User:Durin/Admin voting measures

From Wikipedia, the free encyclopedia

I frequent WP:RFA. I've noticed a number of people voting on RfAs that use criteria for voting for/against admin nominees. I disagree with these approaches, and my reasons are below. I've created this page to show my reasoning for my disagreement, and to offer some additional approaches as an alternative.

Some people have errantly concluded that this page constitutes an insane level of standard that a nominee needs to exceed to get my support vote. One user concluded that 99 of 100 candidates would not pass my standards. Another user said on IRC that I should be hung from a flagpole so that we can get more admins. These people are very much in error.

What you see below is not a set of hurdles that a nominee must exceed in every respect, without fail, or I will oppose. What I've outlined below are methods I use to craft an opinion of a nominee. They are not hurdles in any respect. It's certainly possible that a nominee could not exceed any of these measures and still find me voting in support of them.

I have been actively working to reduce the standards many people use to vote against candidates, not increase them. Prior to my efforts, just 48% of nominees with less than 2,000 edits succeeded. Since I began trying to reduce standards, that number has risen overall to 57%. I can't take credit for all of the rise, but I'd like to think that my efforts played a part in reducing standards.

Contents 1 Objective criteria approaches 2 My approaches 3 No big deal 4 My standards for bureaucrats 5 Summary

[edit] Objective criteria approaches

I think strict objective criteria developed by some for determining appropriateness for adminship are often misguided. Some of these objective measures are:

• Edit count: It's not the number of edits. An avid stub sorter can make their edit counts skyrocket in short order. Meanwhile, a person doing substantial improvements to a series of articles (say, on Spruance class destroyers) could be rated significantly lower using a strict edit count basis yet contribute a LOT more material. I've done both of these types of activity. Both deserve merit, but are entirely different sorts of activities. Gauging their merit by the number of edits each receives is misguided; it's like judging a spacecraft vs. a car based on their miles per gallon (liters/mile) ratings. It's a very poor tool. I think we use it because it's available, and little more. Kate's recent upgrade to the edit counting tool to break out namespace edits is helpful in this area, but still leaves us with a hammer and us looking at everything as a variety of nail.

• Similarly, saying "must have 2000 edits" is far too arbitrary. If a person has 1999 edits, they won't qualify for adminship, but when they make that 2000th edit they are magically imbued with the ability to be an effective admin? I don't think so. The qualities that make a person a good admin existed before they ever came to Wikipedia. Thus, a person with 1 edit has the same qualities as a person with 10,000.

• Time on Wikipedia: I used Wikipedia for seven months prior to making my 10th edit. My first two edits were by an IP I don't use anymore. Users like me could conceivably know Wikipedia:Administrators' reading list by heart, and have read every ArbCom case there has been. They could also have similar knowledge of all Category:Wikipedia official policy and Category:Wikipedia guidelines. They could know all of this information without contributing a single edit to Wikipedia. Yet, most people voting on WP:RFA look to how long a nominee has been actively contributing edits. This does nothing to judge how long they have been reading and studying Wikipedia.

[edit] My approaches

With that in mind, I've developed the following approaches (not criteria; that's too limiting):

• Edit analysis:
  
  A chart showing editing behavior. This is an objective tool, not a measure, to help show pattern of behavior based on edits. Yes, I know this somewhat contradicts what I said above. It is just a tool. Not a measure. An example of this is shown at right. This is useful for showing patterns of behavior. If, for example, we were to see a large number of edits over a month period, followed by a long duration absence of edits, there might be a reason why the person vanished. So, ask the question of the nominee. It can also be useful for showing other large fluctuations in participation.
• Average edits per day: This is a derivative of the above edit analysis. Yes, it also disagrees somewhat with some of my comments above. But, if a nominee has a low edit average per day output, they are likely to remain a passive editor in the future. An admin needs to be active, not passive. I would generally rate an average of less than 10 edits per day as reason for concern, but not a reason to eliminate. I voted neutral on the nominee analyzed by the above chart because of a low average, not oppose.
• >3000 edits, no vote: Almost all nominees with more than 2000 edits become admins. From 23 June 2005 to 11 September 2005, of the 126 nominees less than 9% of the nominees with more than 3000 edits were rejected (less than 6% withdrew). In general, if I see a nominee with more than 3000 edits, I won't vote unless there is debate on the user. In a few days, the number of positive votes will usually so overwhelm neutral or negative votes that it's a moot point, and a positive vote is lost in the maelstrom. User:Func's recent promotion to adminship (Wikipedia:Requests_for_adminship/Func) with 112 yes votes highlights this. Yes, I know a neutral or negative comment might cause other potential positives to vote otherwise, but I think this is a rare case. I don't tend to pursue low hanging fruit on Wikipedia; there are plenty of people for this sort of work.
• Familiarity: Do I know the nominee? If so, I'm more likely to vote on the RfA (one way or another). I don't keep a black-list of people I would vote against on an RfA. That said, my interactions with them in the past may influence how I vote for some nominees.
• Personality: Some users I do not think have the personality to be an admin. They may make fantastic contributors, but are ill-suited to being an admin. In general, an admin needs to be able to maintain a level head at all times. They also need to know when to disengage themselves from a conversation. For example, I had a discussion with an admin (User:Bkonrad) at Talk:Gordon Springs, Georgia. That discussion nicely concluded, and I don't think any feelings were hurt or tempers raised. It could have easily devolved into a mud slinging match. I exited the conversation when I felt progress could no longer be made. This was appropriate. A sense of humor is not required, but usually helps.
• Distribution of edits: A nominee needs to demonstrate participation in a broad variety of admin appropriate areas. Understanding policies and guidelines is very important to a nominee's ability to be a good admin. A nominee having a solid distribution of edits across namespaces shows a willingness to participate in a variety of facets of Wikipedia. I think this is important to the success of an admin. Further, contributions to areas such as village pump, AfD, RfC, RfA, and vandal fighting help show that the user is trained in admin appropriate areas.
• Conflict: Sometimes, nominees answer the "Have you been in any conflicts over editing..." question with saying they've never been in a conflict or anything approaching a conflict. If a nominee has not demonstrated their ability to handle such a situation, I count this as a strong negative towards adminship. Admins must be able to effectively deal with conflict.
• Lack of contribution control: As a Wiki editor, a nominee should not show a desire to control the outcome of an article towards their particular world view. I'm not talking NPOV here. What I've seen in some editors is considerable upset because their particular contributions were edited by others. If a nominee is not willing to let go of their contributions when they press the "Save page" button, they will make a bad admin.
• Edit summaries: While it is true that vandals frequently fake their edit summaries to try to get past the vandal watchers, it is still an important function to help clarify what an editor was trying to do. I like to see nominees almost always use edit summaries when not editing their own user pages. I recently saw a nominee who in their last 100 edits used edit summaries just 60% of the time. Yet, the vast majority of people are voting in favor of this admin. This is disappointing to me. I voted in opposition to this nominee, even though the user had nearly 6,000 edits.

[edit] No big deal

While it is true that admin status should be 'no big deal', I think it is important to be careful in appointing admins. It is far more difficult to remove an admin than it is to make one. An admin's voice carries more weight in controversial areas, and as such needs to be wielded by a wikipedian who has the skills needed to properly handle that responsibility. For me, the "no big deal" part comes in the form of the effect it has on people who are nominated to adminship; it shouldn't cause some dramatic shift in their behavior or outwardly demonstrated character.

[edit] My standards for bureaucrats

• Impeccable, unquestionable behavior in all regards.
• An administrator for at least one year.
• Significant involvement in RfA and associated talk pages.
• I do not feel that "we don't need more bureaucrats right now" is a valid reason for exluding someone from being a bureaucrat.

[edit] Summary

I will add more to the above list as time goes on. What is important for my own 'criteria' (I dislike that word in this context) is that the approaches are malleable depending on the circumstances and the people involved. Each nominee deserves to be subjectively judged on their merits, not on an arbitrary quotient achieved through mathematical machinations. Working with the above approaches, I generate a reasonable picture of an admin candidate.