Proposal talk:Administrative hierarchy

and if admin numbers are falling?

Far from a large and growing body and a position that every serious editor aspires too; On the English Wikipedia active admin numbers are dwindling by about 1% a month. We actually need to spend time on working out how best to use our dwindling admin corps, and how to avoid the situation degenerating into a hierarchical one. WereSpielChequers 17:59, 3 October 2009 (UTC)

Can you back up that position though? How would a hierarchy be a "degeneration", and why do we need to avoid it? Can you provide an argument against any of the motivations in this proposal? Equazcion 18:48, 3 October 2009 (UTC)

Sure, though I'm afraid I will have to draw my examples almost entirely from the English Wikipedia as that is where I'm most active - but discussions I had at wikimania lead me to believe that some of this will apply to other wikis. Lets break up those motivations and deal with them separately:

• "Administrator" has almost become an inevitable rite-of-passage for every serious Wikipedia editor. There are many ways in which a wikipedia editor can "progress" in the community, such as getting an article to FA status. Becoming an admin is an increasingly rare one - we have only appointed 96 so far this year on the English Wikipedia, and if you look at en:Wikipedia:Successful requests for adminship running the gauntlet of the current RFA process is not something that many editors are willing to do. In particular there is a marked lack of candidates running for adminship who edit in their own names, in my view a sensible reaction to an increasingly vicious process. Who would want anyone googling their name to find an online interview with oppose reasons such as "This editor is naive, uniformed [sic] and has an appalling record regarding content", "The candidate's perspective on the deletion process is best described as "completely out to lunch". The dump I took this morning was, I assure you, a sincere contribution. Unfortunately, it was still crap." and "because I can"

• They have become a rather large body of users that are incapable of efficiently coming to decisions or settling disputes. Every day hundreds of decisions are made by admins, users are blocked, articles deleted, pages protected. Few of those decisions involve more than one admin, few of those decisions are wrong. I would argue that not only are admins capable of efficiently taking decisions, but also that the vast majority of admin decisions are efficient both in taking minimal effort to make and in being "right first time". I'd agree that the community has some issues where it has difficulty coming to a decision or thinking efficiently - but I struggle to think of examples where admins are the flawed decision making group.

• Even with regard to fairly clear-cut cases, administrators are hesitant to act due to the massive body of other administrators that might disagree with their actions. You have a point here, there are some en:vested contributors who can flout rules and are immune from sanctions that would apply to most users.

• In addition to this inefficiency, there is a certain lack of oversight in day-to-day events. A bad administrative decision requires the oversight of that administrator's own peers, which can be an uncomfortable social situation as well as a logistical mess. The advantage of having hundreds of admins is that if you make a mistake, an admin you've never encountered before could be the one to correct you.

• Disagreements among administrators, of which there are many, more often that not flair into lengthy heated wars. I'd agree that there are many disagreements amongst administrators, but my experience is that more often than not they get resolved without escalating into lengthy heated wars.

• Even a simple case of error or misconduct on the part of an administrator must spawn a lengthy debate, whereas if there were an oversight level, such cases could be dealt with swiftly instead. On the English wikipedia we have en:wp:trout and various quick sanctions, simple errors rarely cause lengthy debate. Complex ones sometimes do, but not simple ones.

• Large organizations generally require hierarchies in order to operate efficiently, and Wikipedia may be no different. Large organisations are vulnerable to diseconomies of scale such as becoming increasingly hierarchical and bureaucratic - for example compare NASA's handling of Apollo 13 with their mishandling of the Columbia disaster. Wikipedia's success has been driven by en:the wisdom of crowds, the more hierarchical we get the further we drift from what made us a success. WereSpielChequers 16:10, 4 October 2009 (UTC)

• Perhaps "incapable" was the wrong word. You're right: Many decisions, perhaps most, are made smoothly and without significant incident. Nevertheless what's left ending up at ANI makes for lengthy heated debates that could be avoided. Not that debates are a bad thing, but the type that end up lasting on ANI are generally pretty bad. They are very often the result of a disagreement over an admin's decision or error.

• You say the benefit of having many admins is that the admin who corrects you "could be" someone you never met; which might be fine if it were always the case. Nevertheless, a separated oversight level could eliminate the possibility altogether, as well as the possible appearance of impropriety.

• As far as complex vs. simple errors, this is very debatable, and it depends what you mean by "simple" and "complex". I've seen admins make "simple" errors, as in, uncomplicated straightforward bad decisions, such as making irrational blocks, and it takes a ridiculously long time just to get enough discussion going to decide to do something about it -- especially if the admin was in good standing with the community. You fear that hierarchy leads to be bureaucracy, and assumedly bureaucracy is undesirable because it slows down decisions. However what I'm actually proposing is a way to speed them up, by adding a level that doesn't need to debate, wait for discussion, then wait for the heat to die down, then wait for someone to have the guts to close. What we have now is already a kind of bureaucracy -- we have no organization, so we have to wait for everyone to weigh in before anything can happen. I'm proposing a level that could make quick decisions. A bit of hierarchy might be good, and mustn't necessarily lead to bureaucracy.

• As for NASA: You've brought up the most notably-failing hierarchy to show that hierarchies are bad. I don't think I need to tell you why that's a terrible argument. There are lots of examples of the opposite, as well as many examples of a single authority level tasked with policing itself that failed. Equazcion 19:04, 4 October 2009 (UTC)

Thanks for your reply. I agree that the minority that reach ANI include some that get "lengthy heated debates", and I can understand why at first glance it might look like that is the norm at the drama boards. But if you check the archives you will see a very different pattern, with several times as many quickly resolved and archived incidents as lengthy heated debates. But there's a broader issue here, natural justice requires that people accused of something have the opportunity to put their side of the case. That can get long and heated at times, but I don't see how you can avoid that without creating an arbitrary decision body that makes decisions for which it is not accountable, and from which there is no appeal.

Assuming you appointed members of this new level from within the community you would still have the risk that they would have had previous interactions with a transgressing admin. If anything amongst a relatively small group of people who hang out at the drama boards there will be individuals who have encountered everyone at that new level.

As for the need to minimise hierarchy, NASA is not the only organisation that degenerated into a bureaucracy, GM, the Byzantine, Ottoman and Spanish Empires are three more that come to mind. As for there being examples of "a single authority level tasked with policing itself that failed" I don't dispute that and I for one would not want to disband arbcom and turn wikipedia into such a single authority level organisation. WereSpielChequers 15:06, 5 October 2009 (UTC)

• So which is it? Hierarchies are bad, or we have one that we should preserve? You say you don't want to disband arbcom because it functions to police the admin level, which is a clearly hierarchical position, yet you seem adamant in the claim that hierarchies are inherently bad. If we have a hierarchy already, and it's a good thing, then who's to say a little more hierarchy might not be an improvement?
• My proposal drew a clear distinction between "bureaucrats (for example)" (emphasis added) and my proposed oversight level. The example was (obviously) meant to be inclusive of arbcom. Arbcom doesn't get involved in daily events, and is actually a rather drawn-out bureaucratic process, much as it is a form of oversight. I'm advocating, in certain daily situations, a more visible hands-on oversight level.
• I didn't say whether or not there would be an avenue of appeal for the oversight level's decisions, nor did I suggest how its members would be appointed. I neglected to mention these things because I haven't worked out the details -- but that doesn't mean they couldn't be adequately worked out, if the general idea for this proposal is something that people would like to see. Proposals are often argued down because of premature detailed feasibility arguments. It might be better to just weigh the merits of the general concept first. Do you (people in general) like the concept, or not like it? We'll work out the details and try to plan a way to make it work later, should it be deemed a popular enough idea.
• As for the past failed hierarchy thing, despite your self-contradiction, I'll answer this once again: Agreed there are many examples aside from NASA of failed hierarchies. Disagreed that this helps your argument. I've heard a lot of people die in car accidents, one every 12 minutes globally, I believe; ergo cars are bad? I've heard of many white people who went berserk and slaughtered people. White people are bad? Again, taking the worst examples of the failure of [something] does not say anything about its overall merit. That's what is known as propaganda. So, big picture, please. Hierarchies are the basis of nearly all organizations, both failed and successful. Equazcion 15:58, 5 October 2009 (UTC)

My position of being opposed to Wikipedia becoming increasingly hierarchical is not the same as the anarchist argument for having no hierarchies at all. In fact my examples, NASA when it put a man on the moon GM when it overtook Ford in the 1920s, the Spanish and Ottoman empires in their heyday - all were hierarchical; just nowhere near as hierarchical as they became during their declines. My argument is against adding another layer of hierarchy, not against hierarchy in principle, perhaps it would have been clearer if my original statement had included the word overly as in "how to avoid the situation degenerating into an overly hierarchical one." As for my criticism of your proposal lacking an appeal mechanism, well I based that on your words Even a simple case of error or misconduct on the part of an administrator must spawn a lengthy debate, whereas if there were an oversight level, such cases could be dealt with swiftly instead. Either you allow people a fair chance to put their case and accept that occasionally you will get lengthy debate, or you can have cases as you put it "dealt with swiftly instead". WereSpielChequers 19:34, 5 October 2009 (UTC)

It's not a binary decision. Swiftly does not necessarily mean without avenue of recourse; and as I said, we can work that out later, should the general concept of an oversight level appeal to people. If your argument is not against hierarchy per se but rather against 'a level of hierarchy that will cause the project to fail', then we are in agreement there (of course). On the issue of 'what particular amount of hierarchy would cause such a failure,' that's rather subjective and we'll probably have to agree to disagree there. Though I must clarify that I'm not necessarily adamant in my support for this proposal as it stands. I'd not only like for it to be considered and discussed as a possible solution to problems I've perceived (which we are again in subjective disagreement over), but for it to also evolve as necessary. Equazcion 20:08, 5 October 2009 (UTC)

My experience is that there is a trade off, and I'm not convinced that an extra level of supervisor above admin would cause any less discussion than now. As I pointed out earlier Either you allow people a fair chance to put their case and accept that occasionally you will get lengthy debate, or you can have cases as you put it "dealt with swiftly instead". If you think you can change the current process to be swifter whilst retaining or even improving the current fairness and right for the accused to be treated fairly and put their side of the story, then I for one would like to see more detail than "we can work that out later" as to my mind the bit you intend to work out later is the nub of the issue. WereSpielChequers 14:15, 13 October 2009 (UTC)

This proposal is only playing the Wikipedia game

What serious user would want to be an admin? Adminship is there for those who are interested in other aspects of the project, rather than the real task of providing content. - Brya 05:44, 4 October 2009 (UTC)

Whether or not adminship consists of "real" tasks and whether admins should be considered "serious editors" is a different argument entirely. I believe there's a proposal or two regarding the general usefulness of adminship. You may want to comment there instead. Equazcion 06:44, 4 October 2009 (UTC)

yes adminship is falling but plenty of people are still rejected and quit the project because of it. some poor guy, shoessss is up right now and these oh-so-overburdened admins have the time to go vote him down.

where 200 edits probably was enough to be an admin before, now you probably need 1000 and have to make sure you didn't make any political mistakes. others don't bother running since they know they will be denied. it would be easy to increase the numbers but fact is, admins don't want it that way. smaller numbers increase their supposed importance, and lets them whine on and on about the workload on admins and how they can't be bothered to spend time on task X because there are too few admins..... in any case most work is now done by bots, admins have sort of lost their usefulness. 202.111.2.186 08:32, 4 October 2009 (UTC)

I agree that plenty of editors who apply for adminship are being rejected and deeply regret that so many leave the project over it. I also agree that standards have risen to an absurd degree - far worse in fact than your figure of a thousand edits. I believe that only one successful candidate in the last 100 on the English Wikipedia had less than 3,000 edits, - most have many more. But I disagree that it is admins who are raising the standards and making it more and more difficult for an editor to become an admin, if you look at the RFA for the candidate you mentioned you will see that among non-admin !voters in that RFA there are 6 supports and 15 opposes. My observation of close calls where RFA candidates fail with between 50% and 70/75% support is that few get as many as 50 opposes, most of those opposes are not admins, and very few RFAS get opposed by more than 2% of the admin corps. As for Bots doing most of the work, Bots do a lot of gnomish stuff that editors used to do, in particular they do a lot of the vandal fighting that used to be done by people on the path to becoming admins. But the community is very cautious about automating admin functions such as deleting and blocking, so arguably we are becoming more dependent on a dwindling number of admins. WereSpielChequers 17:44, 4 October 2009 (UTC)

This answer reflects what is wrong with Wikipedia. A culture has sprung up where the only "serious users", that is the only users that are taken seriously are those who play the Wikipedia game, running up edit counts, pasting templates, enforcing a certain kind of look. Actually, at least originally, Wikipedia was intended to be encyclopedia, that would provide verifiable information, presented from a neutral point of view. From that perspective the serious users are those that are involved with adding and improving content. Of course, with much of Wikipedia now serving only as a platform to expound points of view, serious contributors are very much a minority, and now here are openly excluded from being "serious users". - Brya 05:50, 13 October 2009 (UTC)

It is unfortunate that edit statistics are one of the criteria we need to use, but we don't have much choice there. There's no software that can provide a summary of how many of those edits were actually useful, and we do need to guage an editor's level of experience somehow. It's kind of like legal drinking age -- a poor indicator of maturity, but all we have nonetheless. That said, it's not all we have. I think many if not most people who vote at RfA do spot-check the actual edits to make sure they're useful, and furthermore most passing RfAs are based on many votes by people who have encountered the editor in the past and can attest to their value. Besides, I don't mind if people run up their edit counts by making useful edits. Yeah we have lots of relatively easy ways to increase our edit counts now, but things like article tagging are arguably useful, if done thoughtfully, and admins should have experience with templates, I think, as they're a major facet of the encyclopedia. I can understand the argument to judge quality over quantity, like letting people with fewer, more significant article contributions become admins; however not only will that be difficult to gauge, but article contributions aren't really what we need admins for. Equazcion 06:15, 13 October 2009 (UTC)

To return to the question: why assume that any serious user would even want to be an admin? - Brya 06:56, 13 October 2009 (UTC)

You're not even asking a question. You're just trying to make a point, and I'm not playing along. Correct me if I'm wrong, but you either think adminship is useless, and/or all editors with admin aspirations must not be serious editors. We all hear you loud and clear, but I don't know what you're doing here. Go start a proposal to do away with adminship. If you have a problem with the word "serious" appearing here, you're just splitting hairs, as it's really not centric to this proposal. Equazcion 07:47, 13 October 2009 (UTC)

Actually, I did not say or even imply that "adminship is useless", only that it is just a single aspect of what happens at Wikipedia. I would not want discourage anybody from wanting to be an admin, after all, somebody has to be. Also, I did not say that "editors with admin aspirations [or admins] must not be serious editors": such users may be serious contributors indeed; or they may not be. These are unrelated things. - Brya 07:58, 13 October 2009 (UTC)

Then kindly state the point you do hope to arrive at, rather than asking a question that seems rhetorical and tertiary to the proposal. Equazcion 08:02, 13 October 2009 (UTC)

What I hope to achieve is to keep the idea alive that Wikipedia aims to be an encyclopedia, rather than just a political game to be played. - Brya 08:36, 13 October 2009 (UTC)

K, and that relates to this proposal how? Equazcion 08:38, 13 October 2009 (UTC)

The proposal pictures Wikipedia as a political game only. - Brya 09:53, 13 October 2009 (UTC)

And how so? Equazcion 17:01, 13 October 2009 (UTC)

One example is that the proposal still says "Administrator" has almost become an inevitable rite-of-passage for every serious Wikipedia editor despite at least two people on this page pointing out that this is untrue. As a fresh example en:Wikipedia talk:Requests for adminship#RfA victim...cough, volunteer trawl includes one example where a crat on EN wiki approached 6 editors and only one agreed to run for admin, it also includes a request from me for Emails from people wanting a nominator. I had one response. WereSpielChequers 08:31, 26 October 2009 (UTC)

Still not getting it here. How does a proposal for an administrative hierarchy perpetuate the view that administration is a political game? Or have you just piggybacked on someone else's complaint to make your own point yet again that there are assessments on the proposal page you don't agree with? Equazcion

Citation needed

There are several statements in the proposal that I have refuted on this page, however the IP who wrote this hasn't yet amended the proposal. Would people prefer that I correct the facts or merely show they are challenged? WereSpielChequers 14:19, 13 October 2009 (UTC)

I wrote this proposal; forgot to log in at the time. I would prefer to give our little debate some more time at least, before deciding definitively whether you've successfully 'refuted' me. Ideally we should have more people than just you and me to decide, but attention on this wiki seems spread rather thin. If you have specific problems with statements or language here, state exactly what you'd like changed so it can be discussed. Equazcion 17:03, 13 October 2009 (UTC)

Replacing

"Administrator" has almost become an inevitable rite-of-passage for every serious Wikipedia editor. They have become a rather large body of users that are incapable of efficiently coming to decisions or settling disputes. Even with regard to fairly clear-cut cases, administrators are hesitant to act due to the massive body of other administrators that might disagree with their actions.

In addition to this inefficiency, there is a certain lack of oversight in day-to-day events. A bad administrative decision requires the oversight of that administrator's own peers, which can be an uncomfortable social situation as well as a logistical mess. Disagreements among administrators, of which there are many, more often that not flair into lengthy heated wars. Even a simple case of error or misconduct on the part of an administrator must spawn a lengthy debate, whereas if there were an oversight level, such cases could be dealt with swiftly instead.

With:

"Administrator" has become a rite-of-passage for some Wikipedia editors. They have become a large if dwindling body of users who don't always efficiently make decisions or settle disputes. Even with regard to fairly clear-cut cases, administrators are sometimes hesitant to act due to the massive body of other administrators that might disagree with their actions.

In addition to this inefficiency, there is a certain lack of oversight in day-to-day events. A bad administrative decision can require the oversight of that administrator's own peers, which can be an uncomfortable social situation as well as a logistical mess. Disagreements among administrators, of which there are many, occasionally flair into lengthy heated wars. Even a simple case of error or misconduct on the part of an administrator can spawn a lengthy debate, whereas if there were an oversight level, such cases could be dealt with swiftly instead (unless it allowed parties to the case to put their side of the story and had an appeal mechanism).

Would address some of the flaws in this proposal. WereSpielChequers 08:51, 26 October 2009 (UTC)

I took some of your suggestions, but most of these would negate the assessment that there's even a problem worth trying to fix. You need to keep in mind that this is a proposal based on one or more editors' feelings that there is indeed a problem. Not everyone will agree on its stated assessments of the situation. They don't need to. You don't need to. Equazcion

I don't dispute that there are problems, but I'm afraid we disagree as to what the problems are, and as our diagnoses are different it should be no surprise that our prescriptions are very different. I am however happy to reconsider my position if you can find data to support your assessments. WereSpielChequers 16:33, 31 October 2009 (UTC)

It's not provable. I could bring evidence of heated debates and you could bring evidence of swift decisions. So where would that get us? I'm seeing a problem that you're not. I don't need to either convince you or change the proposal. Rather, this seems to be a proposal that you don't support. And that's fine. You don't need to. Equazcion

Unintended consequences

I fear for the unintended consequences that a bureaucracy (however ad-hoc) like this could create. We have enough power-plays on the administrative level in the first place (granted, I'm thinking largely of en.wp - but that's where most of the scaling issues lie).

And although the political aspects are unfortunate, at least the power is spread fairly horizontally. Will adding a vertical element improve administration? Xavexgoem 18:21, 25 October 2009 (UTC)

I'm thinking it'll cut down on the "power-plays". If administrators have their own "administrators" above them to worry about, they'll be more careful. Equazcion 20:54, 25 October 2009 (UTC)