Wikipedia:Wikipedia Signpost/2018-05-24/Op-ed



Wikipedia Talk:Requests for adminship, once the most lively forum on the project with the exception of ANI, is becalmed. The babble of noise at peak times akin to the background din of a noisy Manchester pub on a Saturday night has dropped not just to a whisper, but to a stony silence. It's become an empty space. Walk through it and you'll make conspicuous footprints in the dust gathering on the floor. Your footfall echoes in the deserted room, "Is there anyone there?" you call halfheartedly before turning round and going back outside into the sunlight. In the street there is also little activity. A kid sits on a doorstep poking vandalism into K-Pop music bios through his cheap Chinese smartphone. A bunch of teenagers sit round their Samsungs seeing who can tag the most new pages in sixty seconds.

The whole place has the feel of a deserted Wild West film set. Across the almost traffic free road, you enter a half open door with a dilapidated sign hanging on one nail: Café Anna it announces in sun-faded coloured letters. Inside the pretty room with many colourful interactive pictures on the wall, there is a stuffed effigy of 'The Founder' in an armchair, but there is no busy bustle in this place either. There's a lady editing an article about a golf complex in Hainan. The article is almost as complete as the complex. She's looks startled when you enter – she obviously wasn't expecting anyone. "Yup, it's kinda quiet here," she says. "April's customers are down to a fifth of the usual month's average."

You go out again, follow a sign pointing to 'Administration' hoping find a bit more action. A sound of music gets louder as you approach. Inside the office there's an elderly man idly picking out a Telemann Fantasia in funk on a piano. On a table is a sign that says 'For Adminship apply here'. The pianist lightly bangs a discordant cluster and drops the lid over the keys raising a cloud of dust. "Yup, it's kinda lonesome here. Not many enquiries. Ya wanna sign up?" he asks. "No, no," you hasten to reply "I'm one already". You go out and walk some more in this almost ghost town until you hear some loud tapping. Following the sound you come across a guy nailing a board to a telephone pole. It's list of names. "What's this?" you ask, "Lynchings?"

"Nope, desysopings" he replies. "What did they do?" "Nothing. It's what they didn't do" "What didn't they do?" "Edit. That's what they didn't do."

You make your way back to your car. Driving back towards civilisation you think to yourself: "Golly, there's enough in that creepy place for a Signpost article..."

Has the AdminShip finally run aground?

This month another five admins tacitly lose their tools
These admins were all quite active until they suddenly stopped. None of them placed a retirement notice on their user pages. We hope they are well.
 * Cenarium joined Wikipedia November 2007 and made ~27,700 edits. RfA (42/2/2) 20 June 2008,  abruptly stopped editing April 1, 2017
 * Al Ameer son joined Wikipedia March 2007 and made ~55,000 edits. RfA (85/1/3) March 2009, abruptly stopped editing April 24, 2017
 * Lupo joined Wikipedia December 2003 and made ~25,310 edits. RfA (17/0/0) August 2004, abruptly stopped editing April 19, 2017
 * AliveFreeHappy joined Wikipedia December 2003 and made ~20,600 edits. RfA (44/1/2) December 2007, abruptly stopped editing April 27, 2017
 * MichaelBillington joined Wikipedia March 2006 and made ~18,000 edits. RfA (89/1/1) 24 March 2007, abruptly stopped editing April 25, 2017

The admin corps have lost 39 members a year on average since 2012 (that's net, counting the 20 created annually via RfA). But as everyone knows, averages are a weak metric and the actual rate of attrition is increasing: down 50 in the last 12 months. The rate of loss greatly exceeds that of replenishment and is steadily getting worse – only 3 new admins this year and we're nearly halfway through 2018. We aren't at the crisis point yet, but at some time in the not too distant future we will be.

We are not alone
In a Signpost special report in 2012, – now Lead Manager, Wikimedia Foundation – describes some RfA reforms that were made on the German Wikipedia, but that was six years ago. Otherwise noted for its exceptionally dynamic chapter infrastructure, its Administratoren system now also seems to be stuck on a sandbank. In a 2018 12-frame presentation Admins and  attempt to explain why the wind has gone out of the sails of the central European sysops:

"Why is admin work so unpopular?" they ask.


 * There are fewer and fewer users to choose from
 * Too much responsibility
 * High pressure
 * Expectations of the community
 * Old things are dug up, "mud slinging"
 * Always exposed to big discussions when controversial – increases fear of doing something wrong when making decisions

Causes of admin decline
 * More and more admins are stepping back
 * No more desire
 * Too stressful
 * Boring discussions
 * Focus more on real life
 * Takes too much time
 * Inactivity

Solutions
 * Address and encourage experienced users
 * Suggestions in recent surveys on the administration system:
 * Separate voting and commentary phase
 * Abolish AWW system (Administrator recalls) and introduce recurring admin elections

Conclusion
 * Even small modifications to the admin election system could defuse mud slinging and encourage more users to vote – initiate RfC?

Does that all sound familiar?
At the English Wikipedia we went through this introspection seven years ago in a massive 2011 research. None of the posited solutions were proposed to the community for debate.

How it all began: In this discussion from Wikipedia Nostalgia on creating a corps of Sysops, although some of the speakers were aware of a need for controls, they were not expecting the English Wikipedia to become what it is today and needing over two thousand admin accounts to be created over the course of time.

Years later in the January 2013 issue of The Signpost, produces a very accurate exposure of the situation five years ago.

However, it is debatable whether the intended purpose of RfA is at least maintaining the number of administrators; the candidates need to be ushered in from somewhere else first. Nothing changed except the slope of attrition got steeper and the mud slinging continued. Following an extremely complex system of low-participation RfCs at Reform Phase II in December 2015 (see screenshot), the number of RfA voters was doubled by allowing additional RfA publicity, and the pass mark was reduced. And still nothing else changed except the slope of attrition got even steeper and the mud slinging still continued.

The most recent serious discussion Planning for a post-admin era begun by Hammersoft on 16 December 2016 about the vanishing of admins and expected qualifications for candidates went on for eight days and ~70,000 bytes, probably because an unusual cluster of five RfAs was taking place in the same month: Bearing in mind that we are now reading this in mid 2018, it's interesting to note that he went on to say "...the decline in active admins, overall admins, nominations to RfAs, successful RfAs, and re-sysoppings shows no sign of slowing down. In fact, all factors are arguably getting worse. We may see some dead cat bounces, but given declines showing consistency over these last many years, it seems unlikely to change".

More recently in one of the increasingly rare threads at RfA Talk one user, admin, claims "RfA is a discussion, not an election, so disagreements should be discussed. That said, if ten people have already raised the same criticism of a certain !vote, one does have to consider whether they really need to add another response." while the reply from is almost the antithesis: "Except that it is actually an election, with numeric cut-off points for pass and fail, and even 'crats acting as sort of electoral college for grey-area cases. RfA is both an election and a discussion, and no amount of Wikipedian distaste for vote counting is going to change that." Whoever is right, today's RfAs are generally passed (or failed) on such a large consensus that messing with the math in any section is not going to change anything. The bureaucrats rarely intervene because there's nothing for them to do without taking some flak themselves. As long as they can be fairly sure that their action won't change the outcome, they won't do anything either. And the mud slinging will continue.

Interestingly, Wikipedia's two greatest RfA success stories happened within a few months of each other in the second half of 2017. In 227 edits comprising 133,277 bytes the talk page of the second most successful RfA Megalibrarygirl (Susan Barnum), librarian and mother,  saw what has probably been the most lengthy dispute over voter behaviour. At 282 supporting votes against only 3 in the opposition section, the candidate has been accused of incompetency and sexism to an extent that leaves one wondering just how much gender bias or even blatant misogyny is indeed embedded in this male dominated database. Barnum and her Wikipedia work were featured on Web Junction and in LibraryJournal in March this year.

On the most successful RfA of all time, the reluctant candidate who had to almost be dragged kicking and yelling to the process, one of the two opposes was based on the very civility and sensitivity that has made such a well liked editor. Perhaps his user name was an unintended premonition of the number of support votes his RfA would reap. Certainly his nominator's predictions were not wrong. One of the three neutral votes came from a well established user who finally admitted they confused Cullen328 with another editor.

The underlying main cause for the reluctance of candidates of the right calibre to come forward, despite claims of exaggerated demands for tenure and/or experience, appears to be the environment at RfA itself. Some oppose votes are a death’s head at a feast, deliberately spoiling an otherwise immaculate run for the bit. It's borderline vandalism like sauntering down the street and throwing mud at someone's freshly hung out clean washing; the actual rationale for such votes is often contrived. Sometimes they arrive with a totally inappropriate question, and when that doesn't shake the candidate, they think up some other reason to oppose. While a lone vote like that isn't going to make any difference to the outcome, the community has the right to show their distaste of it, but it causes a long discussion on the talk page where passions run high and otherwise constructive byte time is wasted. Due to its tradition of being the one place where our 5th pillar gets struck off its plinth with impunity, RfA sometimes brings out the worst in our best editors.
 * Conclusions

It has been suggested that the community's fears of 'adminship for life' allowing admins to perpetrate their perceived tyranny forever, is a contributing factor to the exaggerated criteria, but it has also been posited that this aspect is the least thing users have in their minds at the moment of posting their votes. A close examination of the oppose sections of RfAs demonstrates that a large number of the votes are one-off aggrieved users who have correctly been warned or whose work has been correctly reedited by the candidate, or often simply bad faith that causes a "category 5 hurricane in an industrial-sized tea urn". Sanctions as a solution or even as a deterrent have proven ineffective. Partial T-bans (allowing a vote but no further commenting) have been suggested as a remedy and just two in all the years have been enacted. Punitive? Maybe, it's like being slapped in the face in public: the sting to the pride, especially to that of a prolific FA editor, for example, is far more severe than the sting to the flesh, and the shame lingers longer, but admonished users risk becoming even more irascible towards candidates and other admins even to the point of unprovoked harassment.
 * Solutions

Restricting participation to users who meet a minimum threshold for voting, such as the practice on the German Wikipedia, may not help. It may contribute towards more objective voting, but the disingenuous votes and personal attacks seem to come from seasoned editors who seem to be fairly sure of themselves knowing that they can get away with saying anything they like. Clerking by abstaining users; clerking by abstaining admins; clerking by bureaucrats; splitting the RfA page into two distinct sections like the German model, with numerical votes on one page and threaded comments on the talk page only; secret poll every 3 months on the Arbcom model with its 'voter guides' and questions, with all candidates reaching a certain score being accepted. These are all ideas that were discussed in 2011. In fact looking back, it seems as if every imaginable solution was at least mentioned.

Often cited as a remedy: "Fix the voters and RfA will fix itself", Risker, an influential editor and former long-term Arbcom member, indeed states: Qualifying her opinion further:  It's a detail that  has not hitherto been discussed in depth. "Nobody should have more than 50% opposes" she adds.

Where it stalls however, is the paradox that where no official entry point exists for admin candidates, one can hardly impose regulations on those who vote. It's a Catch-22 question. The English Wikipedia is the only major language project not to operate such restrictions. But until it happens, the mud slinging will continue...

In comparison with the number of admin actions carried out daily, issues regarding admins actually appear to be rare, but they are significant enough to discourage candidates from coming forward. Leading towards new, positive reforms, maybe another 'dead cat bounce' is needed in the dialogue surrounding RfA and adminship issues.

In next month's issue, we'll be taking a closer look at what the actual work of an admin entails.