Wikinews:Water cooler/policy/RfDa discussion

"Requests for de-adminship" policy

Thanks to Amgine for bringing this matter to my attention on my talk page. I was one of 4 people who voted for a Rfda (de-admin)before it was cancelled[1]. I decided to research the matter because Amgine had said on my talk page"This constitutes a disruption of the site, as you are aware. The RfdA was removed because there was no basis in policy" and Amgine was also kind enough to direct editors to the policy he was referring to with his cancellation edit[2]"This RfdA does not have a basis in policy. Please consider bringing an RfdA which meets the policy guidelines on this page." meaning the Wikinews:Administrators[3] page and the "policy" on the page seems to be;

"Further, evidence that an admin has abused his privileges must be provided. Any listing that cannot produce sufficient evidence will be removed from this page; further disruption may result in blocks."

The stipulation that removal hinges on admins abusing their "privileges" (blocking and page saving powers) is incredibly narrow and basically means an admin could be aggressively vandalizing the site in the way our biggest vandals do and we still could not put him up for de-admin. This narrow stipulation is nothing short of patent nonsense; and to couple it with the threat of a "disruption" block is really over the top,imo.

I spent 4 hours researching the Wikinews:Administrators page and found this "policy" was actually established through 2 edits by 1 inactive admin.,Dan100, on Oct.6th.- during the heat of a previous de-admin. battle- with no community consensus or discussion at all.

[4][5]

The previous policy on the Administrators page[6] had said;

"Requesting de-adminship: Stewards are the only users who can remove administrator privileges. They will not de-admin unless there is community consensus for this to happen. Anyone may request de-adminship and everyone may object to a request for de-adminship, and if their argument is reasonable, stewards will seriously take this into account."

The point is; our policies and guidelines page[7] says specifically; "Most Wikinews policies are developed through consensus". Therefore, I submit that there has not been a consensus to adopt the edits of Dan100 on October 6th. and I dare say that most of us were not even aware those edits had been made.

Since the policy page has the purpose of addressing and developing policy, I will begin the discussion by saying I support the previous policy (prior to Dan100's Oct.6 edits) as being a basis for our policy and will be bold with a slight alteration; Neutralizer 16:21, 13 November 2005 (UTC)[reply]

Policy is fine the way it is. --MrMiscellanious (talk) (contribs) 02:50, 14 November 2005 (UTC)[reply]

For discussion; New De-sysop Policy

"Anyone may request de-adminship and everyone may object to a request for de-adminship, and if the majority of the voters after 48 hours express the view that the Rfda is frivolous; then the Rfda will be cancelled."

Neutralizer 15:53, 13 November 2005 (UTC)[reply]

So in essence, this policy revision would mean there is a vote, to see whether or not there should be a vote, for de-adminship. I'm afraid this may have reached that point. -Edbrown05 18:18, 13 November 2005 (UTC)[reply]

Excellent! A policy discussion on the policy section!

I would like to raise the question of whether the deadminship should be an unpopularity contest, or a way to address wrong-doing? Since we are in the process of creating an arbitration committee, the question of addressing abuse of admin privileges should really be given to the arbcom.

But that doesn't address when an admin loses the trust of the community. If we allow deadminship requests as unpopularity contests, anyone can disrupt the community repeatedly because they do not happen to like someone. It can (and is) used to harrass someone. So there should be some form of justification requirement. For a de-adminship, as has been clear on Wikipedia, that should be abuse of adminship privileges because other forms of misbehaviour are already covered by policies.

Not all cases of loss of trust will have justification to request a de-adminship; this should not mean an admin is not protected from harrassment. But there clearly should be a way for the community to remove privileges from someone they do not wish to have them, but who has not actually done anything wrong. One way which is used on Meta is to require a regular community review of admins, where the community automatically re-votes to support or not support an admin for another period of time.

Under the current suggested policy, above, if it were applied to any other editor (to remove their ability to edit articles, for example), there are contributors who might be brought to vote on a weekly basis. This would obviously not be fair.

Instead, I suggest two policy items instead of one:

* Abuse of administrative privileges is grounds for deadminship by the Arbitration Committee, and any user may bring evidence of such abuse to the ArbCom.

* Administrators work for the community, and their adminship should be reviewed regularly, as a community review of their performance and continued acceptance by the community. Admins will be up for review in the first week of the next quarter beginning 6 months after their election, and every other quarter there after so long as they remain an administrator. (Any administrator who been elected more than 6 months prior to the community acceptance of this policy will be reviewed in the first week of the next quarter, and every other quarter there after so long as they remain an administrator.) - Amgine 22:47, 13 November 2005 (UTC)

Clarification please; would this mean that, for example, since MrMiscellanious was sysoped Aug.2nd., the first time he would come up for review would be in April,2006 ? Neutralizer 00:33, 14 November 2005 (UTC)[reply]

If that's the case; I would suggest tweaking your proposal like this; change the time period from 6 months to 3 months (which can be a long time with a dynamic project like this)....and just to get things rolling; why not start off by putting all existing admins up for review right now who have been sysops for over 3 months? Then the next set of reviews would be the first week in January etc.? Neutralizer 00:42, 14 November 2005 (UTC)[reply]

I agree that a 3 month time frame is more suitable, however given the recent events on Wikinews I think if this policy goes through the first "review" should take place in the new year. Just to give time for any hard feelings to be smoothed over. --Wolfrider 02:33, 14 November 2005 (UTC)[reply]

<nod> I can see your points, but consider a couple things:

This might be interpreted that Neutralizer does not agree with the community vote taken just before he arrived.
Every 3 months every admin on the site will be up for community approval. What if none gain 70% approval? What if they choose not to stand? (Some, like myself, will likely consider it both onerous and insulting.)
Should this apply to bureaucrats? Should this apply to all editors? Should this apply to stewards?

I will be frank: this seems like policy-making to target a single person. That's a very bad way to make policy. If you have a problem with a single user, push the creation of the arbitration committee and bring your case to them; don't risk the security of the site or the good will of all admins by subjecting them all to quarterly reviews. - Amgine 03:11, 14 November 2005 (UTC)

I understand that this looks like it is being created to target a single user, but it was something I myself was considering proposing a while ago. I really don't see an issue with no admins gaining approval (does it have to be 70%?) but I do believe that admins should be under constant review by the community. A vote of confidence never hurt anyone. ;-) I really can't comment on the bureaucrat situation, as I do not know how they are chosen. I don't think a policy like this should apply to stewards, as they tend to stay out of affairs do they not? --Wolfrider 03:38, 14 November 2005 (UTC)[reply]

Oppose. No need for policy change, user is acting on personal feelings toward another rather than making an unbiased proposal that would help Wikinews and its community. --MrMiscellanious (talk) (contribs) 02:57, 14 November 2005 (UTC)[reply]

Actually I see a change like this assisting the community. Regardless of whether the proposal is made out of a reaction to a particular event, it's a good idea and warrants consideration. Anyway, an administrator that has gained the trust of the community need not worry, I see this as a policy intended to weed out administrators who (while not overtly breaking any rules) are deemed by the community to be damaging to the project. --Wolfrider 03:41, 14 November 2005 (UTC)[reply]

Glad someone started this discussion and great to see it remaining constructive! I would like to understand what happend when the rules were changed. I was not around at the time, and maybe there were good reasons to change the policy at the time (although that is no excuse for changing the policy without community concensus). This might shed some light on the question whether there actually is danger of deadmin requests degenerating into (un)popularity contests. From what I have seen this does not happen, most users act very responsibly. I would hope that one could avoid a regular review process as outlined by Amgine as this may be a compromise that makes nobody happy. Some people will feel that the waiting time between reviews is too long and other feel annoyed that they have to constantly vote on admin reviews instead of writing articles. I would like to see some very general guidelines of when one may start a deadmin request. Violating rules is a given. With the powers of an administrator, there come certain responsibilities. Most importantly, I expect an administrator to act as a role model for the other users, someone that could serve as an example especially to new users. (When i started to write here about a month or two ago, that was my expectation!) How about:

"Any user may request de-adminship if the user feels that the administrator in question has repeatedly acted in ways that set a poor example for other contributers, despite good efforts of contributors to resolve these problems."

This is admittingly vague and does not exclude 'frivolous' deadmin request, but it provides some scaffolding for a deadmin argument and it points people to try and make an honest effort to resolve problems before listing a deadmin request. --vonbergm 03:22, 14 November 2005 (UTC)[reply]

I don’t see a need for an Arbitration Committee, which to me is just another bureaucratic mess to deal with, and I don’t see a need to change anything at Wikinews except for letting go forward a de-admin vote that was proposed.

In two instances, the de-admin vote was squashed for reasons bordering on esoteric that I don’t care to puzzle over, and only served to keep an open sore… what… un-medicated?

I think it’s fairly obvious that any admin should be allowed to be called for de-admin vote just as easily as someone labeled a vandal is blocked. It’s all a matter of opinion so what’s the big deal about this? -Edbrown05 03:46, 14 November 2005 (UTC)[reply]

I see your points, but constant requests for de-adminship (is that a word?) could really bog down the community and lead to harassment of admins. Not exactly fair. Also, I agree there really should be a reason for relieving an admin of his/her duties, but it might be best to have a secondary system in place to keep admins in check. It's quite possible for administrators to be abusive and harmful to Wikinews without breaking policy. I think this proposal combined with our current system addresses both concerns. --Wolfrider 03:55, 14 November 2005 (UTC)[reply]

When I first saw the "admin re-up vote" idea, I liked it. Then, like all ideas it sort of soured like an old apple, because I'm afraid of low voter turn-out! But it is the sort of secondary system that could work. -Edbrown05 04:17, 14 November 2005 (UTC)[reply]

Trust: In the discussion vote on de-sysoping Simeon, while I voted the other way from MM, I found one of his comments very useful. He was saying that he felt he had lost "trust" in Simeon as an admin, and that because trust is a key qualification for that role, he saw this as a reason (he also had other concerns as well) to support the de-admin vote. So although, as I say, I took a different view on that particular discussion (I was happy express trust in Simeon), I think MM had hit on a very good "rule of thumb". What do people think about including some kind of "trust" criterion in a de-sysop vote - ie: if a majority (be that absolute, 70% or whatever) feel unable to express "trust" or "confidence" in a particular admin, then that can be the grounds for de-sysop? It would effectively be a vote of "confidence"/"no confidence".

Good faith:Maybe we could tie this in with the principle of "assuming good faith" - ie. we should always assume the best of each other but if the behaviour of a particular admin makes it impossible to maintain our assumption of good faith, then that's where the trust/confidence breaks down, and de-sysoping becomes an option. "Good faith", in my view should mean at the very least being courteous and non-confrontational, . Of course the occasional lapse isn't the end of the world, but it's when the problems become consistent that there's cause for concern.

Dan100's imaginary rule: Neutralizer's to be congratulated for his detective work in tracking the origin of the "rule" which permits the blocking of people deemed to have brought "frivolous" de-sysop requests. It's my understanding that rules have to be discussed openly and agreed by community consensus. One unilateral edit dreamed up on a whim, without any community consultation, by Mr. Dan100 does not a rule make. So unless I'm missing something here, isn't Dan100's imaginary "rule" therefore null and void? At the very most, it's a proposed rule that has not yet been discussed or approved by the community.

Quorum as a solution to the "frivolity" problem: I think there are far better ways of deterring "frivolous" de-sysop requests than threatening to block people. How about a "quorum" system - ie. no de-sysop request can proceed unless four (to pluck a number out of the air) or more users have voted in favour? So basically the de-sysop request wouldn't even need to be taken seriously until a significant number of wikinewsies had expressed support for it. This automatically screens out any cases where it's just one or two aggrieved wikinewsies trying to cause trouble. Lone "axe-grinders" can initiate as many de-sysop actions as they like, but they'll just wear themselves out and it won't be a disruption to the community because their efforts will just be ignored unless they can convince others there's a serious issue in play. If, on the other hand, there are four (or whatever number we decide should be the quorum) wikinewsies concerned enough to support de-sysop, then that's already by definition a more serious situation. And even then, the vote may still fail if a sufficient number of other wikinewsies vote against de-sysop. I'm not making a firm proposal about exactly which number the quorum should be set at - but I do think that the PRINCIPLE of quorum could help us out here quite a lot. --

Rcameronw 22:09, 14 November 2005 (UTC)[reply]

Trust, good faith, quorum. I like it. --vonbergm 22:56, 14 November 2005 (UTC)[reply]

Comment I noticed that the last few nominations for admins have produced about 6 votes in the first 24 hours; so it might be too restrictive to require the quorum of certifiers to be 4 (since some of the votes will likely be against de-admin); unless the certification period could be extended to 36 or 48 hours. Neutralizer 03:59, 15 November 2005 (UTC)[reply]

Better example; Even after 7 days there was only 4 votes in total re MessedRocker's nomination for admin. [8]. We need a better way to alert the community when these votes are underway,imo, and that better way should be found right away if we are going to the quorum approach. Neutralizer 15:15, 15 November 2005 (UTC)[reply]

What about a staggered quorum? Just adiscussion idea. Rfda would stay alive;

24 hours with a listing by an identified user...

2 days with one additional "certifier"

3 days with 2 additonal certifiers and

goes to completion with 3 additional certifiers. Neutralizer 16:52, 15 November 2005 (UTC)[reply]

Rcamernow and others, Dan100 simply changed a page which can be edited by anyone -- most of the policy changes in this Wikis history were made similarly (without voting, which is evil). If someone makes a change and no one objects and the policy is used then it becomes accepted policy -- at the root of all that is consensus, real consensus. --Chiacomo (talk) 23:07, 14 November 2005 (UTC)[reply]

OK then, I object to Dan100's change! I'd suggest that it's been a (perhaps even uniquely) divisive, disruptive, and damaging rule that has led to endless upset within this community, when there would have been much better (ie. more efficient, effective and fair) ways of dealing with the problem it was designed to solve. A quorum system is just one example - I'd welcome other suggestions from the community. I don't think there ever was any real consensus for the "Dan100 clause", and I certainly don't think there is now. I think it's dividing us and I think we can do better. Down with the "clause"! Up with Quora! --Rcameronw 23:47, 14 November 2005 (UTC)[reply]

(PS - thanks for the clarification about how the wikirules work, Chiacomo - you are indeed wise in the ways of wiki)--Rcameronw 23:47, 14 November 2005 (UTC)[reply]

No problem, Rcameronw! I simply didn't want folks to think that Dan100 had done something nefarious. I am greatly pleased that this discussion is underway... Quorum isn't exactly the word I'd use, "certify" is better, I think. As in, "Chiacomo, initiated the RfDA and Neutralizer, Amgine, and Recameronw certified the request within the requisite 24 hour period; it can now go forward." What do you think? (I know we're arguing about language now.)... --Chiacomo (talk) 23:52, 14 November 2005 (UTC)[reply]

Sounds perfect to me. I have no firm attachment to the word "quorum" - except for the bizarre nostalgia of all the student meetings I attended at university that were then immediately cancelled because "quorum" hadn't been reached (think of a de-sysop vote as a kind of "student union meeting" and you'll see how I got there). I've no strong objections to "certify" although my only concern is that it implies, in my mind, some kind of specialist knowledge, which I'd never claim to have! But basically I'm really fine with whatever word we use, so long as the structure is a sound one - and I think the one you've outlined is very good. What I like about this system is that I feel it a) addresses the legitimate concerns admins have about the risk of repeatedly having to defend themselves against requests apparently based solely on a personal-grudge, while at the same time b) giving a fair and robust mechanism for wikinewsies to raise a request, without fear of "punishment" when a significant number of them honestly feel that there is a clear problem. Everyone's a winner!

Rcameronw 10:56, 15 November 2005 (UTC)[reply]

Do we need de-admin request limitations at all?

Just checked the history and As of Aug. 27th; we had gotten by since the birth of Wikinews with no limitations at all on bringing de-admin requests (Rfdas)[9].

Then came the first Rfda on MrMiscellanious and the rules began; so I am wondering;

A; Are any rules needed? After all, we have admins who are not shy about blocking anyone who is disrupting the site by registering spurious Rfdas or any other form of disruption. We also have rules allowing for blocking open proxies.

B; Why in the world would anyone want to keep the admin job if the majority of voters on a 7 day Rfda voted them out regardless of the merits of the Rfda?

C; Is the community really better off by retaining an admin who would want to stay after being voted out on a 7 day Rfda?

Therefore, I propose we just go back to Aug.27th Rfda limitations...which were zero. Neutralizer 05:18, 17 November 2005 (UTC)[reply]

Well, I wouldn't want to contradict, but apparently those guidelines came about when Dan100 (I think) talked to a steward about what would be required to de-admin. Like most policy, they look like the traditions which stewards have developed over time just written down. I don't remember, was that before or after the request to de-admin NGerda? (I was sailing at the time, or just back, but don't recall the events.) - Amgine 05:33, 17 November 2005 (UTC)

Wrong, Amgine; just plain wrong. Why not just have a look at the history instead of "apparantly"izing? Aug. 27th; no limitations, Aug.28th Rfda on MrM, Aug.28th, Dan100 unilaterally cancelled the Rfda (vote was about 3-3) and then (Dan100) proceeded unilaterally within the next hour to make the first limitation edits that he called "common sense" rules.

Amgine brings up a good point in that the NGerda Rfda, proposed by Dan100, was dealt with about a month earlier (July) and had a full hearing (no accusations at all of abusing admin privileges) which went very smoothly without any limitations being in place; showing limitations really aren't necessary at all. Neutralizer 06:10, 17 November 2005 (UTC)[reply]

rfda policy proposal

This is ideas from many contributors on talk:Administrators. Major point with this is to avoid "lonly axgrinder" rfda and appeal to solve problem with an administrator before rfda.

Here is a proposal to policy for rfda:

Requesting de-adminship: Stewards are the only users who can remove administrator privileges. They will not de-admin unless there is community consensus for this to happen.
Before a rfda can start 4 users must certify its need on WN:A within a timespann of 48 hours.
As a guideline a good faith effort should have been made to resolve the issues leading up the RfdA. Bring up complaints against an administrator on talk:Administrators or another suitable forum.

International 20:14, 5 December 2005 (UTC)[reply]

The only other point that should be added is something to discourage/prevent the same admin being nominated every 48 hours until one succeeds or the same admin being put up week after week with no consensus to deadmin but four users who do not like the person. I am not particular in favour of this version, as I believe requiring proof of violation of the admin’s powers is the only reason a person should be deadmined. However if some sort of restriction is placed on frequency of the RfDA I will not oppose this proposal. --Cspurrier 20:30, 5 December 2005 (UTC)[reply]

Here is a suplimental/counter proposal: Add a Case Building subsection for RfDA, called RfDA-CB, where people may collect arguments for some future RfDA. Advantages:

Allows people to officially vent without annoying everyone with stupid voting. Not nearly as confrontational as real RfDA. Reduces wikistress. etc.
Such venting is quasi-constructive as neutral parties may edit RfDA-CB evidence into more a concise form. Admins can find any legitimate or quasi-legitimate complaints, and try to curtail any minor abuses.
More accurately reflects the fact that people want to see a patern of abuse, not just a single abusing incedent, for an RfDA. It is true that any sufficently serious offence is grounds, but RfDA voters let the small one-off stuff slide, as they should.
RfDA-CB pages would actively discourage users from making an RfDAs, as frequent RfDAs discredit them.

RfDA-CB pages would essentially be a very specilized form of talk:Administrators, but one extremely obvious & seductive, as well as calming, to RfDA submitters. Anyway, I doubt any formal restrictions on RfDA are needed if you have RfDA-CB pages, but one could obviously adopt the suggestions of International & Cspurrier too. Thoughts? Nyarlathotep 15:41, 6 December 2005 (UTC)[reply]

Good idea, it is in the same spirit as user:vonbergm good faith idea. This make a good strukture for construktiv problemsolving. I agree that formal restrictions on RfDA is bad and I dont think anywiki yet (the last months) has missused RfDA. But certifikation from a few wikinewsis may prevent future missuse. Good argument will round upp enough certifyers and nobody will in advance dictate what a good argument is. International 17:18, 7 December 2005 (UTC)[reply]

Legitimate Rfda Cancellation/deletion

MrM deleted my Rfda on Amgine although the previous Rfda on Amgine has been withdrawn. I have a health situation that allows me online rarely; I trust the community will replace the Rfda I just put on; as I do not have the energy for an edit war with MrMiscellanious. The use of such obvious technical types of page control should not be tolerated,imo. Paulrevere2005 02:08, 14 December 2005 (UTC)[reply]

Leaving Rfda Votes up until consensus has been reached

It is my belief that when a admin is put up for Rfda and the vote has finished, the vote should be left up until a consensus has been reached that it should be removed. (ie. Amignes vote is the perfect example of a vote that should be removed because a consensus has been reached) this enables people to commment and work out their difference so another vote doesnt eventuate. Seeing it will obviously be a large group of people with a problem with the admin then getting the admin to work out the problems individually is Ridicules and will just make life harder for the said admin and everyone involved especially if wikinews gets "big".

the Rfda is alot diffrent to the vote for admin--Whywhywhy 00:50, 19 December 2005 (UTC)[reply]