Wikipedia:WikiProject on Adminship/BN

This is already a long page. Please continue discussion on talk.

copied from WP:BN at 08:49, 13 September 2006 (UTC):

Making it up as you go along

I don't recall approving bureaucrats to hand out arbitrary lengths of adminship service because they don't really know how to make the decision. Adminship has never, ever been handed out for 8 weeks at a time as a result of an RfA. Did you actually tell anyone anywhere taht you planned to do this if they supported Carnildo? Or are you just making it up as you go along? You don't own RfA, you merely implement its decisions, and no part of its decision referred to two months. -Splash - tk 13:30, 5 September 2006 (UTC)

I also object. I did not comment on RfA/Carnildo3; I can't recall having anything to do with this editor; so perhaps I'm neutral on him. I have been occasionally active on RfAs. This editor may be rehabilitated, a fine admin -- or not; I have no idea. I do see that the RfA was contentious and the tally borderline. My instinct in such matters is to err on the side of caution; I would invite the candidate to reapply in a few months and see if community sentiment changes. Guesstimating from the change since RfA/Carnildo2, clear consensus may well form in favor -- then.

I definitely don't like the precedent set by "temporary" adminship. Has this ever been done before? Has the community expressed any desire for this measure in general? Where does this door lead? B'crats are expected to make tough judgement decisions; it's routine. Is this a creative solution or a failure to make the call? Also, exactly under what circumstances and by what method will Carnildo's sysop privs "expire"? Who will make the decision to revoke or extend them? Sorry, but evaluating admin performance is not a b'crat function; it is ArbCom's or a steward's, if not the community's. John Reid 15:14, 5 September 2006 (UTC)

I was not involved in the decision that led to Carnildo's resysopping, but to clarify: his adminship is not temporary exactly. It is under probation. This means that, after a given period of time (two months), the ArbCom, not the Bureaucrats, will review his record as an admin for this period of time and decide on whether or not to remove his admin bit again or let it stay. If the decision is made to remove his adminship again, a member of the ArbCom will post on Meta, at a forum called Request for Permissions, where a Steward will take care of the request and remove the flag. Redux 16:02, 5 September 2006 (UTC)

It's worth noting that RfA doesn't own bureaucrats either. Presumably they are empowered to take actions that in their best judgment are for the good of the project. Christopher Parham (talk) 04:27, 6 September 2006 (UTC)

There is no precedent for probationary adminship and I dislike to see it set by fiat. I object to the term "own"; it is an inappropriate straw man. RfA determines who will and who will not become an admin; b'crats are empowered only to implement community will -- not to paternalistically watch over the community and do what's best for us. If we want a probationary admin, we'll ask for one.

This is a very old, well-established principle: Power is divided between the rank-and-file on one hand and Jimbo and the Board on the other. ArbCom is a comparative novelty; power is expressly delegated to this body. Power is not delegated to admins or b'crats -- not in any way. These classes of users exist solely to implement community will.

If there are any b'crats who don't agree with that last statement, please speak out clearly right now. John Reid 11:32, 7 September 2006 (UTC)

I too am concerned about the Carnildo decision, and am reposting my comments here from Wikipedia talk:Requests for adminship/Carnildo_3. Discussion has also reached User_talk:Carnildo and WT:RfA.

The role of bureaucrats is to gauge community consensus. WP:CRAT says that bureaucrats may grant administrator access "only when doing so reflects the wishes of the community", WP:RfA says that bureaucrats "review the discussion to see whether there is a general consensus for promotion", and Taxman, one of the three bureaucrats behind the decision to promote Carnildo, acknowledges that "As a bureaucrat it is my job to determine consensus in RfAs." Yet the decision makes no mention of consensus, and only references community opinion to say that while many users oppose Carnildo's adminship, he is being promoted anyway.

The threshold for consensus was not met here. Although RfA is not a rigid vote, consensus tends to be gauged by percentage, with bureaucrats "generally held to a 75-80% discretionary zone", in Taxman's words. Carnildo's RfA closed with 112 supports and 71 opposes, a 61% ratio. The true ratio may be even lower due to users (like [1] and [2]) not bothering to oppose when they saw that the RfA was already well below the threshold for promotion.

There has arisen a dangerous misconception that when deciding whether to promote, bureaucrats are permitted to "weigh the arguments" or make arguments of their own at the expense of community consensus. Neither WP:CRAT nor WP:RfA gives them such authority. Rather, it is the job of the community to make arguments and weigh each other's arguments during the discussion. The job of bureaucrats is strictly to gauge consensus. Of course, they cannot do so by mechanically counting supports and opposes, but must watch carefully for sockpuppets and other consensus-obscuring factors. And of course, bureaucrats are part of the community too, and in their role as community members, may participate in the discussion themselves. But in their role as bureaucrats, when deciding whether to promote, they are bound by the community's will.

In addition to lacking consensus, this decision had other problems. First, although bureaucrats may participate in RfA discussions, it is best for impartiality that they not close discussions in which they have participated, unless the outcome is uncontroversial due to an obvious consensus or non-consensus. Danny, who supported the RfA, nonetheless holds "primary writing credit" for this very controversial decision. Second, transparency requires that bureaucrats discuss issues of promotion or non-promotion openly. Taxman pledged to do so when he became a bureaucrat, but now admits that with respect to the current decision, of which, as far as I know, not a peep was made prior to its surprise announcement, he "did different from what [he] said [he] would", a "mistake", in his words.

In short, the decision to promote Carnildo was made without transparency and without consensus. If the bureaucrats believed that "special consideration should be given to the extenuating circumstances of this case", they could have presented that view to the community and sought consensus in the ensuing discussion. Instead, they issued a highly irregular decision without public consultation. It is disappointing and worrying to see trusted users exhibit such disrespect for the community which granted them their positions, and I urge them to recant. Tim Smith 14:37, 8 September 2006 (UTC)

• I think what transpired is that Carnildo was put up for RfA by User:UninvitedCompany. Word rapidly reached the members of ArbCom, sparking a discussion on what to do. Three members of ArbCom that voted to have Carnildo's sysop powers removed subsequent to the pedophilia userbox war came out in support of the RfA in the first day. As the RfA's time period waned, it became obvious that the RfA would not pass. Discussion ensued between some bureaucrats and some members of ArbCom on what to do. A decision was made to promote Carnildo, and use the closing RfA as the vehicle for doing so. I could be wrong in some of the above; it's speculation.
• In my opinion, what should have happened is the RfA should have been allowed to fail. Subsequent to that, a case should have been submitted to ArbCom re-opening Carnildo's role in the pedophilia userbox war, to re-evaluate whether the desysopping should have been permanent, or temporarily suspended.
• What should happen now, and so far (to my knowledge) has not happened is a full accounting of exactly what transpired. This is a highly controversial case, and transparency is being demanded by a number of users. If the case can not be presented in a convincing way that the community can support, then it would seem there would be basis for overturning the decision. Mistakes can and do happen. In the least, some statement by the bureaucrats as to why this case should or should not be a precedent for future considerations and what impact this decision has on future (and some would ask for) past RfAs should be made.
• The bureaucrats didn't just make a controversial decision. They shattered the mold that RfA has held to for years. To date, this has been without satisfactory transparency or explanation. I am not going to say the bureaucrats made a mistake. I've been witholding personal judgement on that pending explanation by the bureaucrats. However, such explanation has not been forthcoming. This is...disappointing. --Durin 15:50, 8 September 2006 (UTC)

Durin's speculation as to the chain of events is interesting but not wholly accurate. Please note that Carnildo's nomination was offered by UninvitedCompany, a former Arbitrator and a member of and regular participant in the Arbitration Committee's private discussion list, the same as I. UC specifically raised the issue in our discussion list, and the matter was discussed at some length. UC's nomination was after this discussion and the Arbitration Committee was appraised of, and generally not opposed to the nomination, in advance of it being made. The nomination was not a surprise to the Committee, as Durin suggests. Several of us (acting as individuals, but with the gravity that a members or former member of the Committee has) deliberately supported the nomination early on.

The decision to promote him was made by the three bureaucrats who claimed to make it. The Committee was not (as a whole) consulted on the question as to whether he should be promoted. One member of the Committee objected after the fact to the promotion, although that member's opinion appears (from the discussion that followed) to be a minority of one within the Committee. I was present for the tail end of the discussion to promote Carnildo, but by the time I joined the discussion the decision had been made; I played no role in that decision. I did not represent the ArbCom in that discussion. I am not, at this time, at liberty to name who else was present, although I will state that at least one member of the current ArbCom was present at the time I joined the discussion.

It is my belief that if the ArbCom were asked to consider whether Carnildo should be returned to sysop status, prior to the RfA, the Committee would likely have agreed to do so. Even after the RfA was completed, I believe the Committee would have done so. I believe that I am not alone in this evaluation of the Committee's likely action, and that any member of the Committee would likely come to the same conclusion after reasonable consideration. I believe the three bureaucrats who decided to promote Carnildo, were advised of this state of affairs by representative(s) of the Committee, and concluded that promoting Carnildo on their authority, rather that on the Arbitration Committee's, was in the best interests of the project at that time.

I see that someone above has made some starkly legalistic argument about bureaucrats and sysops not being delegated their authority from the Board and therefore restrained to act only as agents of consensus or some such folderol. That is, frankly, nonsense. Sysops and bureaucrats, as with all editors on the English Wikipedia, are charged to act in the best interests of the encyclopedia. Consensus is a tool we use to gauge the best interests of the project, but it is not an end in and of itself. I applaud these three bureaucrats for having the moxie to break from the stifling expectations of the pseudoconsensus that typically erupts from any given Request for Adminship and instead evaluate the broader picture and make a decision that reflects more than merely the shifting moods of a fickle and ill-informed populace. Kelly Martin (talk) 04:09, 11 September 2006 (UTC)

Perhaps I'm that someone. There is nothing "legalistic" (i.e, BS) about my position; community control of admins and b'crats is a core principle. B'crats do not have the power to decide who will be promoted to adminship or b'cratship; neither does ArbCom. Only the community has that power. ArbCom does have the power -- granted by an unsteady meld of community consensus and Board fiat -- to deadmin; they exercised that power in this case. Other admins involved in this case were temporarily deadminned; Carnildo was simply deadminned and invited to reapply in due time. That's all. ArbCom is out of this.

There is a growing trend among admins to assert authority within this community. Yet it is a core principle of our community -- our social model -- that admins have no authority, only the responsibility to carry out community will. Their authority, such as it is, is strictly limited to understanding and interpreting general expressions of community will and applying them to specific cases. When the community speaks directly to a specific case, admins have no latitude to ignore us. I object in the strongest possible terms to the creeping arrogance infecting the admin group. This threatens to destroy the community and convert it into a hive of worker bees controlled by straw bosses.

What goes for admins goes double for b'crats. The least sign of autocracy among the b'crat group must be stamped out, lest admins be promoted against community consensus who themselves lean toward autocracy. Consensus does not come before the encyclopedia; it is not a mere tool but a core principle of our little society.

In my last comment I called upon dissenting b'crats to come forth and so state. Kelly Martin is not a b'crat so I will not hold her to this higher standard of accountability. Now I call on all b'crats to formally state that they continue to serve community needs by acting in accordance with expressed community consensus. John Reid 09:35, 11 September 2006 (UTC)

Admins serve the encyclopaedia, not the community. They are charged with acting in the best interests of the encyclopaedia, not the community. The fact that they are selected by the community does not mean they have a community-given mandate which may be withdrawn as the community chooses. The community is merely expressing its view that the candidate concerned will be able to act in the best interests of the encyclopaedia; community consensus is meant to be a practical and pragmatic solution to the problem of selecting the best-qualified admins. Recall that prior to RfA, we selected admins only based on the views of a few people equivalent to 'crats. Today, for reasons of scalability, we have RfA.

If we must compare WP to a democracy, as some people seem to enjoy doing, let us recall that representatives in Congress/Parliament and the President/Prime Minister are not charged with enacting the will of the people, but doing what will best serve the people. Congressmen do not vote on motions as they think their constituents would, but vote as they think would best serve their constituents. The analogy breaks down, of course, when one considers that in a democracy, the government is subordinate to the will of the people. Wikipedia is not a democracy. It is closer to an encyclopaedia publishing company, and there is a clear hierarchy which must be followed. The fact that the shareholders elect the board of directors doesn't imply that the board of directors must hew to every whim and fancy of a majority of the shareholders. Johnleemk | Talk 12:16, 11 September 2006 (UTC)

• I think we need to pull this back and constrain the discussion to bureaucrats and RfA. We've extended the discussion to be a debate about the entire project. This is really not the place for such a discussion.
• There seems to be two sides to this; bureaucrats are charged with either (a) implementing the will of the community as expressed by the consensus garnering mechanism known as RfA or (b) acting in the best interests of the encyclopedia, even if that act is contrary to community will. At a guess, I would venture to say that the majority view is the former, and not the latter. There's the divide. The divide is further exacerbated in that the people who advocate (b) are generally regarding the opinions of the people who advocate (a) as worthless. Further, the people who advocate (a) regard the people who advocate (b) as power mongering and cabal creating people. Both sides are right, and both sides are wrong. At this juncture, neither side seems interested in the possibility of compromise. --Durin 13:09, 11 September 2006 (UTC)

Indeed, this is the usual division within Wikipedia: the division between the people who view Wikipedia as an encyclopedia being written by a community and the people who view Wikipedia as a community that happens to be at the moment writing an encyclopedia. The problem with the people in the second group is that the Wikimedia Foundation has a charter that really rather prohibits Wikipedia from turning into (for example) a K/S novel, no matter how much the community may wish for such a thing to happen. The encyclopedia is and will always be paramount, and any wishes the community should have, no matter how strongly held, that are inconsistent with that are simply wrong and must be ignored. The community is not empowered to decide to stop writing an encyclopedia; or, rather, if it should do so, it will no longer be welcome to participate in the project. We are all servants of the project, not of the community.

Every debate about the proper role of editors, including that administrators and bureaucrats, ends up being a reflection of this fundamental division. You can't get away from it by wishing it away. Kelly Martin (talk) 13:31, 11 September 2006 (UTC)

• Nobody is wishing it away. To conclude the people who advocate (a) as being in opposition to the goals of the project is improper. Both groups are interested in the success of the project. People who advocate (a) do not necessarily view Wikipedia as a community that happens to be writing an encyclopedia. --Durin 13:42, 11 September 2006 (UTC)

No, you're wrong about that. People who advocate (a) are not committed to the project. They are committed to the community. The fact that the community at this time happens to be writing an encyclopedia is merely a coincidence, and if they are truly advocating (a), then they will happily go along when the community decides, by consensus, to instead change goals and start writing K/S fanfic. Now, you can, of course, argue that "well, of course the community will not do that", but that's not the point. The point is whether your committment is to the encyclopedia, or to the community. And I argue that the people advocating (a) are committed to the community and to whatever the community, in its ever-shifting moods, decides it feels like doing, rather than inerrantly committed to writing an encyclopedia no matter what butterfly the community decides to chase this week. And to that extent I am convinced that the people advocating (a) are not merely simply wrong but actually harmful; not only do they dilute our focus and encourage us to waste time on irrelevancies, but generate friction when they refuse to allow the car to be steered back on the road simply because the community has decided to go drive off after pretty flowers. Simply put, the people advocating (a) are only committed to writing an encyclopedia because currently the community still has something of a consensus for doing so. However, that consensus is fading fast and will continue to fade as the community grows.

We have large segments of our community who don't do anything to write an encyclopedia. There are segments that are writing advertising copy. There's a segment that is writing a sports gazetter (masquerading as encyclopedic coverage). There are segments that are writing a social networking site. I shouldn't be surprised to find out that there's a segment somewhere that is writing interactive fiction somewhere in some corner of Wikipedia. Our community is absolutely not unified in its desire to write an encyclopedia. Anyone who thinks it is simply is not paying attention, or is stuck in 2003 (when there really was such a consensus).

We have not a small number of people whose participation in Wikipedia is solely so they can become admins (because being an admin is cool, ya know). We have people who are here only to pimp their wares. We have people who are here only to primp their egos. And the people advocating (a) would give all of these people just as much license to decide how this encyclopedia project should be run as they would the people who are actually writing an encyclopedia. Kelly Martin (talk) 15:15, 11 September 2006 (UTC)

• Thanks for your input Kelly. Per earlier disagreements and their outcome, I am exiting this conversation with you. For the record; I strongly disagree with you, but will not debate you on these points as you and I working together is like oil and water. --Durin 15:28, 11 September 2006 (UTC)

From where I'm standing it looks like >95% of the people who participate in RFA are committed to help build an encyclopedia. People who have goals not in alignment with that tend not to last long around here and tend not to participate in processes like RFA. Haukur 15:26, 11 September 2006 (UTC)

Kelly you are putting up a straw man. People, like myself, who believe that RFA should be about implementing the will of the community don't automatically believe that the total direction of the project as a whole is a matter of consensus. Though thank you for making clear the root of your objection, which appears to be that you don't like or trust Wikipedia's current community. Dragons flight 15:38, 11 September 2006 (UTC)

It is helpful that a knowledgeable participant has shed some light on the murky internal machinations behind the recent debacle, rather that us being left to speculate for ourselves. From Kelly Martin's first reply above, it looks like she is saying that the ArbCom had decided that Carnildo should be re-admined, and thought the RFA process was the least painful way to achieve it (the alternative being to recind part of an earlier ArbCom decision). Certain former and current members of ArbCom piled on the support votes "in a personal capacity", but with their customary "gravity", to achieve that result. When the RFA went pear-shaped, there was a discussion between certain bureaucrats and certain members of the ArbCom, informing the bureaucrats that the ArbCom would be minded to re-admin Carnildo anyway, at which point the bureaucrats decided bite the bullet to re-admin him, notwithstanding that the balance of opinions expressed in the RFA fell outside the usual range within which they would, based on past practice, have decided that there was consensus for the RFA to pass.

Perhaps this is an application of WP:IAR, with the ArbCom and bureaucrats know the "right thing" when they see it; perhaps the views of current and former members of the ArbCom have a heavier weight (or "gravity", should I say) that us common-or-garden peons; perhaps this is just an exceptional case and no conclusions should be drawn from it.

Now, where did I leave that FAC I am writing... -- ALoan (Talk) 15:30, 11 September 2006 (UTC)

I've been active here for a shorter time than most who will be reading here, but I have to comment on what's been written above, although I'm not sure whether this essay belongs here or not, but it was spurred by Kelly's comments so I'll go ahead here.

From what I have seen since I've been here and what I've read of past disputes and decisions, on balance "the community" has gotten a lot more right than it's gotten wrong. I define "the community" here as the group of people who participate in both the article-writing and the decision-making functions -- the people who are writing the encyclopedia while also participating in some of the meta-decisions about how it should be written. There are times when leadership is needed, and there have been times on AN/I when I (a non-admin but an interested bystander) have commented that a given sensitive situation is something that should be resolved off-wiki. Those situations exist, and mechanisms are needed for dealing with them, but they are rare.

On balance, community decision-making means decision-making by those users who care to participate in a given process. Those who choose not to participate are putting their trust, generally safely, in those who do. I don't participate, for example, in choosing featured pictures. That doesn't mean that I am indifferent to whether the picture on tomorrow's main page will be attractive and informative or ugly and useless; it means that I think there are other users more qualified than I to make, and more interested than I in making, that particular decision. I don't seek to participate in bot approvals, because I know nothing about the technical aspects of bots. On the other hand, I do participate in RfA, because I believe that my individual abilities lead me to be able to provide useful comments in that particular arena. People say that there isn't enough participation in RfA, but I don't think we want one million casual users weighing in on admin candidates -- nor do we want them chosen by a small committee. Editors who are choosing featured articles, or commenting on AfD's, or just to do their editing, have made their own choice about where their talents and interests lie.

I will also mention in passing that I've seen no evidence that we have a problem with people becoming administrators "because it's cool." To be sure, users may spend their time in areas such as XfD and RC patrol because participation there is significant to many RfA voters; I've debated whether I wish to be an admin at some point (I'm not sure), and I know that if I do, I will have to get involved in those areas even though they are far from the major contributions I can make to this encyclopedia with my own interests and abilities. But the net result, if my efforts were to be redirected in those directions, would be that I would be attending to responsibilities where the encyclopedia needs more resources. Who's to say that pushing the type of intensive user who is considering adminship in those directions is a bad thing. Similarly, I disagree with 1FA voters but if the net result of their position is more FA's, I don't see that as a threat to the project.

The distinction between "the community" and "the encyclopedia" may have a kernel of truth to it, but as it's been presented, it is an entirely false dichotomy, and needlessly divisive. I think of Wikipedia as an encyclopedia that is written by a community -- a community that I am proud to be a part of, that is working together in a grand experiment of volunteer effort to do something that has never been done before. Building such an effort requires teamwork, comeraderie, and a sense that the participants are valued and, collectively, have significant input into the future of the enterprise. It also requires, as Jimbo recently noted, that the contributors be having fun.

One individual decision, on a promotion or anything else, is insignificant in the bigger picture, and I have not joined with anyone who's claimed that if the bureaucrats are allowed to do X or the community doesn't overrule decision Y, the project is worthless and we must all down tools. I'm willing to accept that once in awhile, there may be something that the rank-and-file users don't know. But if highly experienced members of the project, who esteem themselves and present themselves as leaders, proclaim that support for the general principle of community decision-making is a rejection of the project of writing an encyclopedia, I suggest that they could not be more wrong. Newyorkbrad 16:03, 11 September 2006 (UTC)

That sums up my feelings perfectly. the wub "?!" 22:07, 11 September 2006 (UTC)

Well, maybe. You are right in what you say, that in the end we can't have a cabal running things, but actually I think most people may have forgotten that adminship "should be no bog deal". If Carnildo abuses the tools - and he will be watched like a hawk - then the tools will surely be withdrawn again, just as we allow tendentious editors to be unblocked and have a second chance, but we block them if they abuse the privilege. I see plenty of possible benefit to the project from Carnildo being able to delete images. Carnildo is one of the best-informed in regard to copyright issues, and most copyight violators have no intention of paying our nlegal bills if their invalid fair-use claim gets us sued. I don't see much disbenefit of Carnildo having the sysop bit. Not now the userbox wars are over. So: give Carnildo a second chance. Assume good faith. Guy 23:24, 11 September 2006 (UTC)

Speaking only for myself, and not on behalf of any of the others who've posted above, I was commenting on general principles, and not on any one specific decision. Just in case it wasn't clear. Newyorkbrad 23:36, 11 September 2006 (UTC)

We should probably trust the bureaucrats to determine community consensus. Obviously if they only ever added up votes they would be failing the community and failing the encyclopedia. --Tony Sidaway 23:30, 11 September 2006 (UTC)

To Protect and Serve

I agree with Tony Sidaway and indeed I'll strengthen it: We do and even must trust b'crats to implement community consensus. Perhaps Tony and I part ways over his essay; I read it as a call to give b'crats greater latitude, which I do not endorse. But I do agree that RfA is not a vote and b'crats need to use wise discretion in sorting the puppets from the people. That said, there is a certain standard set for promotion. With obvious puppets discounted and reasonable participation, RfA outcomes above a certain percentage pretty much demand promotion; RfAs below a lower threshold forbid it. It is only within the gray area between thresholds that we ask b'crats to go beyond combating "election fraud" and make a close call.

The particular RfA that sparked this thread fell well under acceptable limits for promotion. The fact that ArbCom members lobbied in favor contaminates the discussion, rather than enhances it. That is to say, the suspicion of undue influence requires the closing b'crat to cast a particularly critical eye on an outcome so weighted.

Pompeia Sulla. Precisely because b'crats are entrusted with such an important privilege and with the community mandate to exercise it with discretion, no b'crat can afford to appear biased or subject to pressure from any direction. Admin promotions come at the behest of our community only.

No personal offense intended, but remarks such as We are all servants of the project, not of the community make me shudder.

Each one of us in his role as editor is a servant of the encyclopedia project and I hope that no experienced editor will disagree. We have many different viewpoints on just what an encyclopedia may be; personally, I say that the word itself is becoming self-referential, since for most purposes, Wikipedia is the only general-reference encyclopedia most human beings will ever use. But that's another topic. Those of us fortunate enough to have spent considerable time with printed, bound, multi-volume encyclopedias have a very clear idea what is meant by the word and we hope to pass this value along to the next generation of editors. It is a core value, a foundation issue, and absolutely non-negotiable -- although very difficult to define.

The Board serves our project directly and is empowered to ignore our community, for good or ill. Jimbo is above any law, at present. Those of us who have been promoted to positions of trust -- admins and b'crats -- do not serve our project in these capacities. They serve our community; our community serves our project. There is no value conflict whatever.

The danger is that our trusted servants may decide to serve our project directly, bypassing our community. This is in direct conflict with Sidaway's law of Wikipedia: Wikipedia is cleverer than you are. We do our best to elevate our wisest members but b'crats are still flawed humans. We trust them to interpret and implement community consensus. We do not trust them to be smarter than our community as a whole.

I renew my call for each individual b'crat to declare his or her position on this topic. I want to hear it. John Reid 01:54, 12 September 2006 (UTC)

Seconded. It is notable that the discussion on one side seems to be led by Tony (and, to a lesser extent, Kelly), neither of whom is a bureaucrat, whereas the bureaucrats have remained almost silent. It would be very nice to see what they think of this decision, whether they think it implements policy as it was understood before it or is merely an exercise of discretion, and whether it has any influence on how RFAs will be handled in future. -- ALoan (Talk) 12:48, 12 September 2006 (UTC)

• Frankly I'm rather disturbed that the bureaucrats have gone silent. When I had a moment, I was going to do a thorough check, but it appears that not a single bureaucrat has contributed to the discussion here or at WT:RFA for the last two days and maybe longer. --Durin 13:36, 12 September 2006 (UTC)

I've added a section heading here, to improve continued editing. I disagree with your assertions that the Admins and Bureaucrats serve the community and not the project directly, nor do I feel it conflicts with Wikipeida is cleverer than you are. In fact, it's precisely that reason I trust them in the positions they're in to act for the project first and community second. I don't believe their actions, even the Bureacrats, are completly irreversible - which is the underlying flexability of Wikipedia which allows it to survive. Whilst flawed humans they may be, they're not without oversight, so I don't expect them to outthink the community as a whole - I don't expect anyone to, even Jimbo.

I don't expect my comments to change your mind on this, but I just wanted to offer a similar yet opposing viewpoint. I'd be happy to discuss my thoughts on it further but I have no idea where Tony or Kelly would fall on this. --InkSplotch 23:33, 12 September 2006 (UTC)

The historical role of bureaucrats is to promote candidates for adminship when authorized to do so by the community. Speaking from a standpoint of roles, User:Ed Poor was the first to do this, and carried out his responsibilities for a considerable span of time using SQL queries before a user interface was constructed and the term "bureaucrat" applied. As such, the role predates RFA, it predates voting and percentage thresholds, and predates consensus (the original standard being no significant objections). I don't think the role has changed significantly, though it has been formalized, and the community has made the means by which it authorizes promotion clearer and more objective. The role has expanded somewhat with time, in that bureaucrats now are the sole closers of RFAs, and the matter of bots and name changes has been added. The Uninvited Co., Inc. 04:10, 13 September 2006 (UTC)

Renewing my call to suspend RfA

A couple of months ago I called for the bureaucrats to suspend RfA. I'm renewing that call. I am asking all bureaucrats to refrain from promoting new admins until such time as the community develops a process for promoting admins that isn't broken. The current process is too capable of promoting unqualified candidates, excludes qualified candidates for thoroughly irrelevant reasons, and often generates hostility and bad blood. But as long as it functions at all, there's no incentive to change it. An agreement by the bureaucrats to suspend promotions would force the community to actually deal with the issue, instead of blithely ignoring it. The Sean Black and Carnildo RfAs are great examples.

At least give it some thought. Don't dismiss it out of hand, the way you did last time. Kelly Martin (talk) 08:15, 10 September 2006 (UTC)

I strongly agree that RfA is broken and that it needs replacement. I strongly disagree that this is the place to propose turning it "off". B'crats are not given that latitude, sorry. I will surely entertain a proposal to overhaul RfA and I pledge to work with any serious proposal. I oppose any b'crat sit-down strike.

I'll repeat: In regard to promotion, the sole job of b'crats is to implement community will -- not to determine, control, or correct it. John Reid 13:33, 10 September 2006 (UTC)

And I disagree. Bureaucrats are given that latitude. They are expected to act in the best interest of the project; they are not mere automatons who exist only to promote people who meet certain numerical standards. Kelly Martin (talk) 14:47, 10 September 2006 (UTC)

The role of bureaucrats is to gauge and implement community consensus. WP:CRAT says that bureaucrats may grant administrator access "only when doing so reflects the wishes of the community" and WP:RfA says that bureaucrats "review the discussion to see whether there is a general community consensus for promotion". It is the job of the community to determine the best interest of the project by making arguments and weighing each other's arguments during the discussion. The job of bureaucrats is strictly to follow community consensus. Of course, they cannot do so by mechanically counting supports and opposes, but must watch carefully for sockpuppets and other consensus-obscuring factors. And of course, bureaucrats are part of the community too, and in their role as community members, may participate in the discussion themselves. But in their role as bureaucrats, when deciding whether to promote, they are bound by the community's will. Tim Smith 16:23, 10 September 2006 (UTC)

Not quite. Whatever our written policy may say, Bureaucrats may not, for instance, promote a clearly unsuitable administrator even if it's the community's will. Ignore all rules applies to bureaucrats, too. --Tony Sidaway 16:41, 10 September 2006 (UTC)

Let's also keep in mind that the sampling on RfA is not very reliably a "consensus" of the community. The largest RfA I've seen had some 175 votes. There are over 10,000 active editors on the English Wikipedia. Frankly, I think a sampling of less than 2% of the community can't really be characterized as a "consensus". I know that I don't participate in RfA not because I think it's unimportant or that I don't care, but because I find the toxicity of RfA sufficiently unpleasant that I don't want anything to do with it. My opinion is not counted because I find the process unpleasantly repulsive, not because I don't care. I wonder how many other editors are avoiding RfA because they find the process distasteful; I doubt I'm the only one. Kelly Martin (talk) 17:28, 10 September 2006 (UTC)

Kelly actually participated in five RfAs during the past month ([3] [4] [5] [6] [7]), the last just hours before the above post. Consensus on Wikipedia—about an article, an adminship, or anything else—rarely involves a sizeable percentage of active editors. In practice, because RfA is open to the community, the consensus of editors in a discussion is taken as a proxy for the consensus of the community. It's true that sampling irregularities can skew the result, and bureaucrats must watch carefully for such consensus-obscuring factors. But their job remains to gauge and implement community consensus.

To address Tony's example, different people have different standards of suitability for adminship. If community consensus is to promote, then the candidate is suitable in the eyes of a large majority, and unsuitable only in the view of a small minority. In that case the closing bureaucrat may indeed promote. In fact, that commonly happens. If the bureaucrat decides instead to ignore all rules and not promote, then they must answer to the community for doing so. Tim Smith 21:04, 10 September 2006 (UTC)

I join Kelly in urging the bureaucrats to seriously consider suspending the process of making new administrators. An alternative to this would be for bureaucrats to routinely close out RFAs in a more conscientious manner, instead of just counting up sacks of me-toos from the RFA groupies. But I think that might require an expanded corps of bureaucrats to cope with the workload. --Tony Sidaway 14:56, 10 September 2006 (UTC)

What makes you think bureaucrats are not already closing RfAs in a "conscientious manner"? Angela. 15:02, 10 September 2006 (UTC)

I'm sorry if I didn't make myself understood. My point is not that they're not conscientious (they are), but that perhaps routinely increasing the amount of work being put into closes, distinguishing good reasons for oppose or support from poor ones, listing editors who contributions are against the spirit of RfA (such as those who always vote support), might be an alternative to Kelly's proposal. I also recognise, you will see, that this would entail substantially more work and might require more bureaucrats. --Tony Sidaway 15:11, 10 September 2006 (UTC)

Angela, I certainly do not believe that the bureaucrats are acting unconscientiously. Rather, I think the current structure of RfA itself engenders harm to Wikipedia, and that bureaucrats should refuse to participate in it in order to put an end to that harm. Kelly Martin (talk) 15:34, 10 September 2006 (UTC)

I think there's consensus (that word again!) that there are times RfA isn't working smoothly, though no consensus at all about how to improve it. But this suggestion is not a solution to anything. We have a pending RfA in which the votes expressions of opinion are running something like 90 to 0. Are you suggesting that the closing 'crat should refuse to promote, in a clear case where RfA worked fine, to draw attention to the fact that in other cases it doesn't? Sounds highly counter-productive, and a bit like a WP:POINT situation to me. Newyorkbrad 15:04, 10 September 2006 (UTC)

I think the suggestion is that all promotions should be suspended. --Tony Sidaway 15:13, 10 September 2006 (UTC)

Kelly's suggestion was that the bureaucrats "refrain from promoting new administrators" and "suspend promotions" until the RfA process is revised, to "force the community" to come up with a new process. If that is done, then even the uncontroversial promotions in cases where there's no dispute about application of the RfA process at all will be held up. I still don't see the value to that. Newyorkbrad 15:20, 10 September 2006 (UTC)

Many of the "uncontroversial" candidates are uncontroversial because nobody asks the right questions. We've had uncontroversial candidates who turned out to be bad admins, after all, and a lot of people who make admin these days do so because they've deliberately gamed the system in order to become admins. Kelly Martin (talk) 15:34, 10 September 2006 (UTC)

If the right questions aren't being asked, you are as free to ask a question or present a comment as anyone else. Newyorkbrad 17:12, 10 September 2006 (UTC)

The value is obvious. Produce a non-broken process. --Tony Sidaway 15:33, 10 September 2006 (UTC)

I think the 'crats are doing the best that they can and are attempted to act in a manner that is both professional and effective. If a process is screwed up, it doesn't mean the 'crats are. They are human beings capable of seeing beyond simple numbers or a list of names. They may mess up, but everyone does. I think the 'crats should be and can be trusted in the vast majority of circumstances. Yanksox 15:15, 10 September 2006 (UTC)

Kelly, closing the RFA is not the best solution. I agree with you that the RFA has become a mess, and the quality of admins in the past are much better than some of the present day ones. So why is this happening? I have a few thoughts: Well, power corrupts and as the size of admins increases it's becoming more of a cabal and we find many users shooting down valid candidates on really flimsy reasons. I've noticed that the RFAs that have the highest chances of passing are the ones which have been nominated by well known wikipedians. Calling and end to the process without proposed solutions is certainly not the best way to move ahead. Why don't you give 'crats the mandate some more elbow room in subjectivity? Either way if we go ahead (as Splash has put it above), we're accused of being partial or not considering 'legitimate' concerns. My concerns are mainly the fact that if just *one* well known user objects, you're going to get a whole army of mechanical "nodders" agreeing with the editor, without going in the reasons why the candidate may have acted in that fashion and the opposer could be wrong. What do we do then? I know that the candidate may have just got a little hot one incident, so can it be integrated for all instances in the future? I certainly don't feel so, but if we do promote, there are accusations of us having ulterior motives and what not. Please do suggest some alternatives before calling for a suspended session. =Nichalp «Talk»= 18:08, 10 September 2006 (UTC)

Here’s an alternative that makes sense to me. Most of the comments/concerns are centered around voting and unreasonable oppose rational, and many are suggesting giving Bureaucrats more discretionary power. Instead of turning RFA off, why don’t we try something like this for the next 5 nominations? See what happens. Rx StrangeLove 18:27, 10 September 2006 (UTC)

By sheer coincidence, as it were, recently I started a new guideline called Requests for adminship is not a vote. It has effectively been incorporated in Wikipedia:Requests for adminship/Front matter. It clarifies that bureaucrats have wide discretion in determining community consensus. --Tony Sidaway 18:35, 10 September 2006 (UTC)

Having contributed to the current language, I'd like to say that I consider it a reflection of the existing system and not a license to expand bureaucrat discretion. In my opinion bureaucrat discretion is and should be narrowly construed except when exceptional circumstances exist. It is important to the sense of community involvement and fairness that the closing of average/normal RFAs be handled in a way that is transparent and predictable. Also, I would be very dissappointed in any Bureaucrat who felt that RFA should be suspended in the absence of a community mandate to do so. Bureaucrats are trusted to implement community standards and expectations; they are not empowered to force their own views upon the community. Dragons flight 19:44, 10 September 2006 (UTC)

I am unconvinced that suspending RFA will solve anything. However, I do once again urge everyone who is interested, and most particularly Tony Sidaway and Kelly Martin, to participate in earnest in the Wikipedia:WikiProject on Adminship or otherwise start a discussion independent of RFA itself. I believe that a major barrier to any sort of change at RFA is that no demonstrably better alternative has been presented in sufficient detail to withstand scrutiny. The Uninvited Co., Inc. 20:21, 10 September 2006 (UTC)

• This is in part because no concerted effort has been made to determine just what is wrong with it and what the goals of such an effort would be. Lots of people say its broken (of course, lots say it isn't). Fine. You can say its broken. But, without any idea of what is wrong with it and what the goals are, you're just taking shots in the dark and are likely to make it worse, not better. --Durin 01:21, 11 September 2006 (UTC)

It's obviously broken because it's changing rapidly in unexpected directions. Hopefully it will become unbroken pretty soon. Perhaps those who are changing it know what they're doing, perhaps not. --Tony Sidaway 01:27, 11 September 2006 (UTC)

• I'm not arguing whether it is broken or not. --Durin 02:32, 11 September 2006 (UTC)

The brokenness of RfA is directly connected to our failure to communicate our core values to new editors. This is such a massive failure that most of us are unwilling to address it in the whole. Indeed, this may be impossible; perhaps all we can do is chip away on the edges. The explosion of policy proposals is another effect of an influx of editors who are poorly informed or unwilling to uphold core principles and foundation issues. This failure threatens to overshadow every other issue.

The UBX war is a case in point. Some anti-boxers feel that pro-boxers are only amusing themselves or that they come here in order to shove their biases in total disregard of neutrality. Unfortunately, far too many kiddie editors are here for just such purposes and they do create and use kiddie boxes to serve these ends. My personal belief is that sometimes UBX serve important social roles; they act as a social binder, an occasional lubricant, and are a practical tool for reputation management -- they are small change in our reputation economy. However, this viewpoint of mine is overshadowed by the hordes of kiddie editors slapping kiddie boxes everywhere because we can and it's free speech, man. The discussion of whether UBX serve community needs, thus indirectly but importantly serving the project, has tended to degenerate into the question of how to deal with kiddie boxes. Those of us trying to promote a moderate policy are caught in the crossfire between flesh-eating zombies and exhausted, embattled shooters.

Thus, our community has been infected with two opposing camps of internal enemies. The first are those who join up and start editing without any grasp of core values -- and who are taught all the wrong lessons and none of the right ones. The second are editors who understand these core values all too well but who, in reaction to the first, determine to abandon some values in order to protect others. When one group or the other have killed out all the moderates (if you're not with us, you're against us), there will be nothing left except total war and destruction of the project. Choose your nightmare: MyWiki, NuNupedia, BoardWar, Invasion of the Lawyers, WP Brand Inc., or Jimbo Goes Postal.

I do not have a solution to this most pressing of all issues. Sorry. Heck, I'll admit it: I do consider myself smarter than the average monkey. But this has me stumped. I tend to think that our foundation principles are internally inconsistent, that at least one must be abandoned to accommodate inevitables. But I also suspect that to remove even one such brick will be to bring the whole house down. I just don't know.

So long as editors in general either do not embrace core values or willfully abandon them, naturally admins will do no better -- and a lifetime of experience tells me that bad apples tend to float up in hierarchies. No method of promoting admins can draw gold from straw.

I do not think RfA can be fixed. I think it works now as well as it can now. If this failure to communicate our core values to new editors -- to attract and retain editors who share core values -- is reversed into success, then existing process will start working again. If this failure persists and worsens, then it really doesn't matter what else we do. John Reid 02:43, 12 September 2006 (UTC)

Well put. Clearly there needs to be a better process for RfA than the one that is in place now (there seems to be community consensous for a change), however, how to go about re-factoring RfA is probably not going to be an easy nor agreeable task. // Pilotguy (Have your say) 20:27, 12 September 2006 (UTC)

Ha ha, Pilotguy. You say my comment is "well put" but then you go on to disagree and say RfA can and should be fixed. No offense taken or, I hope, given; I can see the other side, too. But I'm afraid that any fix will, by the law of unintended consequences, do more harm than good.

I'm wary even of floating ideas (WP:BEANS) but I can think of a few band-aid fixes that might allow the current situation to stagger along for a bit:

• Make RfA a straight vote with explicit suffrage requirements. I like 1 month. Forbid discussion.

• Make it a secret ballot, so nobody can earn brown-nose points for supporting candidates.

• Create Wikipedia:Requests for Deadminship. 'Nuff said.

• Deadmin (and de-b'crat) everyone. Let Jimbo and the Board choose 3 b'crats to their own taste. For a limited period of time -- say, 1 month -- allow these new b'crats to elevate anybody to adminship they personally feel worthy. This new class of admins is limited to, say, 100 members. When this class fills or the free time frame elapses, we simply return to the old system.

• Establish the rule that new admins are always temporary; at the end of the (3 month?) term they are automatically deadminned...
  • ... and cannot reapply for adminship for another year. OR
  • ... unless they pass a second confirmation evaluation.

• Abolish elevation-by-merit or by discussion. Admin all editors who maintain some minimum level of contribs for 1 year without getting blocked for more than a day or two or being banned by ArbCom.

• Dissolve RfA as a page and as a process. Create a subpage for every user, e.g, User:John Reid/Review. Encourage all editors to comment on this editor's performance there. When an editor feels this page looks favorable "enough", let him apply directly to any b'crat for elevation. Same general guidelines for elevation but there is no open or close time. Editors who whine and repeatedly reapply pointlessly are hung by their thumbs from Main Gate.

• Create a secret guideline that prevents anyone from becoming an admin who has never been blocked.

• Create a new class of admin-candidates. These do not get any special new powers but are required to serve under the mentorship of an existing admin for 3 months, during which they do dog work. The mentor gets a veto at RfA.

• Require admins to prove legal majority. Seriously.

If I get any more wacky ideas, I'll add them later. But I really don't think any of these will bring lasting improvement. John Reid 01:03, 13 September 2006 (UTC)

If RfA is suspended, it will be viewed by some elements in the community as the existing corps of Admins trying to close ranks and prevent anyone new from joining them. I'm not saying that I believe that this is the intention of the suggestion to suspend RfA - just that it will be perceived this way in some quarters. That will lead to immense pressure for a new system to be devised and implemented extremely quickly - and thus may well lead to a poorly thought-out new system which may well have more problems than the exitising one. If someone has a clear proposal on how to replace the existing RfA system, they should put it up for discussion like any new process proposal. And the existing system should stay in place until a clear decision on how to replace it is reached. In the meantime, discussing the vague idea of replacing the system without a proper proposed alternative is getting us nowhere. Utimately any new system is probably only going to draw the active involevment of a small percentage of users. If those are the only users who take an active interest in the role of adminship and evaluating the people who wish to hold that position, I don't see that as a problem. At least the people who are involved take it seriously. Zaxem 00:22, 13 September 2006 (UTC)