It seems that based on a recent Signpost article, another round of RfA reform has come into play. So here are my thoughts on becoming an admin, being an admin and ending adminship. The concepts are inter-related and should be considered together.

According to the stats, there are 703 active admins today. However, "active" means 30 edits in the past two months. By that statistic, there were 777 active admins on the 31 March 2011. It's interesting to note though, that on that date, there were only 107 admins on this list who made at least 1000 edits in a month. So yes, there are probably a few hundred active admins, but very active ones? (For comparison there were about 500 admins on the list, who had made 100 edits)

Perhaps this is because vistor base is falling? Not really - we've fallen from a peak of 2910 "very active" editors in 2006 (100 edits per in a given month) to 2440 in 2012.^[1] That's a 16% fall. The total number of database changes has fallen by about 25%.^[2] For comparison, the number of active admins fell during the period 2008 - 2011 by about 30%.^[3] The number of RfA requests has fallen by 85%.^[4]

"RfA is a horrible and broken process"
We all know the mantra - RfA is broken, it's a gauntlet, a place where incivility is rife and only perfect editors get through. How true is this though? There are a number of things we know for certain:

• The number of administrator candidates is falling drastically. At it's peak, there were 2-3 requests for adminship per day, now we're looking at that number in a fortnight. In December 2005 we promoted 68 admins in a single month, between January 2011 and June 2012 we've promoted 61.
• Potential good candidates refuse to go through the process, due to the perception. Anecdotally, I've reviewed over 30 editors. About a third, I believe would pass an RfA. Only 2 have accepted. Speaking to other editors on this, I find the same. It is clear that not all good candidates are refusing, as we still get passes.
• Standards are rising. Since January 2009, no candidate has passed with less than 3000 edits.^[5] The average number of edits a successful candidate has increased from ~16,000 edits in 2009 to ~25,000 in 2011^[6] In the past 12 months, only 2 admins were promoted with less than 7,000 edits, and only 4 with less than 2 years experience.

There are also a number of other complaints about RfA, which are generally opinions held by members of the community. These points can be disputed

• Too many questions are asked
• The "wrong sort" of questions are asked - complaints about stock questions, irrelevent questions etc
• People can support/oppose for the "wrong" reason
• People can support/oppose without giving a reason
• Civility rules are "thrown out the window" at an RfA
• The candidate has to have (content creation/deletion knowledge/X edits etc)
• RfA was "better before"

It's a simple question, and one that is hard to answer. The problem comes in defining the cause and the effect - did the standards rise because fewer editors came through or vice versa? Do people avoid it because it is a terrible process, or it it considered a terrible process because people avoid it?

If we could get a definitive answer to this question - we could "fix" RfA. Replacing it with another system, when this question has not been answered will not improve the situation.

The psychology of adminship is an interesting one. Adminship starts as a bunch of tools which are used enact the community's will. Deletion should only be done when the community says a page should be deleted, blocks shouldn't be controversial and so on. From that point of view, it truely is "no big deal" - in fact admins are just the community's dogsbody.

However, that's not the point of view taken by most editors. It's easy to see why. It starts off small, admins are expected to weigh up decisions at XfD, because they are the only people who can press the final button. When there's no consensus, the admin will maintain the status quo... and suddenly they've got a supervote. Admins are put on a pedestal - they have the power to delete anything, block anybody, see what's been deleted in the past. The mystery of the role is enough to engender respect in some, and distrust in others.

As a subset of wikipedia editors, administrators are often stereotyped. One stereotype is that administrators cover each other's backs or apply one rule for one editor and another rule for another. Another stereotype is that incivility is only incivility when it comes from a non-administrator. These stereotypes are almost always true for some admins.

Adminship is currently a lifetime role, once you've being given the bit it stays with you. However, there are a number of circumstances where the bit can be removed.

• Voluntarily giving up the bit. This can include upon request - if the administrator is open to recall.
• Having the bit forcefully removed - currently this is only done by Arbcom.
• Inactivity - the bit is removed after 1 year inactivity. It can be returned on request in this situation.

There are upsides and downsides to "lifetime" adminship. On the upside, with the fear of accountability removed, it allows administrators to make tough decisions. However, on the flipside, it pushes forward the concept of abusive administrators.