Meta talk:Requests for adminship/Policy issues

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search

Admin eligibility[edit]

There are currently two approaches on the "table". One based around fairly tight criteria of edits and rights by Lar and one based around "suitability" by Pathoschild. Both seem to have merit. I have moved this discussion so far to here and left a link to this on the main talk page (here). --Herby talk thyme 12:34, 13 December 2007 (UTC)

This is something I have given some thought to over the past couple of months as I have come to know Meta better myself. I fully understand the concept that as a trusted admin elsewhere people may be considered for joining the "Meta admin club". There is a sense in which I felt that way when I joined. 100 edits really can be achieved in a very short time on RC and the like. Equally I guess the 100 edit criteria on the home wiki was something that seems to be taken from the air - I would imagine on most Wikis most admins had 100 edits.

Now some months on in working actively on Meta I have some serious doubts over the standards of eligibility. It does not seem to take into account what can be done (& how much is affected) by an admin here. Taking two simple cases there is the inter wiki mapping. Something which affects all wikis and requires some reasonable Foundation experience to satisfactorily evaluate in my opinion.

Equally there is the spam blacklist. This effectively controls content across all Foundation wikis and beyond. Allowing the fact that at least some small experience of regex is required, I'd prefer to think that those who do deal with this are experienced wikimedians. I certainly stayed firmly away from such pages when I started - however can we count on all new admins to be the same?

I would like to see Meta admins be genuinely experienced in the ways of Meta itself. There is plenty here to do with discussion of new projects & project closure for example which require no admin rights. Involvement in these and other Meta pages would allow far better consideration of admin suitability.

I would like to propose that the eligibility criteria are changed to 500 Meta edits and 500 home wiki edits to allow prospective admins more opportunity to get to know Meta. I look forward to the communities views. Thanks for the time & attention --Herby talk thyme 12:39, 11 December 2007 (UTC)

Generally agreed. I would point out the current criteria should have never taken spam blacklist into consideration (interwiki map, on the contrary, may have), since such a thing didn't exist at that time as far as I know. I don't think the criteria which fitted 2003 circumstance can go through 2007 situation without any modification - the project is growing and there are something new we want to consideration here and there. So raising the bar sounds me making a sense ... but I am not sure 1) the proposed count, 500, is a good measure and 2) edit count without any specification is a good measure. I would rather love to hear other opinions. --Aphaia 13:47, 11 December 2007 (UTC)

I also generally agree. While 500 edits on meta are quite a lot, I'd rather suggest to specify a minimum time frame, both on meta activities, both on local adminship, IMHO a certain familiarity with WMF projects requires at least six months and preferably that the activity is not limited to one single project.
This is my own idea:
  • Increase up to 500 edits the minimum requirements
  • Specify a minimum of n months demonstrable activity both on and in a local project
  • Suggest that the user is already involved on multiple projects
--M/ 14:27, 11 December 2007 (UTC)

A few comments: (That we ought to consider revising our criteria does seem goodness to me.)

  • There isn't a "defined process" for setting this sort of policy here. So we need to come to a reasonable consensus among a wide enough set of participants in this discussion that we all are willing to accept the outcome... when we have a set of things we seem to agree on maybe we do a straw poll or a consensus check by republicising and drawing more voices in.
  • I want to separate suffrage to comment from qualification to be an admin. I think there is merit in having both, see upwards, we were getting close to some sort of consensus (some time and edits elsewhere, some smaller number of edits here) on suffrage, would like to see that driven home too while we were at it (if we do a wider call, let's get them both, and maybe even tackle suffrage for other things like stewards, to maximise the utility of the call?).
  • I think 500 edits HERE to be an admin is a bit much. Remember that the original qualification was to be an admin elsewhere. I'm not sure there is a lot of stuff for people who are not admins to actually DO here that wouldn't be "make work"... or if there is (stuff like translations, helping write help texts etc) I'm not sure it is all that germane to adminship. I totally agree with Aphaia that the interwiki map (for example) is very important across many many WMF and non WMF projects and, further, we've had some edits to it that I don't support, there wasn't enough consensus and discussion first, in my view. But I don't see how 500 edits on help texts will help someone learn the interwiki map nuances. I'd support a lower number HERE and some wider experience elsewhere (adminship on one large or two small wikis, meaningful participation on at least 2 other wikis, some higher edit count at at least one wiki (1K??) , some significant time (3,4,6 months?), something like that, not sure what exactly) before I'd support a higher number here. ++Lar: t/c 16:27, 11 December 2007 (UTC)
I welcome this discussion on the appropriateness of the current criteria for requesting adminship. Whilst appreciating that meeting the criteria is no guarantee a request will be successful, I consider it useful to have a minimum criteria which provides a reasonable guide to the what the community are likely to accept. That is the the interest of both of community and the user in question.
Regarding specifics, I would agree with some other comments that 500 edits on Meta might be too high. I think I'd suggest a the increase in edit count to be focused on activity elsewhere as being an admin on Meta of course requires a good knowledge of Meta policies but the scope of the project demands a good understanding of the broader Wikimedia Foundation projects. Specifying a time period might be difficult, people might join a project and whilst accumulating the level of edits required, may have large gaps when they don't edit which may mean they're less understanding of WMF affairs. I think it is more appropriate to consider the time an editor has contributed on a case by case basis. Adambro 20:32, 11 December 2007 (UTC)
500 on meta: is a bit much, but I'm all in support of: 500+ at home wiki; 3 months on meta:; 250 edits on meta. xaosflux Talk 01:17, 13 December 2007 (UTC)
500 on Meta seems ok - I'd support it - and 500 edits on home wiki is a great idea. 3 months activity - perhaps. I think it may be good, considering the complexity of activity on Meta. --Anonymous DissidentTalk 05:04, 13 December 2007 (UTC)

stake in the ground[edit]

Maybe we should set up a policy subpage and edit it till we're all at consensus. But failing that here's my stake in the ground... or reed, or straw man, whatever.

  • to stand for admin here:
    • currently hold Admin or higher on two wikis (wherever used, wiki == WMF sponsored wiki, my 'crat ship on GLW doesn't count since GLW isn't WMF sponsored) elsewhere (radical new suggestion that just one adminship doesn't cut it).
      • Both adminships have to have been granted by community consensus (even if a small community), they can't be temporaries. Both adminships (or higher) have to be held currently
      • At least one of the Admin holding accounts has at least 1000 edits from the candidate on that wiki
      • At least one ID has to have been an ID held for at least 6 months (doesn't have to be the adminship wiki, but some wiki or another)
      • At least one adminship has to have been held for at least 3 months... the other one can be newly minted for all I care, but it shows two communities trust.
    • 200 edits here. Disclosure of what namespaces (via some tool) so that users can judge for themselves whether they are relevant or makework.
    • 2 months time here since first edit here.
    • disclosure of any banning or significant (more than 1 day) blocking at any wiki under the current id so voters know about it (it's not a bar to adminship, but should not be a secret that is concealed until someone ferrets it out)
    • Crosslinking of relevant accounts, or a fullblown Wiki Matrix to allow for checking the eligibility info
    • Currently serving stewards, developers, foundation board members are exempt from all of this and can have adminship here for the asking, but lose it if they lose their stewardship (dev, fbm), no reconfirmation needed. If they want one that they don't lose, they go through RfA like everyone else, pass or fail. If they fail they still can have the special one anyway. (the wider community trusts them, the board trusts them, etc.)
  • to have standing to comment on an adminship or reconfirmation or deadminship (if such are introduced) here:
    • At least 3 months at some wiki from which you are not currently banned (any ban during any part of the election invalidates the vote (HARSH!!! I know... this is a straw man))
    • at least 100 edits at that wiki from which you are not currently banned (ditto)
    • at least a week here prior to the election/discussion start
    • at least 20 edits here a week prior to the election/discussion start
    • Crosslinking of relevant accounts, or a fullblown Wiki Matrix to allow for checking the eligibility info.

That's my stake. Shoot away. What did I miss? Where am I all wet? ++Lar: t/c 03:41, 13 December 2007 (UTC)

What do you mean by "higher"? Hillgentleman 04:13, 13 December 2007 (UTC)
That probably means higher user classes. For example, a bureaucrat has one higher step of the user class than an administrator.--Jusjih 04:29, 13 December 2007 (UTC)
What do you mean by "higher user classes"? In mediawiki, different user_groups give different rights. There is no "higher". Hillgentleman 04:48, 13 December 2007 (UTC)
Given this offering I think a subpage is necessary - could tackle voter eligibility while we are at it. And "no" there are no higher classes than a user only different rights. I think that kind of "ranking" may be a can of worms too far but much of Lar suggestion combined with the other thoughts here look worth some discussion. Thanks to all - I'll do a subpage later and copy or move this --Herby talk thyme 08:54, 13 December 2007 (UTC)
Higher was a bad choice of words if it is read to imply "better" or status or class divisions... But you'll almost never see someone made a (permanent) 'crat or a CU at a wiki without having first been made an admin, so if you want to think of it as an experience or trust progression, it is an apt shorthand. But I won't debate the point, feel free to change the wording as desired. I should also point out that I don't necessarily think a very numerical approach is the right way to go. Maybe it's more of a straw dog than a straw man, but I wanted to capture all the stuff I'd seen others put forth. On the other hand, a strict numerical system is easy to calculate (and easy to game). Actually talking in depth and analyzing candidates takes a lot of work from people dedicated to doing things here. How many of those do we have? It may set up a defacto aristocracy. (Herby, Pathoschild, Aphaia and a few others who spend more time here proportionally than most)... that's a straw man too, by the way. ++Lar: t/c 13:22, 13 December 2007 (UTC)
Hey I'm aristocracy or maybe just a straw dog - better a cloth cat I think. While numbers are easy to game I think there has to be a base level minimum. Watching some RC RfAs here are rarely a surprise to me. Edits & time would be good & I certainly think "unblocked" would be sensible in some form. Some sort of matrix/linking would be essential.
However - beyond that I would favour Pathoschild's approach. I've not yet got how that would work in my mind & will reflect. Equally while arguments are good voter eligibility would be an issue - we know cavassing in some form exists but it must not overwhelm the views of those working on what is in the end a small wiki? --Herby talk thyme 13:36, 13 December 2007 (UTC)
Small yes. but No Other Wiki (tm) has admins that can affect how things operate on EVERY wiki with a single edit (to the spam blacklist or to the interwiki links). So let's not get the importance of this wrong. Heck, maybe only stewards should be admins here. (another straw man for those keeping score at home). ++Lar: t/c 13:41, 13 December 2007 (UTC)
Agreed (though stewards only .... hum let me think :)). The point is good - equally should it ever arrive I have argued elsewhere & still would that possibly Meta admins could have rights to block open proxies cross wiki for example (stewards, especially new ones, being far to busy) --Herby talk thyme 14:01, 13 December 2007 (UTC)
Small..yes..but ironically, there are more vandalism on these wikis than some of their bigger sister projects...As a member of the SWAT team, I'm saddened by the "HUGE" decrease in the number of editors that actually took interest in watching over/patrolling smaller wikis since the number of members stated here is wrong and we barely have about 12 maximum number of editors patrolling this wiki and only about 2 Stewards and 2 Steward prospects have also joined the project recently and we hope it succeeds since we really need more help...--Cometstyles 14:05, 13 December 2007 (UTC)

No numbers, no votes: presentation and discussion[edit]

(I started a new section because this is completely different from the above suggestions, so we might want to discuss them separately. Please excuse the unusual length.)

A user should not be barred from adminship based on simple numbers, like not being an administrator on two projects. For example, I'm more active on Meta (see stats, self-summary, edit log, admin & steward logs) than many users are on their home wikis. Activity in crosswiki coordination is a better demonstration of good faith and sense than adminship granted on two small wikis populated by a few users (who may even be sock puppets, for all we know), or simply counting edits.

I think the requirements proposed above are primarily hoops— useful for weeding out candidates not seriously intent on Meta adminship, but not particularly meaningful in measuring actual suitability to the role. The problem is that this also weeds out excellent candidates who might simply not want to go through the hoops, while allowing bad-faith users intent on gaining Meta adminship for selfish purposes.

Consider these hypothetical situations (in my opinions):

  • an excellent candidate: active on the spam blacklist talk page and in small-wiki monitoring for three months, Adel has shown excellent skills in conflict resolution, finding compromises, and making logical decisions. It doesn't matter much that he's only a bureaucrat and administrator on one wiki, but he technically doesn't meet the requirements.
  • a bad candidate: Billy routinely attacks or insults other users, has a short temper, and promotes conflicts. He has 600 edits on Meta and an administrator account on two barely-active wikis, so he technically meets the requirements.

The above examples show how numbers mean very little in judging suitability, but are easily exploited.

Instead, I suggest we avoid simple numbers— which I think en-Wikipedia has shown can go horribly wrong— and move more towards meaningful consideration. Users should explain why they want adminship, demonstrate prior long-term activity in areas where they intend to be active (for example, regular discussion on the spam blacklist talk page long before they start editing it themselves), and generally present their cases to the community. "Support" and "Oppose" votes should be accompanied with well-explained reasons; "support" is meaningless, while a lengthy comment extolling a user's qualities (or faults) is excellent. —{admin} Pathoschild 10:11:49, 13 December 2007 (UTC)

I actually think this is very sound (& will move/copy) it to a page covering all this today? However maybe there still should be some form of under pining minimum? Equally I had no idea I would be as active on the spam blacklist when I became an admin as I now am (& like Pathoschild I am rather more active on Meta than I am elsewhere, Commons excepted maybe) --Herby talk thyme 10:22, 13 December 2007 (UTC)
My straw man included just about every possible numerical criteria I could think of that could meaningfully be argued to have relevance... a lot of hoops. But if Pathoschild's proposal could actually be made to work, I'm all for it. I would nevertheless retain the process exemption for stewards/devs/board members. ++Lar: t/c 12:00, 13 December 2007 (UTC)
I agree with Lar and Pathos idea is excellent, though I do believe that Meta should have a stronger policy against sock puppets,and Herby's suggestions on "Supports" and "opposes" is great since on enwikipedia, opposese doesn't really have to accompany a valid reason, though on Meta, I believe opposes must entail a really good reason and one small suggestions I would like to make is regarding the "You have at least 100 valid contributions on the Meta-Wiki.", which I believe should be changed to "You have at least 200 valid contributions on the Meta-Wiki Mainspace" since welcoming new editors and fixing minor errors and making multiple edits to the same page(resisting using 'Preview') can easily get you 100 edits in a week on Meta and the person should have been editing on Meta on a regular basis for atleast 3 Months and it also better to those suffering from editcountis, to judge on edits made overall including Deleted edits then based only on tools..--Cometstyles 12:29, 13 December 2007 (UTC)

Hmpf... Generally I would not support the idea to take a certain edit count or something like this as criteria. I think one who wants to get sysop rights should have experience with the sysop tools on another project (and experience means using the tools actively/regularly), and should be active on Meta for a while, and should make plausible why he/she really *needs* the sysop tools on Meta. As for me, just "I want to help out with this and that" is not a plausible reason. The activity so far should make visible that the candidate is indeed interested in coordinating or cleaning up or spam listing or whatever else a Meta editor might find sysop tools useful for. I would not determine a certain lower bound of edit number or membership or the like. This can be done individually by any voter. (As for me I would not count any edits but I would just look at the edits themselves and decide from them instead of bare numbers.) The most important thing for me is that the candidate gives a good reason for needing the tools. If he/she is active for two months only but has done a lot of good work in that time, why not giving him/her sysop access (if and only if he/she is an active sysop somewhere else). That's just my point of view, just so you know. ;o) --Thogo (talk) 16:07, 13 December 2007 (UTC)


So let's restart this then. Unlesss it's being discussed elsewhence. Perhaps someone else wants to take a cut, we've seen two vastly different approaches, and I think both have some things about them that need to be taken into account. ++Lar: t/c 04:10, 14 January 2008 (UTC)

Hopefully it doesn't complicate things too much, but what would you think of having two different requirements systems?
  1. By demonstrated need and trust
    • Requires knowledge of:
      1. Regex
      2. Intimate knowledge of the primary protected-page that the user wishes to work on.
      3. At least passing knowledge of the procedures of the other "functional" protected pages.
      4. Meta admin policy.
    • Requires a current Meta admin, developer, or steward to vouch for the user.
    • Otherwise qualifies for temporary adminship on Meta.
  2. By community consensus
    • Requires knowledge of: Meta admin policy.
    • Requires agreement to avoid editing "functional" protected pages unless the user is familiar with the function.
    • Requires numeric minimums, e.g. 100 useful edits on Meta, 6 months activity, adminship on at least one other Wikimedia project.
This method, I think, would allow both "regular" adminship for the majority of users as well as "technical" adminship for those who would otherwise apply for temporary adminship, but wish to perform the function long-term. (Users from other wikis helping maintain the SBL, which affects them?) ~Kylu (u|t) 15:17, 6 March 2008 (UTC)

Voters' requirements[edit]

There has been some discussion on this previously - in part to prevent any apparently canvassing I think. If the community were to take up Pathoschild's suggestion above then this might be the eligibility to "comment" rather than vote.

Most of the previous discussion is here. Part of me says that maybe this should be left until we have done the above section but others may not agree. --Herby talk thyme 12:34, 13 December 2007 (UTC)

Seconded. It looks me a logical course. --Aphaia 07:20, 17 December 2007 (UTC)

Policy changes[edit]

Well the discussions rightful place is here, so to begin with first I sort of like A .B.'s proposal though with a few adjustments such as to No.#2 i.e participants must have at least 50 useful, non-trivial edits on meta and must have had an active account here for at least 3 months., I agree with the 3 months part, but "50" useful edits doesn't make sense since we dont want editors asking every admin or crat on their talkpage if they are eligible to vote or not, so to make it more clear, i'd prefer maybe "minimum 50 edits total excluding edits to userspace" to prevent edits only being made for the sake of voting...--Cometstyles 15:06, 12 February 2008 (UTC)

I like A.B.'s proposal too but I'd rather not see too much verbiage about exactly what is a useful edit, etc. I'd rather see wording that gives the bureaucrats discretion to evaluate the contributions and decide, within some limits, whether the contributor is eligible to vote, instead of wording that is a welter of subclauses, exceptions, and special cases which then could be applied by a bot if you could program it cleverly enough, instead of depending on judgement, common sense, and doing the right thing :) ++Lar: t/c 15:12, 12 February 2008 (UTC)
If they are not going to be "useful" edits (:)), can we make it 100 please, otherwise, great & let's get on with it --Herby talk thyme 15:52, 12 February 2008 (UTC)

What I'd like: for candidates, 200+ useful edits... of course, what is useful will be decided by voters - we've had candidates who do nothing but welcome people/edit their user page which were rightly opposed by the community. It is up to the community what is good and bad, not any one user. I don't mind a time limit - some of our best admins were only here a month before they were promoted. If the case is they couldn't even vote after three times that much, something is very wrong. I'd say if there had to be a minimum, a month, but honestly, I find some people "get it" after such a short time.

For participants: even less, obviously. I'd expect a fair amount of time on another project (after all, Meta only exists because of other projects), say about 3 months. But time here, no. I wouldn't impose a limit. On my RfA, for example, there were users who had made barely 10 edits here, but knew me and trusted me from other places. Same with my RfB, and I expect many other people's too. I repeat: Meta exists only for the other projects. I personally welcome new editors to take part in RfAs here... it's a good way to become familiar with the community. What the problem is, is when they vote differently to the more active members in the community. This is where bureaucrat discretion comes in. If I see the community all votes support, but newer editors (but not necessarily new to Wikimedia projects) vote oppose or vice versa, I'd be more inclined to not give so much weight to those "newer" votes. Personally, I find the current numbers proposed way too much, and that adminship is no big deal, and voting on it even more so. Majorly (talk) 16:37, 12 February 2008 (UTC)

Considering the purpose of Meta, perhaps we should base the decisions on the status of the users on their home wiki(s)? Say, "500 edits total on any non-Meta project, must have a userpage on meta with links to those projects." I have to agree with Maj that we're here for the sole purpose of serving the other communities, so we need to have our participation be as open as possible. Admin candidacy on Meta can be more strict, given that we have a temporary adminship position in place for those who don't qualify for perma-adminship and the work they do will affect many locations, not just Meta. ~Kylu (u|t) 01:37, 24 February 2008 (UTC)

Majorly, There is a difference between providing opinions and casting vote on decisions. Everybody can have opinions and yes we should listen to all that is raised . BUT not everybody who is unfamiliar with meta is in a position to decide. There should be no condition for comments, but stricter condition for voting. If anybody has raised legitimate concerns on the trustworthiness of a candidate, that will be reflected in the votes. Hillgentleman 02:16, 24 February 2008 (UTC)