of keeping Keepscases' name on the task force list. --Kudpung กุดผึ้ง (talk) 01:20, 29 September 2011 (UTC)

Perhaps in a day or so - at the moment, I feel an important irony that he has stated that there is a problem with RfA and is actively working towards change, whilst at the same time disrupting the process. Worm^TT · (talk) 01:44, 29 September 2011 (UTC)

I agree with Worm, let this RfA play itself before moving on to this business. Best to keep the drama centralized *rolls eyes*. But I think the issue goes beyond Keepscases to the overarching issue of RfA trolling in general and the (in my mind, misplaced) belief that "everyone is entitled to their opinion" no matter how big of a horse's ass they present themselves as. Trusilver 03:00, 29 September 2011 (UTC)

You may both be right, but I've just received another email from someone who wanted me to nominate them. They have withdrawn interest in running for adminship until this crap is resolved. This simply corroborates yet again the fact that serious candidates of the right calibre are not prepared to run this gauntlet. Those who say that Keepscases adds a touch of necessary comic relief or who say that admin candidates should be made of sterner stuff, are misguided. Anyway, it looks as if NYB is going to do something, so I agree that we should wait and see. --Kudpung กุดผึ้ง (talk) 03:44, 29 September 2011 (UTC)

Well that's bad news. Worm^TT · (talk) 03:56, 29 September 2011 (UTC)

I just had someone who I thought would have been an excellent candidate turn me down when I offered to nominate him for his second try, and I honestly don't blame him. I wouldn't wish RfA on anyone, and I almost feel bad to offer to nominate, knowing what I'm about to put someone through. Trusilver 03:58, 29 September 2011 (UTC)

With 36,000 edits under your belt, it's time you ran yourself before you make enemies. The longer you work on Wikipedia, the more likely it is to attract vindictive comments from people who vote with a vengeance. It's the mistake I almost made by leaving it so late ;) Kudpung กุดผึ้ง (talk) 04:05, 29 September 2011 (UTC)

Oops! My bad :( I forgot. Struck. Sorry for opening old wounds. Kudpung กุดผึ้ง (talk) 04:08, 29 September 2011 (UTC)

Dante was an author before his time, I think that his vision of the 5th circle of hell in Inferno is probably a good representation of Wikipedia, where the wrathful spend eternity attacking each other. Does that NOT sound like the average RfA? It just shouldn't be that way. However, the collectivist nature of the project makes it hard to be anything but. Trusilver 04:41, 29 September 2011 (UTC)

--Kudpung กุดผึ้ง (talk) 00:05, 6 October 2011 (UTC)

WMF support

We've known how Jimbo feels personally for a long time, although he's never technically endorsed our particular project.

As you guys (probably should) know, Philippe has recently made it clear that not only is the WMF supportive of our efforts (i.e. this project), but they're actually willing to allocate working hours to support us where they can. I'm sure you'll all agree that this is hugely significant, and that we should fully utilize the support that's being offered here. So let's try to answer Philippe's question: how can the WMF help? While this doesn't mean the board will suddenly implement all of our proposals, it definitely opens new doors for us and I think we should discuss what to ask of the Foundation. So, thoughts? Swarm 21:20, 5 October 2011 (UTC)

Support of the foundation opens doors that were previously closed, obviously. I don't want this to sound like I'm advocating an end-run around policy or consensus, but having the Foundation open to assisting with real RfA reform means that a system can be enacted more readily that would have gotten bogged down (and already has... for years) in megabyte upon megabyte upon megabyte of discussion and political hand-wringing. We have the possibility now of going back to basics. Almost nine years ago, Jimbo gave the oft-quoted no big deal speech, and we have done everything in our power to stray from the spirit of that message from that day forward. To anyone who has gone back to the first year or so of admin discussions (and most of us that are here talking have, ad nauseum), have seen a happier and friendlier version of the project where adminship truly was "no big deal". Something I repeat fairly often in RfA discussions is that adminship is nothing more than a few extra buttons to help with the maintenance of the project; a few extra buttons that a chimpanzee with a few hours of training can use adequately. It is not the nuclear football. The project is not going to cease to exist in a fiery cataclysm if the wrong person becomes an admin. Look back at those early days when three or four people voting support led to admins being appointed... did the world come to an end because a lengthy, mud-slinging discussion didn't ensue? Nope, and it's not going to today either. That is why what we should ask for is the ability to implement a new trial RfA procedure. I'm not saying what the procedure is (though I have an idea), only that it should be the simplest viable system for confirming an admin. The mechanics of such a system are for a different discussion, but I think getting the Foundation on board for a trial run would be the most appropriate and efficient step. Trusilver 22:34, 5 October 2011 (UTC)

Adminship is seen as a seal of approval, and the statement that adminship 'is not a big deal' is one of the most often misunderstood and misquoted aphorisms of Wikipedia. To some it is indeed a big deal, particularly, for example, those who are minors with a short tenure and low record of participation (edit count). - simply said, the SNOW/NOTNOW variety. There are candidates of a more sinister nature: adults who have sockpuppets, who practice lies and deceit, and who would use their tools and perceived aura of power to their own advantage; a look at some (not all) of the desysopped individuals will demonstrate the ones who fit into this category. In the main however, most admins are mature people with enough clue about writing articles for an encyclopedia, and knowledge of its policies and software, and above all, who have a trusted and keen sense of judgement and know how to use it with civility. Most of those who are admins, I am sure, will tell us that once they got 'promoted' and shifted some of their work to admin duties, being an admin really is no big deal, even, if it does bring them some rare admiration and thanks from others for their work. Admins generally feel humble when they get a kind word and it helps confirm that they are not a bad people, and that they are doing most things right. Some admin tasks are less pleasant and can, and do, attract a lot of flak - some of it very nasty, all of it generally undeserved, but in the end they are just doing what they can: keeping the Wikipedia clean and free of junk and vandalism.

The problem is nature, who in her wisdom has decided that most life forms have a pecking order, hence we have this inaptly used word 'promotion' when it comes to according admin rights, and and the congratulations that abound when someone gets them. That adminship is not a big deal is proven by the silent majority of the 14,000 or so regular contributors to the project who might be possible candidates of the right callibre and who are reluctant to expose themselves, not to the scrutiny, but to the humiliation and downright nastiness and trolling that takes place on RfA with impunity. Those who do good work, and mainly steer clear of polemic and policy-making will sail through with hardly a murmur, on their first run, and with 100+ support, while there are others who have worked in contentious issues such as blocks and deletions and will have accumulated some detractors, and whose RfA turns into a very bitter experience even though they pass.

There are lots of ideas for alternative systems, such as for example a secret ballot, but there would still be the question of who would be the judges, and through what mechanism the judges (arbcoms, crats, stewards, editors with highly trusted user rights?) would be appointed. There are also suggestions to give the bit to everyone who has a certain length of tenure and number of edits. What cannot be kept out of the equation however, is a candidate's character, for which there is no metric.

The Foundation has taken criticism recently where it has stepped in and rejected strong local consensus for several important policies. Whether they will learn from this is another discussion, but they do have access to tools and meta data that we don't have, and the WMF is staffed, in my most recent experience, by mature and competent people, and we can listen to them and take their advice and suggestions. It's up to us however, to come up with solutions for reform that will meet with the community's blessing. The bottom line is that the voters have made the process of admin selection 'a big deal'. Changing the way the voting is conducted is the most important method, IMHO, of finding a fix for the problems. Kudpung กุดผึ้ง (talk) 00:05, 6 October 2011 (UTC)

Just for the sake of clarity, the Foundation will definitively not be implementing any changes. They're willing to work with us and officially support an alternative (in the hopes that WMF support will help overcome the usual community resistance), but it's all up to the community still. So, we have to decide, are we going to refocus and start seriously looking at alternatives? We all know why we haven't done so thus far. Is this new found support from the Foundation enough to make us change course? On one hand, this may be the best position anyone's ever been in to propose an alternative to RfA. On the other hand, if Foundation support isn't enough, we'll be in the realm of everyone else who's attempted it. Swarm 01:48, 6 October 2011 (UTC)

I knew this statement from the WMF was coming, and that's partly the reason for my long post above. I welcome their response , which could possibly help us in the future with any technical demands for solutions we may reach. I believe that no refocus is necessary, but that we no longer need to proceed half-heartedly in the shadow of whatever we do, it may be overruled by the WMF, or that an imposition of a system might come from them. This should now encourage us to get started on drafting and agreeing on the proposal statements for any RfC for reforms that we want to put to the broader community. My personal opinion is that in view of the apparent urgency for more admins, we should seek to significantly improve the existing system rather than devise some other radically different process. Kudpung กุดผึ้ง (talk) 03:19, 6 October 2011 (UTC)

I don't see why we can't do both - improve the existing system AND have a radical alternative, Wikipedia:Redundancy is good could apply to more than just the number of admins, but the processes to create them. If the existing system is not working perfectly, why not have an alternative to allow a potential candidate to chose one that would match their strengths?

As for the foundation's involvement, I am curious to what sort of statistics they might have that we don't have access to. I think it might be a good idea to see if we can get an opinion from the foundation on whether the devs would be able to implement a technical solution to the minimum requirements, and which radical alternatives would be unacceptable. Worm^TT · (talk) 09:24, 6 October 2011 (UTC)

How about agreeing to interim changes that would improve the current process while conceiving a completely different process? The second will take a long time, but at least the system won't be quite so maleficent in the meantime. Like Worm, I'd also be very interested in seeing the statistics the WMF have at their disposal. While the changes we make are ultimately decided by community consensus, it is heartening to know that the WMF are watching and agree with what we are doing. — Oli ^OR Pyfan! 05:40, 7 October 2011 (UTC)

Let's not misinterpret what's being discussed here. The WMF is not saying that they have anything up their sleeves; they are offering advice and technical help, if and when we need it. I am not against radical alternatives, I'm just in favour of reaching and implementing more urgent solutions, which in my opinion are those that involve doing something to the current system to improve the quality of the voting and cut out out the drama that is driving potential candidates away. What we need to do now is get those proposals worded, agree on them, and offer them to the community for RfC. Any volunteers?Kudpung กุดผึ้ง (talk) 06:50, 7 October 2011 (UTC)

That's basically what I meant. We could begin with the Clerks proposal, since that's already quite advanced. — Oli ^OR Pyfan! 06:58, 7 October 2011 (UTC)

Significant downturns

I'm just going to point this out because I wouldn't be surprised if our members don't realize this, and I think this data should be included in our next newsletter/announcement.

Good news

August 2011 saw the lowest number of unsuccessful candidacies (4) since August 2004. Last month, September, saw the lowest number of unsuccessful candidacies (1) since we began recording them (in April 2004). As of now, we're two weeks into October and there's only been one unsuccessful RfA so far. One current one is quickly approaching a successful closure, and the other current one (which incidentally happens to be mine ;P) is looking good so far (fingers crossed).

Bad news

August 2011 saw only one promotion, a low that has only been reached one other time in RfA history (December 2010). Also, the number of candidacies in general is reaching a near all time low. August and September saw only 5 RfAs in total, apparently the lowest level since February 2003.

Hopefully the former is indicative of a pattern while the latter is simply an outlier; only time will tell, though. Swarm 08:22, 13 October 2011 (UTC)

I'm certainly aware of it and the plummeting parabola of a logarithmic projection would show that we're going to hit negative equity in around 6 months. For those who want it in simpler terms: We will no longer be covering the natural decline in the number of admins. Let's not forget that the metric for active admin is based on an extremely low level of participation. Kudpung กุดผึ้ง (talk) 09:06, 13 October 2011 (UTC)

You're right, this isn't a triviality, we're seeing all time lows (even more recently than our unsuccessful candidates page). We'll have to wait for the final October returns, but if the trends solidly continue, our outlook will have to change. When this project started, RfA was a "nasty" place but now it's quite seriously on the verge of going into a downward spiral. Let's hope October provides some relief from the slump; we're currently right on track with last year's monthly numbers so it's too early to tell. I've started a list at Wikipedia:RfA reform 2011/Downturns for now, let me know if you can think of any other statistics to add. Swarm 14:33, 14 October 2011 (UTC)

It may be that this project is too little too late, and that adminship needs to start failing before anything useful happens. I'd rather keep a more positive outlook though. I'm still thinking that an RfA drive in December might be a good idea :) Worm^TT · (talk) 14:39, 14 October 2011 (UTC)

That sounds like a great idea, Worm. Swarm 14:52, 14 October 2011 (UTC)

An RfA drive, while sounding like a good idea, doesn't solve anything. It's certainly an underwhelming finish to a year-long project. Everyone talks about the decline in active admins as though it's a big surprise, but it's really not. You don't need to do a lot of impressive math to divine that the NP backlog won't go below seven days, or that in the most active hours of the day, AIV reports can sit for an hour or more without being processed. This would never have happened just two or three years ago. I have to admit that Worm may be right, and that the only way to provoke real change is for things to get worse. I may be pessimistic, but I feel that project will need to go into a major decline before the community will generate a consensus for real RfA changes. Trusilver 15:54, 14 October 2011 (UTC)

I don't think it solves the problem, and I certainly wouldn't have it be the end/transformation of this project. But I still think it's an idea worth trying. As an aside, it's remarkable how my creeping pessimism was dashed by Worm's optimism, and then that ray of hope was swiftly and mercilessly crushed by your pessimistic comment. xD Seriously, though, it's depressing. :( It wasn't really relevant to bring up here anyway, this is something I'm going to we should keep tabs on and if it continues, submit it to the Signpost. Swarm 05:48, 15 October 2011 (UTC)

Hah! Sorry about that :). I don't think the RfA drive is a bad idea, I think we should go for it. But I am pessimistic. I came here with the idea that RfA reform under an existing system was possible, but I'm becoming more and more sure that a more radical approach is going to be necessary; only because it's not the process that's broken, it's the people participating in the process. For every ten people participating in an RfA, you have AT LEAST ten distinct sets of RfA qualifications. I think the only way RfA can improve under the current structure is if there is a concrete set of parameters for adminship. Like if we took your admin qualifications and said "okay, this is policy". Then RfA would become much like an AfD where the sole point of discussion is "Does the candidate fulfill the qualifications." Trusilver 08:15, 15 October 2011 (UTC)

These are all suggestions that have already been, but I don't think they are really a project coordination issue. They are probably best discussed on the relevant sub pages where they will get more exposure. Kudpung กุดผึ้ง (talk) 08:41, 15 October 2011 (UTC)

Unbundling

Unbundling sometimes gets brought up on this project. Just in case anyone is arguing for the right to view deleted pages, refer them to this: (WP:PEREN) - both the current and former Wikimedia legal counsel are on record as opposing these proposals due to legal concerns (see Wikipedia:Village pump (proposals)/Persistent proposals/Straw poll for view-deleted. Another new proposal for it at the VP was closed today as inoperable.
--Kudpung กุดผึ้ง (talk) 05:09, 14 October 2011 (UTC)

RfA results chart

Kudpung

Kudpung กุดผึ้ง (talk) 06:14, 16 October 2011 (UTC)

That is a disturbing chart. Thanks for putting in together Kudpung. The chart shows definitively that WP:RfA is stagnating, and that something needs to be done about it. 25 RfAs in January last year compared with 5 in September this year? It's shocking. It is interesting to see, though, that far fewer WP:SNOW candidates are running now as well. It is also interesting to note that the number of promotions in January last year was only slightly higher than in September this year. — Oli ^OR Pyfan! 06:29, 16 October 2011 (UTC)

Well done. Don't know if you saw the above discussion, Pyfan, but you might also want to check this out. We're seeing quite a few drastic downturns of late. Anyway, let's see what Philippe says. Swarm 07:02, 16 October 2011 (UTC)

I did, but the graph somehow makes it seem even worse. It'll be interesting to see what the research team will come up with. — Oli ^OR Pyfan! 07:05, 16 October 2011 (UTC)

Research team? --Kudpung กุดผึ้ง (talk) 08:18, 16 October 2011 (UTC)

[See diff in my previous comment] Swarm 08:30, 16 October 2011 (UTC)

Phillipe has asked someone at the WMF to make a logarithmic line graph prognosis for us. I'll give her a ping and see how she's getting on with it. --Kudpung กุดผึ้ง (talk) 05:41, 26 October 2011 (UTC)

Consultation with a few experts in statistical analysis (well, social scientist PhD students who use stats analysis in their research). Unfortunately they all said the same thing: the sample size is too small to show any trend. No mathematical formula they tried produced anything more insightful than we'd get from just superimposing a hand-drawn line over the bar graph. --Kudpung กุดผึ้ง (talk) 08:58, 29 October 2011 (UTC)

Hi Kudpung!

You might ask for help at the statistics project.

The simplest distribution for modelling count data is the Poisson distribution, and its mean and variance are equal: You should expect a lot of fluctuations.  Kiefer.Wolfowitz 21:10, 6 November 2011 (UTC)

Well, October has come to a close, an inappropriate attempt at transclusion was made this week, and only one new nom today. With Rlevese ('crat), and Cirt (an AfD closure specialist) having had to vanish this year, and Sarek having surprisingly and discretely handed his bit back again today after going through all the crap earlier this year after his voluntary re-sysop, things aren't looking so good. There will also have been other admins who have quietly slipped away or who are no longer so active. our category of 'active' sysops paints a false picture because the threshold for 'active' is ridiculously low. There don't appear to many serious backlogs (others may wish to differ), and the deletions of new articles seems to be on the ball. Areas where there may be backlogs are AfD closures, MfD, and FfD. Not many admins appear to want to tackle contentious or complex discussion summaries - I don't either - and files seems to be an area where many of us lack the necessary knowledge of file policies; it's certainly not an area I'm familiar with. Kudpung กุดผึ้ง (talk) 10:05, 29 October 2011 (UTC)

Solutions?

I'm not sure if it is within the mandate of this project, but perhaps there is something we can do to combat this dearth. Based on WereSpielChequers' August 2010 Signpost article, I'm considering writing a new one, but I would prefer to have some help with it. --Kudpung กุดผึ้ง (talk) 10:07, 29 October 2011 (UTC)

I'd be willing to help. I think it would be a valuable opportunity to publicise the situation, and hopefully to make !voters realise that something needs to be done. It could also prompt some people who have been thinking about running for adminship to give it a shot. — Oli ^OR Pyfan! 13:06, 29 October 2011 (UTC)

Awesome, I agree 100%. In fact, I was just going to raise this suggestion yesterday, but you looked way to busy to be pestered even further. I'll absolutely help with this as well. Swarm ^X 17:45, 29 October 2011 (UTC)

I had already started working on a draft of this in my snadbox when I mentioned, and I'm just trying to format a table that I've imported from MW (they don't use the same markup), and waiting for some new charts. When I've got the format tidied up I'll move it somewhere official and then we can further develop it together. There's no desperate hurry. --Kudpung กุดผึ้ง (talk) 05:13, 30 October 2011 (UTC)

Sounds great. Swarm ^X 19:45, 30 October 2011 (UTC)

Frightening away the candidates

An interesting theory. --Kudpung กุดผึ้ง (talk) 09:55, 1 November 2011 (UTC)

I happen to like talking on IRC because you find editors being a little more candid (and honest) than they tend to be on the project. One thing that I keep hearing over and over when it comes to RfA is something like "I wouldn't pass RfA, so why even bother?" or "It's just too stressful to deal with." I think we have already come to understand that these are the top two reasons that people don't accept RfA nominations. I look over their contributions though, and my assessment is that most of the ones that say they can't pass RfA probably would. Trusilver 15:35, 1 November 2011 (UTC)

There is an interesting summary of an extraordinarily detailed report on the findings of research into the Polish system of selection for their admins. Admittedly, theirs is a much smaller Wiki, but their conclusions are not dissimilar to our own. --Kudpung กุดผึ้ง (talk) 18:43, 1 November 2011 (UTC)

You wouldn't happen to have a link on that, would you? Trusilver 19:04, 1 November 2011 (UTC)

Wikipedia:Wikipedia Signpost/2011-10-31/Recent research#What it takes to become an admin: Insights from the Polish Wikipedia. --Kudpung กุดผึ้ง (talk) 03:05, 3 November 2011 (UTC)

More stats showing possible decline

It's possible that my RfA criteria page is among the most visited (1,531 hits, 80 backlinks). Anyone else who has a stats page might like to do the same exercise.

Wikipedia:Advice for RfA candidates

• April - 42
• May - 251
• Jun - 124
• July - 145
• Aug - 118
• Sep - 140
• Oct - 178
• Nov -
• Dec -

Wikipedia:Request an RfA nomination

• Aug - 822
• Sep - 222
• Oct - 263
• Nov -
• Dec

--Kudpung กุดผึ้ง (talk) 03:37, 3 November 2011 (UTC)

Page views Jan 2010 - Oct 2011 of Wikipedia:Guide to requests for adminship.
Kudpung กุดผึ้ง (talk) 05:04, 3 November 2011 (UTC)

Can I ask, what program are you using to generate the stats? I've created some in Excel... but I'm not happy with how they look. Worm^TT · (talk) 13:35, 4 November 2011 (UTC)

XL 14.0.0 on Mac 10.6.8. It's clumsy but does the job. I'll send you the template I use if you like. Kudpung กุดผึ้ง (talk) 16:21, 4 November 2011 (UTC)

Please do. Worm^TT · (talk) 16:23, 4 November 2011 (UTC)

And another painful graph. Worm^TT · (talk) 11:07, 5 November 2011 (UTC)

It's what we needed. Kudpung กุดผึ้ง (talk) 12:03, 5 November 2011 (UTC)

Really?

Is anyone aware that someone went ahead and took the liberty of starting an RfC on the clerks proposal? Yawn. Swarm ^X11|11|11 13:30, 4 November 2011 (UTC)

Yeah, I wasn't too impressed with that, I must say. You've just had an email from Kudpung on the subject... I expect I'm about to get a YGM too ;) Worm^TT · (talk) 13:32, 4 November 2011 (UTC)

We do need to decide on how we want to move proposals forward. RfC, perhaps - though the proposal needs to be in a ready state before an RfC happens. I'm free most of the day tomorrow (I think), so I might be able to skype/IRC/chat here about how best to proceed. Worm^TT · (talk) 13:34, 4 November 2011 (UTC)

Yeah, Kudpung sent out an email literally as I created this section. I was quite surprised to see this pop up on my watchlist. We're not remotely ready to propose anything because we haven't even discussed how we want to make our proposals, much less whether our drafts are ready to be proposed in the first place. Way premature. Happy to discuss wherever, just let me know. Swarm ^X11|11|11 13:52, 4 November 2011 (UTC)

We need to stop this kind of thing and make it clear to task force members that we work as a team here. If they want to go off at half-cock and do this kind of thing, then please not here. RfC proposals need wording extremely carefully - half the time RfCs fail is because they turn into a meta discussion about the proposal statement. Besides which, I don't personally think the climate is right for these suggestions yet. If we appear to be slow, it's because we need to choose the right moment for things, and that moment is not now.

We need to get an article out to Signpost first, with as much proof as possible that between them, the !voters at RfA have wrecked the system. Look at what we have on a current RfA - most of it well intended, but making far too much of health issues. IMHO, a candidate's health is not part of the equation. We have an admin who has been blind from birth and does a better job than most. Providing a long spiel about health is like saying "I'm dyslexic, blind, have one leg, paraplegic, deaf, dumb and bloody stupid, but please take all this into consideration and make it easier for to become an admin than than for everyone else", and that just asks for trouble. What we're interested in here is: can you use the damn mop with common sense, or can't you? and that's basically what RfA is about for everyone, whether they have green skin and three eyes, and can edit the 'pedia hanging upside down in a tub of cold water on a waterproof iPhone. --Kudpung กุดผึ้ง (talk) 16:38, 4 November 2011 (UTC)

Next steps

Worm and I had a very long conversation last night on possible approacahes to RfA reform and and the worsening drought of noms. We need to get together on this and perhaps temporarily change our focus, but we need to do this with as little background noise as possible to avoid the intermittent trolling on this project. We need to reinforce the notion that task force is for experienced users, and after having slept on the discussion with Worm, I have come up with a unique idea that might just put an end to the trolling on RfA. Some people don't like the use of off-Wiki communications. I particularly dislike the Wiki IRC channels and avoid them like the plague, but recent Skype conferences I have had on various projects have convinced me that it's an excellent way to get some honest work done without the background noise. --Kudpung กุดผึ้ง (talk) 00:38, 6 November 2011 (UTC)

It's worth being clear about whether it's the choice of technology that's the problem, or the way it's used.

To be specific, for doing what you refer to as "honest work" on IRC, you wouldn't generally do it in one of the Wikipedia/Wikimedia IRC channels. That's the equivalent of thinking that a discussion about RFA reform will produce most productive results if held in WT:RFA - except worse, because channels like #wikipedia-en also regularly host trolls, and users who are banned on-wiki, and all sorts of other malcontents, plus people who are bored and want to discuss what they're having for dinner or their favourite meme (like... me).

So no, if you want a specific discussion about a specific topic on IRC, you create a separate channel for it (one command taking two seconds to type), and invite people who have something to contribute. Just as you would on Skype (where it's called a "conversation" not a "channel").

IRC doesn't do audio or video; but Skype does. IRC is useable by all users without needing to install anything, and is very efficient on bandwidth and client resource usage; Skype isn't. IRC allows instant access to a wide range of different Wikipedia editors with different skill sets and tool sets; Skype allows you to limit your interactions to people you've added to your contacts on Skype. They both have their pros and cons. --Demiurge1000 (talk) 09:22, 6 November 2011 (UTC)

Thanks, but the unique idea had nothing to do with communications. It's something I wanted to discuss with the coordinators over a 3-way video - which by the way performs with excellent results. Kudpung กุดผึ้ง (talk) 11:31, 6 November 2011 (UTC)

Yeah that's fine - definitely not something IRC can be used for. --Demiurge1000 (talk) 14:24, 6 November 2011 (UTC)

Defining the problem

I've been considering how one could best use the WMF's offer of help, and I think that one useful area would be to get some research commissioned. One of the problems I've had in convincing people that we have a problem is that while I can measure how many active editors we have who have the admin bit, that includes editors who edit regularly but only use the admin tools once in a blue moon. It would be useful to know how many active admins we have and the hours of the day or week when we are thinnest on the ground. In particular the gaps at AIV are apocryphal, measuring them and the trend for the last few years would either confirm that we have a problem in ways that no reasonable person could deny, or reassure us that we had more of a safety margin than we thought. Some of our differences at RFA are rooted in genuine disagreements as to the facts of the situation, getting an independent professional researcher to settle those issues would I believe make it easier to get agreement as to how things should change. ϢereSpielChequers 14:01, 6 November 2011 (UTC)

There are two ways of looking at this:

• What the stats would show
• What the admins report from their empirical findings.

Conclusions we can draw from empirical study are:

• Most vandalism edits, attack and children's test pages come from American time zones
• Most new uncontroversially inappropriate new pages such as spam, copyvio, nn BLPs /autobios, come from time zones on the Indian sub continent.
• Not all admins work at the same time of day/night
• Not all admins are in the same time zone
• Most admins are located in American time zones
• There are often long periods when AIV is not being watched (e.g. it's not a place I venture into)
• Some attack pages hang around for a couple of hours before they are deleted
• Expired PRODs and BLPRODS get deleted within seconds of expiry - almost as if a bot were on duty
• There is a near critical backlog at AfD closures (e.g.I don't close AfDs)
• There is a terrible backlog at FfD (I don't go there either)
• Volunteers (admins or editors) can't be forced to take on more work or operate in areas that they are not familiar with or are not interested in.
• Among recent RfA (say over the last 12 months) an increasing number of candidates appear to state that they will not be seeking to work in contentious areas. They probably believe by saying that, they will get an easier 7-day trip. problem is, we need candidates who have the experience and who are prepared to take on those problem areas.

Stats show that we fare very badly compared with other Wikis. Every en.Wiki admin has to cope with far more articles and far more editors than for example the German Wiki. Kudpung กุดผึ้ง (talk) 16:44, 6 November 2011 (UTC)

In responses to WereSpielChequers's post, I had just been breaking down the raw data based on one random day (31 October 2011) at AIV and have come to find out that the average time to process a report was 21.1 minutes. I don't know about anyone else, but I find that number horrifying. Obviously, one number is insufficient to base anything on, and I'm continuing to compile that information into useful information with a wider sampling of activity, but at least that points to a definite, measurable problem surrounding admin participation. More in another day or so. Trusilver 21:46, 6 November 2011 (UTC)

That's an excellent idea and one that could contribute to an article we are considering for the Signpost. However, while stats are indispensable to illustrate situation, averages may not necessarily convey the true concerns. While the arithmetic mean is often used to report central tendencies, it is not a robust statistic, meaning that it is greatly influenced by outliers. Notably, for skewed distributions, the arithmetic mean may not accord with one's notion of "middle", and robust statistics such as the median may be a better description of central tendency. Based on my list above, it would perhaps be more relevant to demonstrate in linear form, the backlog over time, such as when America is asleep. That may more accurately portray a possible need for more admins. Kudpung กุดผึ้ง (talk) 07:20, 7 November 2011 (UTC)

Not a lot here[1] that we didn't already know, but it's incredibly vindicating to hear it right from the mouth of a recent RfA candidate. If any of you haven't read it yet, by all means do. "My first ten "Oppose" votes were from editors with a combined 26 blocks." is something I find incredibly telling about our current system's flaws. Trusilver 19:50, 8 November 2011 (UTC)

Thanks for the heads up TS. I've been using the expression 'smoke and mirrors' a lot here when discussing voters - what a coincidence, and I see we haven't hesitated to block disruptive participants in the past. Kudpung กุดผึ้ง (talk) 23:57, 8 November 2011 (UTC)

Upcoming proposal

At User:Moe Epsilon/RFA. I can't understand why he declined to do it with the support of this project, but I sincerely wish him luck with this. It's nothing that we haven't suggested here already, and it's probably one of the perens, but it will be interesting to see the community's response. --Kudpung กุดผึ้ง (talk) 12:46, 7 November 2011 (UTC)

Coordination

I know we agreed that the climate is not right yet to launch any proposals for change, but since then we've suddenly had a couple of highly contentious RfAs that have embodied all the elements that have contributed to turning the process into a snake pit. There's been an AN/I about one voter that ended without consensus (or was closed before one could be reached), and which is now continued at RfC/U. Looking back over much older RfAs than I have in the past, I've noticed that sanctioning voters has not been altogether uncommon. I also think that Steve's RfA whether we have supported or opposed it, is getting out of hand and that at least one voter has probably said more than enough to support their vote but won't let up. I think we should make some serious moves now to coordinate the task force into some action on the voting issues, and do something to get some more nominations. I have come up with this idea but I'm not going to make any unilateral decision on it: We send a message to all the good admins and users we know who have not voted for a while but generally demonstrated clean, intelligent voting whether they support or oppose: Hi {{BASEPAGENAME}}. I notice you haven't participated at RfA recently. The quality of voting has deteriorated to the point that potential candidates no longer want to come forward, and RfA has reached stagnation point. Please consider keeping a lookout for nominations at Wikipedia:Requests for adminship on your watchlist and supporting, opposing, or leaving a neutral comment as need be. Thanks. We can put the Signpost article on hold until the IEP issue has blown over, and the Badger Drink RfC/U has concluded, and TrueSilver has come up with his promised stats. --Kudpung กุดผึ้ง (talk) 02:36, 9 November 2011 (UTC)

current developments

WT:RfA - I wouldn't want to see all our hard work, especially all the gathering of stats, wasted. especially where we're getting so close to a couple of solutions. --Kudpung กุดผึ้ง (talk) 06:52, 11 November 2011 (UTC)

I see that the project page is now getting it's format changed in what I think is not a particularly helpful way. I have reverted the diff. If anyone thinks it should have been kept, feel free to revert again. --Kudpung กุดผึ้ง (talk) 00:19, 17 November 2011 (UTC)

Some recent RfAs have again clearly demonstrated how the system has been progressively broken by the behaviour of the voters. The most recent one incurred a mass delrev and oversighting of over 20 diffs. Whether or not the mission of this project is to find an answer to the diminishing flow of nominations (the current spate of RfS doesn't change the overall trend), something clearly has to be done. According to a current discussion on the Bureaucrats' noticeboard, there appears to be some consensus that it's not forcibly their role to intervene. An open RfC/U on the behaviour of one editor is also being side-tracked and mocked in the same way as many RfAs. One independent discussion for reform is taking place under a euphemism for unbundling the tools. It is a meritorious attempt to do something, but it does not appear to address any specific issues, and may well simply create more bureaucracy, add more flags to the maze of user rights., and be seen by some as a short cut to adminship. perhaps we should be looking seriously at the possibility of clerks. --Kudpung กุดผึ้ง (talk) 05:34, 4 December 2011 (UTC)

Been a touch busy, sorry. The oversighting was due to an editor linking offwiki to problems with a person who has a similar name. I agree that that shouldn't happen, but I don't think we should be looking at that as part of pattern. I don't think that Clerks should be used to do anything new, especially enforce civility. If a 'crat isn't willing to, then this role definitely shouldn't. The role should be based on making sure things go smoothly and run according to procedure, but shouldn't be putting their judgement on civility over and above the rest of the community. Worm^TT · (talk) 11:08, 6 December 2011 (UTC)

Same here, apologies, I've been a bit busy and haven't been keeping up with talks lately. I'll catch up as best I can, but feel free to point me to anything specific that I've missed. Regarding this, I really don't know what to do. The clerks proposal is currently written as Worm says; it implements nothing new. We could change that, but that very well may sink the entire proposal. Crats don't inherently have any more jurisdiction to "police" comments than administrators do, and I understand why they wouldn't want to take up a controversial role such as that. Recent developments have shown that any administrator who makes an attempt to enforce our fourth pillar will receive far more trouble than it's worth. Swarm ^X 11:38, 7 December 2011 (UTC)