News and notes
Nine new arbitrators announced
The percentage of all voters who supported each candidate: green (two-year terms), blue (one-year term), unfilled (unsuccessful). The outcomes are not linear on this chart because they are based on the S/(S+O) formula, not just the proportion of voters who supported each candidate.
A little more than six days after the close of voting, the results of the annual Arbitration Committee (ArbCom) elections have been announced and certified by four non-English-Wikipedia stewards who kindly stepped in to assist in verification: Vituzzu, Mathonius, Tegel, and Matanya.
Of the 22 candidates, 13 managed to gain more supports than opposes (the same as last year out of 21 candidates). The unusual ternary voting system requires voters to choose one of three options for each candidate: support, abstain, or oppose; the formula S/(S+O) (supports divided by supports-plus-opposes) determines the top eight who will take their place on 1 January for two-year terms, and the ninth-placed, who will serve a one-year term. The successful candidates, in order of these formulaic scores, are:
- 28bytes (candidate statement), a bureaucrat
- GorillaWarfare (candidate statement), an administrator, oversighter, and member of the volunteer response team (OTRS)
- Seraphimblade (candidate statement), an administrator
- Roger Davies (candidate statement), a sitting arbitrator, administrator, checkuser, and oversighter
- AGK (candidate statement), a sitting arbitrator, administrator, checkuser, and oversighter
- NativeForeigner (candidate statement), an administrator and checkuser
- LFaraone (candidate statement), an administrator, checkuser, oversighter, and member of the volunteer response team (OTRS)
- Floquenbeam (candidate statement), an administrator
- Beeblebrox (candidate statement), an administrator and oversighter
Participation in ArbCom elections, 2008–13. Voter numbers are blue (left y-axis); candidate numbers are red (right y-axis).
This year, 923 editors voted, up significantly from 824 last year; this continues a trend of increased participation after the record low two years ago (chart at the right). Only one of the 22 candidates, 28bytes, gained the support of more than half of the voters (chart at the top), compared with three candidates of 21 last year. For the first time ever, the S/(S+O) formula made a difference to the result that would have obtained through supports alone: LFaraone gained a seat at the expense of Guerillero, and a two- rather than one-year term at the expense of Beeblebrox, who was placed ninth under the formula (LFaraone was opposed by fewer voters than the other two).
If you were the "average" voter, you supported just over six candidates, abstained for almost ten, and opposed just over six (the percentages were 28.6%, 42.8%, and 28.6%, respectively). The chart at the bottom shows how the 42.8% average abstain vote has risen significantly. In other words, voting intensity (the proportion of non-abstaining votes) rose steadily between 2008 and 2011, from a low of 25% under the labour-intensive manual voting system in 2008, to 65% in 2011; it has dropped over the two elections since to the current level of 57%. Just why voting intensity should have consistently been the inverse shape of the relative numbers of voters year by year is difficult to explain.
Risker, along with three election commissioners (GiantSnowman, Happy-melon, and TParis), played an important role in the running of the event. She told the Signpost:
||This was a successful election, as can be seen by its results: the wisdom of the crowd shone through. There were some procedural glitches that can be avoided in the future, especially if we as a community move now to update information and explain rationales. Many of the issues were due to outdated information, and the difficult and poorly documented SecurePoll interface. None of those glitches seem to have affected the outcome.
As a retiring arbitrator, Risker is working with the arbitrators-elect to help them move into their new roles. "Every one of them has already proven to be diligent, interested, and enthusiastic," she said.
Like last year, the results bear an uncanny resemblance to those predicted by averaging the recommendations of the 20 voter guides. Constructed by one of the election coordinators, Hahc21 (whose signature is ΛΧΣ), the table for 2013 guides got it wrong for only two candidates, both of them close to the success–failure boundary. The actual voter-guide recommendations are summarised here, and anyone who feels inclined can work out which voter guide came closest to the actual result.
We asked Hahc21 for his opinion of the success of the election, operational and otherwise. Specifically on the last phase of the election—the auditing and tallying—Hahc21 told us: "I'm quite sure that results were posted earlier thanks to Risker's and the stewards' hard work." He thinks that overall this year things ran considerably faster and more smoothly than last year, mostly because of the absence of "insane drama events", including the notorious 2012 email leakings scandal. He is pleased that 2013 is the second year in a row with an increase in the number of voters, but is concerned at the 43% of neutral votes: "This is a situation that is constantly increasing, and if we don't find a way to reduce the high proportion of neutral votes, we will have serious repercussions in the future. ... In my view, all elected users should range between 25% and 30% in abstain votes."
Hahc21 is a native speaker of Spanish, so the Signpost asked him whether he has experience of other Wikipedia ArbCom elections:
||I've not run ArbCom elections on other wikis, but I've witnessed some. If I recall correctly, apart from Meta the English Wikipedia is the only one that uses SecurePoll. When the Spanish Wikipedia had an ArbCom, the elections were just like an RfA, but without a "neutral" option: you could only support, oppose or comment. The same thing currently happens at the German Wikipedia, but they have different terms from the English Wikipedia. Whereas we renew roughly half the committee each year at a single election, the German-speakers renew their entire committee each year in two staggered elections—half in May and half in November. They have a lower number of candidates: the November 2013 election had six, of whom two withdrew and the remaining four were elected.
A number of issues have recently been raised on the election talk page, such as the low number of female candidates and arbitrators (including this insight: "How would injecting Alpha females at the Arbcom level lead to "better" decisions?"), the composition of the committee, the possibility of analysing the demographics of voters, and of course the voting system.
Editors are welcome to leave their feedback on the election at the dedicated page.
Average support, abstain, and oppose percentages, 2008–13. Since 2011, abstains have risen significantly, supports have dropped gradually, and opposes have dropped marginally. (*2008 was the last manual-voting election.)
- Farsi Wikipedia anniversary: The Farsi Wikipedia, also known as Persian and primarily spoken in territory formerly controlled by the ancient empire, has reached its tenth birthday. People within modern-day Iran are the primary users of this language, and the project has faced sharp pressure from the country's repressive government. One study, "Citation Filtered: Iran's Censorship of Wikipedia", has claimed that Iran "targets a wide breadth of political, social, religious and sexual themes [on Wikipedia]. Censorship is aimed at restricting access to information related to the Iranian government’s human rights record and individuals who have challenged authorities. Pages about disfavored beliefs, religions and sexuality are also prohibited. Importantly, nearly all of these censored Wikipedia pages contain speech that is protected by international law".
- Toolserver/Labs petition: A petition is open on the German Wikipedia. It was kindled by the fate of the Toolserver, which is operated by the German Wikimedia chapter, and its replacement Wikimedia Labs, which is run by the Wikimedia Foundation in the US. The location change is a key critical theme, given the recent surveillance revelations leaked by Edward Snowden; the signatories are arguing for more restrictive Labs policies concerning the handling of user data, in line with the traditional German-run Toolserver approach that forced user:X!'s edit counter into opt-in mode earlier this year.
- Open positions: The Wikimedia Foundation is advertising several job openings, including a legal internship (summer 2014), a graphic design internship (user experience), and a full 11 openings for software engineers/developers working on site infrastructure, AQ testing, mobile, internationalisation, and features (including one for VisualEditor). In addition, the Cochrane Collaboration is looking for a Wikipedian in residence, preferably with a scientific or medical background. The successful applicant will work for six months and will receive a US$6,500/£4,000 stipend.
- In the media:
- Fallen Astronaut: The famous sculpture on the Moon received attention from Slate this week, which revealed that the involved sculptor and astronaut believe they are "unfairly maligned" on Internet blogs and Wikipedia.
- Wikipedia advent calendar: The Guardian has covered the Wikipedia advent calendar, counting down until Christmas.
- British Library releases free images: The UK's premier library has released more than a million public-domain images from 17th-, 18th- and 19th-century books onto Flickr Commons. Says the library's Ben O'Steen: "The images themselves cover a startling mix of subjects: There are maps, geological diagrams, beautiful illustrations, comical satire, illuminated and decorative letters, colourful illustrations, landscapes, wall-paintings and so much more that even we are not aware of.". There is a guide to using them on Commons at Commons:British Library/Mechanical Curator collection
- Arbitration report: A proposed decision has been posted in the "Ottoman Empire–Turkey naming dispute" case. There are two new open cases, Kafziel and Nightscream. The evidence phase for both will close on 29 December 2013.
- Signpost changes: The "Featured content" section now has two new lead editors, Planet Herald and Brambleberry of RiverClan. "In the media" now has Mark Miller as a contributor.
Keep up with The Signpost