Requests for comment/Proposal for a policy on involved administrators

From Meta, a Wikimedia project coordination wiki

The following request for comments is closed. Closed as successful policy addition to Meta:Administrators (21 Supports vs. 4 opposes) Theo10011 (talk) 21:09, 9 June 2013 (UTC)[reply]


This language is based on the policy at the English Wikipedia, which can be viewed here.

In general, users should not act as administrators in cases where they have been involved. This is because involved administrators can have, or can be seen as having, a conflicted interest in disputes in which they have been a party or to have strong feelings about whether the disputes originated on Meta or on another WMF project. Involvement is generally construed broadly by the community, to include current or past conflicts with an editor (or editors), and disputes on topics, regardless of the nature, age, location, or outcome of the dispute.

One important caveat is that an administrator who has interacted with an editor or a topic area

  • in purely an administrative role, or
  • whose prior involvement was either minor or obvious edits which do not speak to bias

is not involved and is not prevented from acting in an administrative capacity in relation to that editor or topic area. This is because one of the roles of administrators is precisely to deal with such matters, at length if necessary. Warnings, calm and reasonable discussion and explanation of those warnings, advice about community norms, and suggestions on possible wordings and approaches, do not make an administrator 'involved'.

In cases which are straightforward abuse or vandalism the community has historically endorsed the obvious action of any administrator – even if involved – on the basis that any reasonable administrator would have probably come to the same conclusion. Although there are exceptions to the prohibition on involved editors taking administrative action, it is still best practice in cases where an administrator may be seen to be involved to pass the matter to another administrator via the relevant noticeboard.

Some slight modifications have already been made to the language so that it works for Meta, but more tweaks may be needed.


Support[edit]

Oppose[edit]

  • Oppose Oppose Such a policy exists on en:Wikipedia, but history shows that it only creates more problems. It is 'randomly' ignored, or it leads to a discussion about the meaning of 'involved', or the uninvolved administrator will later prove to have been involved after all or was an involved admin's buddy, or a truly uninvolved administrator won't understand the dispute and draw the wrong conclusion. A better solution to the (perceived) problem is to have another user group (not administrators) make such decisions. Guido den Broeder (talk) 22:51, 20 February 2012 (UTC)[reply]
  • Oppose Oppose Too complicated language and unnecesary bureaucracy for me. It is already common practice here that admins should avoid using their buttons when they may fall into a flagrant COI. Please remember that Meta-Wiki is a small project afterall; and while we have a somewhat large number of admins (83) not all of them are dayly active nor works at all admin-workable areas at meta. That is: one works updating portals and does nothing else, another works with abuse filters and does nothing else, 5/6 watches recent changes, etc. They're focused in their area and do nothing else. It would be very difficult not to be "involved" here if this proposal passes (specially if the concept of involvement is construed broadly) because, as I said, some areas are simply handled by a couple of admins such as the spam/title blacklists or the RFH board; and those people knows what they are doing. As per my experience elsewhere I can see this being extremelly wikilawyered to death with bogus claims of "involvement" that will for sure generate more drama than anything specially in this small wiki; and would prevent taking the right decisions when needed. I agree that we, sysops, should avoid the use of the tools when a blatant conflict of interest may involve us (and I myself do it, even avoiding the deletion of things [outside my namespace] where I am the sole contributor); my philosophy is to look if the action is correct, not at the person who did it (blatant COI cases excluded). I'm convinced this proposal will trully harm this site. Better if we develop our own policies based on our own needs, than to start importing policies from elsewhere. —Marco Aurelio (Nihil Prius Fide) 15:02, 5 March 2012 (UTC)[reply]
  • Oppose Oppose while I hold very strict opinions on administrative actions taken while having a conflict of interest or the appearance of a conflict of interest, as those of you who follow Meta more closely will be aware, I do not think it a wise or good idea to import wholesale a concept and a standard from the English Wikipedia in this way. If there's anything that my experience as steward has taught me about the multitude of projects that fall under the umbrella of Wikimedia is that they have fostered very different concept, and what one thinks normal is thought bizarre and unheard of in others. I think it important that Meta have the least amount of policies it can, and that they be limited to the strict essential, to be the most easily accessible to users coming from all sorts of projects and project cultures. It is also noteworthy that in the past four years since Meta removed its yearly confirmations for administrator, nobody's filed a request for deadminship until this week. One last point that I feel compelled to make and that I feel is very significant and should be taken into account is that every single person in the support section above is almost everybody is from the English Wikipedia originally. I am not sure how other communities feel about this, and think it is important that we make sure our policies here on Meta are deemed reasonable and normal from the perspective of both people coming off from the English Wikipedia and other, diverse projects. Snowolf How can I help? 13:25, 8 March 2013 (UTC)[reply]
Well my friend, just after posting this a year ago I asked for help getting this translated so that users who did not speak English could more easily particpate in this discussion. that request was met by nothing by condescension by two of this sites admins, including the one whose bad faith close of this discussion led to it becoming active again. If anything that speaks volumes to why this is in fact needed here. If people in admin positions are so blind to ethics and so quick to dismiss an idea not because it is wrong but because of who proposed it, then ethical standards need to be codified to prevent abuse. Beeblebrox (talk) 21:48, 8 March 2013 (UTC)[reply]

Procedural/motions[edit]

  • Motion to nullify
    1. Policy page was created in anger as a procedural tactic by an editor with contempt for meta: [3] [4] [5]. Note also this response to proposer's anger: [6]
    2. This page was created amid 11-23 February en/meta cross-wiki conflict begun by a post at WP:ANI making an unsubstantiated allegation of defamation thereby tainting future events.
    3. This page improperly seeks to impose the culture of en.wikipedia on meta — and again note it was created amid the extraordinary circumstances of 11-23 Feb 2012.
    • Bottom line: There were many improprieties during the period of 11-23 Feb 2012, and creating this policy as a tactical move during those extraordinary circumstances is one more. The, what could be termed a call to arms at WP:ANI (later moved to WP:AN), was intended to teach meta a lesson. Let this "policy" page not be affirmation of such attempts at teaching lessons. (Posting this preliminary version of this motion immediately since a call to close has been made. Excuse any rhetorical excesses made without sufficient diffs for level of assertion — Comment: The events of 11-23 February 2012 should be reviewed more carefully before any permanent policy is affirmed.)
      -- Proofreader77 (talk) 18:33, 24 March 2012 (UTC)[reply]
    Oppose Oppose nullification. It doesn't strike me as at all improper that, when drama is caused by a gap in policy, the response would be to attempt to fill that gap using standard community methodology (propose change, wait for community to discuss change...). That's not "in anger"; that's "common sense." Since this proposal was opened, it appears to have been carried on quite calmly and constructively, by parties on both sides of the issue. The proposal is not to "impose en culture" on meta; it is to take a bit of policy text that works on en to prevent issues like what happened in February, and see if the community wants to try that as a solution here - notice the part where the community here discusses it and reaches consensus. That's not "imposing en rules", that's "seeing if we want to borrow something here that works elsewhere for this very problem which we just ran headlong into". Fluffernutter (talk) 18:44, 24 March 2012 (UTC)[reply]
  • Support Support de clasificar como nula y sin efecto la propuesta por completo por múltiples razones, entre ellas las aducidas por el usuario Proofreader77. Sólo hay que ver el encabezamiento de la misma para notar el sesgo con el que fue propuesta. El siguiente documento ha sido creado por un absceso de ignorancia tanto procedimental como del funcionamiento de este proyecto. El editor proponente del siguiente documento parece haberse retirado del proyecto no sin antes dedicarnos tan cariñosas palabras como "iros a tomar por el culo" (diff=3475067 [resumen]) por las cuales debería haber sido bloqueado. Las políticas han de ser propuestas y apoyadas por la comunidad en base a las activas necesidades o carencias del proyecto y eso no ocurre aquí. El presente documento intenta efectivamente imponer reglas de otro proyecto, propuesto por usuarios de aquel proyecto. Meta tiene su propia comunidad y a ella, y no a otra, le corresponde el establecimiento de las políticas que regulen el funcionamiento de este sitio. La falta de participación activa en este proceso en comparación con otras propuestas no puede ser ignorada. En segundo lugar y para los que no somos hablantes nativos de inglés, considero que el lenguaje utilizado en la propuesta es de difícil comprensión, enmarañado y tendente a controversia. Las políticas han de ser claras y concisas, no abiertas a cualquier tipo de interpretación interesada, lo cual sí, en efecto deriva en guerras y episodios dramáticos entre administradores y usuarios varios. Por tanto, apoyo su anulación. —Marco Aurelio (Nihil Prius Fide) 18:58, 24 March 2012 (UTC)[reply]
    OK, Marco, he escrito lo que creo que es una versión más claro por debajo. Por favor revise la misma. Gracias. -- Avi (talk) 19:53, 13 March 2013 (UTC)[reply]
    Oppose Oppose per Flufernutter. Alanscottwalker (talk) 21:29, 24 March 2012 (UTC)[reply]
  • Oppose Oppose major "sour grapes" proposal. --Rschen7754 05:48, 4 March 2013 (UTC)[reply]
  • Oppose Oppose motion – Per Fluffernutter. --Michaeldsuarez (talk) 12:43, 4 March 2013 (UTC)[reply]
  • Oppose Oppose per Fluffernutter. --SarekOfVulcan (talk) 05:39, 8 March 2013 (UTC)[reply]
  • Oppose Oppose Common sense is valuable regardless of where it is from. Furthermore, I believe the suggestions I added make this less restrictive (and more common-sensical) than the corresponding EnWiki version. The idea, in my opinion, is sound and should continue to be discussed to see of a consensus of metapedians agree. -- Avi (talk) 19:39, 13 March 2013 (UTC)[reply]

Neutral[edit]

  • Neutral at the present time. [See Changed to support, above] Excellent idea. I take the comments below seriously though. Process question: If wording is agreed on here, what are the next steps? Alanscottwalker (talk) 14:00, 20 February 2012 (UTC)[reply]
    • I'm not 100% certain, but this will likely be open for at least 30 days, and, if accepted, we would have to iron out the final language, and then it would get moved into Meta space as a Meta policy, I reckon. Local projects may choose to link to it as they do some other Meta policies (as Meta is often used as a framework from small projects). -- Avi (talk) 17:45, 21 February 2012 (UTC)[reply]
Thanks. Alanscottwalker (talk) 21:20, 21 February 2012 (UTC)[reply]

Discussion[edit]

Can I suggest that the approach to this proposal should be about the universal approach by administrators at whichever wiki where they have an involvement, and not solely focusing on meta. This should be a clear document on the principle and the community expectation. It is then available to be adopted by respective communities, or a model for their adaptation. I feel that Meta is an appropriate place for the development of the model and the ability to adopt it as a leading practice. Meta can then also have the discussion about whether they wish to operate above the set standard, or to that standard by local rules. billinghurst sDrewth 01:45, 20 February 2012 (UTC)[reply]

I'm certainly open to the idea that this should be a guiding principle for all administrators at WMF sites. I can tell you from direct experience that it helps prevent problems if admins know they will be held to such a standard. In this way the idea of consensus based decision making is also upheld as at least one other admin participates in the decision, although if a noticeboard is involved many users and admins may participate in the discussion. And of course even with this general principle each project can still draw its own line for what is considered "involved" as it may be near-impossible to avoid on the smaller ones. Beeblebrox (talk) 03:15, 20 February 2012 (UTC)[reply]
Billinghurst, It's not clear from your comments whether you are asking for the process or forum of this discussion to change. If the process or forum needs to change can you be clear on exactly how, or perhaps Avi can answer? Alanscottwalker (talk) 14:11, 20 February 2012 (UTC)[reply]

I concur with billinghurst. However, I would prefer the policy to be a little more lax, and allow involved administrators to make closes where the consensus is obvious. Especially in smaller wikis, where the administrators are measured in the single digits, it would make it much easier, if they were to be able to to close a discussion where the outcome was obvious, even if they opined. Which, for the record, is probably the case on most wikis (EnWiki is more the exception than the rule, but with over 1500 admins, there are enough to go around). Also, I would not word this as a "prohibition" at the start, that is up to each and every project to decide whether to make it a violation or grounds for desysoping. I would word it as a "suggestion" or "request". Then, each project can decide on the strength to give this advice and refusal to listen. As such, may I suggest something like this:

In general, users should not act as administrators in cases in which they have been involved, unless the activity is clearly in line with the project's policies and guidelines, and the participant's consensus if applicable. As involved administrators may have, or may be seen as having, a conflict of interest in disputes they have been a party to, or have strong feelings about, it prevents the appearance of impropriety for them to recuse from acting as administrators in such discussions or processes. Involvement is generally construed broadly by the community to include current or past conflicts with an editor, or editors, and disputes on topics, regardless of the nature, age, location, or outcome of the dispute.

An important caveat is that an administrator who has interacted with an editor or topic area in a purely administrative role, or whose prior involvement are minor or obvious edits which do not speak to bias, is not involved and is not requested to refrain from acting in an administrative capacity in relation to that editor or topic area. This is because one of the roles of administrators is precisely to deal with such matters, at length if necessary. Warnings, calm and reasonable discussion, explanation of warnings, advice about community norms, and suggestions on possible wordings and approaches, do not make an administrator 'involved'.

In cases which are straightforward abuse or vandalism, or cases where the consensus of the discussion is beyond dispute, the obvious action of any administrator – even if involved – is usually accepted on the basis that any uninvolved and reasonable administrator would have likely come to the same conclusion. Regardless, it is often best practice in cases where an administrator may be seen to be involved to pass the matter to another administrator via the relevant noticeboard.

The other piece that may be of use is whether or not to explicitly allow reversal of an "involved administrators" action. Personally, I would leave that up to each and every project, with those too small to decide on their own subject to Meta (as per usual). -- Avi (talk) 03:25, 20 February 2012 (UTC)[reply]

Proposed Amendments to Avi proposal (this space is for a short statement of amendment or improvement)[edit]

  • 1. insert "written" before "policies" in the first sentence.
  • Simplify the language and make the concept clear (for those who support the concept as me, but find the language too complicated). -- MarcoAurelio (talk) 13:24, 4 March 2013 (UTC)[reply]
    • I support this very much. The idea is logical and (I believe) already followed here on Meta, e.g. on WM:RFD, WM:PPM and WM:RFA requests are normally closed by someone not involved before. The current proposed text however is not acceptable to me, because it is written in a very complicated style - not very understandable to me as a non-native speaker of English. [I'll try to add a few more exact formulation concerns later] --MF-W 14:25, 4 March 2013 (UTC)[reply]

Discussion continued[edit]

I expect that it would not be too difficult to modify the wording to allow for substantial "wiggle room" for the smaller projects. Generally, I would think that if an admin seemed involved and their actions were subject to scrutiny after the fact, they should be reviewed on the same basis as any other admin action. If found to be justified despite the involvement, that can be noted wherever appropriate and/or the action can be undone and then rapidly re-instated by another admin who would then be "holding" it. In fact, recent experiences suggest this is already practiced at Meta. Beeblebrox (talk) 03:32, 20 February 2012 (UTC)[reply]
"Small" wikis can be very challenging for admins - that is my original background and when you are the only person around you have to deal with what you have to deal with. For what it is worth I have always considered myself a servant of the community and would happily listen/respond to any questions they have over my actions. I would expect this from any other rights holder - sadly experience has shown that that not everyone is responsive. However too much policy will tie down unreasonably smaller project admins. Bear in mind too - Meta is a smaller project which has global affects. This is a reason why I tend to be rather fussy about RfAs here - the impacts can be wide ranging and folk who things that because the buttons are the same the job is too do not understand Meta. --Herby talk thyme 11:47, 20 February 2012 (UTC)[reply]

One word, [written], bracketed amendment to Avi's proposal. See above. If you would prefer suggestions/amendments be made another way let me know. Alanscottwalker (talk) 13:14, 20 February 2012 (UTC)[reply]

Honestly it seems to be a bit redundant, avoiding COI (without being paranoid) is a common sense practice. --Vituzzu (talk) 15:10, 20 February 2012 (UTC)[reply]
I'm not sure if your comment is directed at mine. Are you saying "written" is redundant? Redundancy does have its use. Alanscottwalker (talk) 15:22, 20 February 2012 (UTC)[reply]
I think that adding "written" adds a layer of bureaucracy unnecessary in all but the largest projects, all of whom have local policies that would supersede this. As Vito says, it's common sense to avoid this kind of impropriety, and if we limit it to "written", we would either need more "rules creep" by formulating an explicit policy, or we open the door to rules lawyers. For most projects, the more we can rely on good common sense the better. It is when a project reaches critical mass, and many groups of editors with conflicting motivations arise and begin to engage in long-term edit warring as we see on EnWiki, then clear and explicit rules are needed to ensure the continued operation of the project. I'd say that 270+ of the 300+ projects (and I'm likely underestimating the number of projects) do not have the issues that EnWiki has.
As for how to add amendments; I think it better not to change the text of someone else's edits, as it makes it unclear who wrote what. So, if you would, I'd appreciate your describing your suggestion at the point of your edit, but allow the text I wrote to remain as I wrote it. What may be helpful is to set ub a subsection directly under the initial suggestion and prior to the comments which can be used as a "sandbox" for adjusting the wording; one not tied to any particular editor. Thanks. -- Avi (talk) 15:35, 20 February 2012 (UTC)[reply]
I'm sorry, I don't understand. We are creating a best practices model, correct? So, a lot of projects won't adopt it in this form anyway, correct? If we are creating a best practices model, are you saying it's not best practice to write the rules that everyone is expected to follow? I will correct the amendment. Also, can you respond to my process questions above? Thank you Alanscottwalker (talk) 16:12, 20 February 2012 (UTC)[reply]
I am a sysop on a project with two active sysops. If I have COI, the other sysop will most likely have it too. No, we will not import this policy. Our sysop-policy-document is 138 words, and that is all we need now. Less is more! -- Lavallen (talk) 19:20, 24 March 2012 (UTC)[reply]
Is it possible that it would be helpful to create a separate "best practice" model for admin policy on small wikis? This could help clarify issues that small wikis face, and especially how those issues are different from the big ones. For instance, a small-wiki admin policy might say that responsiveness to questions and comments is the main thing (because the "involved" concept can't be practically applied). Rd232 (talk) 20:51, 24 March 2012 (UTC)[reply]
"responsiveness to questions and comments is the main thing", response and comments about what? - On "my wiki" we have blocked 8 times for graffitti, once an open proxy, and once a bot running wrong, and that is the total amount of blocks for 8 years. We actually do not have enough "practice" to say any words other than the common sense tells us. -- Lavallen (talk) 19:30, 26 March 2012 (UTC)[reply]
The collective experience of small wikis is obviously greater than that, and sharing it and elaborating conclusions may be helpful. Even if the conclusion supports the status quo in terms of practice, that's helpful... Rd232 (talk) 22:23, 25 May 2012 (UTC)[reply]

I applied grammatical principles to above statement may to can; if we are having then it is a conflicted interest, etc. The third paragraph has issues in that the first sentence starts with a conditional component, but the thought is not completed, so the second sentence hangs in isolation. What was the intent of the condition? — billinghurst sDrewth 14:56, 4 March 2013 (UTC)[reply]

Rewritten for clarity[edit]

Marco raised the point above that both the suggestion, and perhaps my emendations, are unclear. As meta is populated by many for whom English is not their native language, I will attempt to rewrite the proposal in clearer language. I would hope that people support or oppose here again so that we reduce confusion.

To prevent the appearance of impropriety, editors should not act as administrators in situations in which they have been involved, unless the activity is clearly in line with the project's policies and guidelines, and the participants' consensus if applicable. Involvement is generally construed by the community to include current or past conflicts with an editor, or editors, and disputes on topics, regardless of the nature, age, location, or outcome of the dispute.

An administrator who has interacted with an editor or topic area in a purely administrative role, or whose prior involvement are minor or obvious edits which do not indicate bias, is not considered involved and is not requested to refrain from acting as an administrator with that editor or topic area, as one of the roles of administrators is to specifically deal with such matters. Warnings, calm and reasonable discussion, explanation of warnings, advice about community norms, and suggestions on possible wordings and approaches do not make an administrator "involved".

For straightforward abuse or vandalism, or cases where the consensus of the discussion is beyond dispute, the obvious action of any administrator – even if involved – is usually accepted on the basis that an uninvolved and reasonable administrator would have come to the same conclusion. Even so, it may be better to pass the matter to another administrator in cases where the actions may appear inappropriate.

Please provide your comments below. -- Avi (talk) 19:52, 13 March 2013 (UTC)[reply]

Second rewrite[edit]

Administrators should avoid using their tools in situations where they may not be impartial. Those situations include using the tools in disputes in which they are involved parties. If the administrator is in doubt, the administrator is encouraged to ask for opinions, or to ask another administrator to take the case at WM:RFH.

I feel this second rewrite may not be as perfect as Avi's above, but I think it covers the basic common sense rule that one should not be the prosecutor, the judge and the executioner at the same time. -- MarcoAurelio (talk) 15:08, 14 March 2013 (UTC)[reply]

Perhaps, but my experience on EnWiki makes me leery of not explicitly mentioning common sense rules of when even an "involved" editor may act as an admin. Thankfully we are much less drama prone here, but I fear that if something is "committed" as policy, someone is going to try to ruleslawyer it, and just because someone expressed an opinion in a deletion discussion, if it is unanimous, (or 27 to 2) they should not be called on the carpet for closing it in accordance with consensus. We tend to be more common-sensical here, though, so perhaps my paranoia is not needed and your rewrite is certainly clearer and more concise 8-) -- Avi (talk) 02:05, 15 March 2013 (UTC)[reply]
I like Marco's paragraph. There can still be a footnote about exceptions, as in Avi's second draft. Both look good to me. (I made some minor rewording for clarity.)SJ talk  22:52, 25 March 2013 (UTC)[reply]

Note - This RfC has been open for over an year. I am starting to close all these old discussions. This one didn't have a stipulated end date, and there are recent comments above. But I can also see they have no bearing on the outcome and suggest a second rewrite. If anyone wishes they can ask me or another admin to re-open the discussion. Regards. Theo10011 (talk) 20:58, 9 June 2013 (UTC)[reply]