Wikipedia has a long history of collaborating with educational institutions. The Schools and Universities Program—international and in many languages, but dominated by American institutions—started in 2003 and evolved case by case with little system. This changed in 2009 as Wikimedia embarked on its formal strategic process, and outreach in higher education came to be seen in terms of achieving explicit goals—especially that of increasing editor participation.
The first "strategic" Wikimedia Foundation program to emerge was the Public Policy Initiative (PPI) in 2010–11 (see Signpost coverage: September 2010, July 2011), a highly successful pilot to improve the English Wikipedia’s articles in the field of US public policy, centering on the participation of college classes supported by Wikipedia ambassadors, and funded by the Stanton Foundation. The foundation then expanded the model beyond the US, most notably with the Indian Education Pilot 2011 in Pune. This ended in abject failure (Signpost coverage), largely due to a lack of community engagement by the consultants running the pilot, and it resulted in significant damage to the reputation of the education program within the Wikipedia community. Now the so-called US–Canada education program is in reform mode, attempting to resolve the structural problems and controversial issues that have arisen over the past year.
Other programs have since included chapter-run pilots in Europe, and the Cairo pilot as part of the foundation’s Arabic language initiative, which gathered pace under the guidance of the PPI core staff just as the problems in the Indian program were coming to light and the US education program was scaling up the PPI structures.
The Arabic language is a priority for the Wikimedia Foundation because of the significance of the language and its peoples as opposed to the relatively small size of the Arabic Wikipedia and editing community. As the sixth most popular language of the world, Arabic is the native tongue of about 300 million people. It is spread laterally thousands of kilometres from Morocco, the Spanish Sahara, and Mali on the Atlantic coast of North Africa to its millenium-old eastern boundary on the Iraq–Iran border. Arabic's cultural and scientific heritage alone make it an important part of the goals of the movement to provide free knowledge for everyone; but its importance as one of the world's great contemporary languages spoken by great numbers of people in societies undergoing radical change have highlighted the value of supporting movement-related education programs there. Egypt is the largest of the Arab-language nations, with some 80 million speakers. Egyptian Arabic is just one of many dialects; although there is a small Wikipedia site devoted to the dialect, the education program has focused on the standard-Arabic Wikipedia.
The recent pilot deployed methods to overcome significant challenges in both real-life and on-wiki environments, and has produced positive results despite the current political instability in Egypt and more widely the wave of demonstrations and protests in the Arab-speaking world since December 2010, known in some quarters as the Arab Spring.
While the Indian pilot and some participating college classes in the US education program suffered from structural problems, such as an unbalanced student–ambassador ratio, and were widely criticised as being out-of-touch with the editing community, the Cairo pilot was carefully designed to marshal strong support for the program and individual students from the local university and Arabic Wikipedia communities. This was achieved through careful organisation at the institutions and by reaching out in advance to the Arabic Wikipedia community, especially local community members on the ground in Cairo, and limiting the number of participating classes to seven, each with a maximum size of fifteen.
The Cairo pilot—which involved article creation and the editing of existing articles, and both writing in Arabic from scratch and translating Spanish and French Wikipedia articles into Arabic—operated at two state universities, Cairo and Ain Shams, with combined enrollments of more than 370,000 students. Up to thirteen hand-picked students were in each class. All lecturers underwent a mandatory orientation procedure in which they tackled questions such as how Wikipedia handles copyright in practice (an issue unearthed at the last minute during the Pune experiment). WMF staff published all pilot-related documents at a central place on-wiki to maximise transparency, and the course pages at Cairo and Ain Shams (Google translations) made it clear who was responsible for what, and displayed running progress reports for each article edited in the pilot.The program produced solid overall results, with 246 articles edited by 56 participating students. Major contributions to the Arabic Wikipedia by article creation, improvement and translations like of the Spanish articles of the author Laura Restrepo (Google translate) and the painter Ernesto Sabato (Google translate) or the Panic of 1907 (Google translate) into Arabic were made during the pilot. Both lecturers and participating students as well as community members involved were largely satisfied with the initiative and its conduct and a large number of students taking part in this terms project signed up as campus ambassadors for the next term, thereby expanding the previously small local Cairo community.
Thus, while not every article was improved, and not every student contributed substantially, the output showed more than just establishing that the format can be exported into the Arabic language version of Wikipedia and participants already tabled ideas how to improve the interaction between classes and ambassadors further generally. The program managed to go beyond its original project design goals and has provided lessons for tackling the US education program reform’s large-scale issue of rebalancing classes with local and editing communities consistent with the English Wikipedia’s guidelines.The Signpost was unable to speak with local participants at short notice, but LiAnna Davis, the foundation's Wikipedia education program communications manager, made contact with us after an all-day meeting in Cairo to discuss how the movement can use the results to further improve its education initiatives. "We had no idea if the Wikipedia Education Program would work in Egypt, especially given the uncertainty following the Arab Spring. We wanted to start very small in case it didn't work. Our goal was to see if the program could work in Egypt, and if it was appropriate for professors and students to edit the Arabic Wikipedia with help from ambassadors. The answer was that it absolutely worked. Taking students, professors, and ambassadors together," LiAnna said, "50% were very satisfied, 38% somewhat satisfied, 4% mixed reactions, and 8% somewhat dissatisfied."
We asked LiAnna to elaborate on the statement in the draft pilot report that "the campus ambassadors' skills for effectively helping the students need to be improved". For the initial campus ambassador training," she said, "we recruited current Cairo Wikipedians to lead the training in Arabic. We suggested some topics, but it was a community-led process focusing on what they thought was most important for students to know from their ambassadors. We expect ... a learning process that will enable them to continuously improve their ability to support the teachers and students." LiAnna and her colleagues are delighted that "the students who participated in the pilot were very eager to be campus ambassadors next term and help improve the training to focus on what they as students receiving help from Ambassadors felt was most needed."
A notable aspect of the report is that some female students were uncomfortable in communicating online with people they didn’t know, and had a male student in the class write as proxies. We asked LiAnna whether this signals a larger problem in getting female Arabic-speakers to participate in WP. She pointed out that it just wasn't possible to recruit female online ambassadors because the community isn't yet big enough. The 15 online ambassadors (as opposed to the campus ambassadors) were drawn from the ranks of experienced Arabic Wikipedians, who number fewer than 70 over the entire Arabic-speaking population worldwide, and are almost entirely male. "One positive is that 87% of the students who participated were female, and they've expressed a lot of enthusiasm for continuing to contribute to Wikipedia, so we're hopeful we can have an impact on those numbers as the program grows larger."
Was there a strategy of avoiding any topic that might be controversial, such as those related to public policy/politics in modern Syria and modern Egypt? "Not from the foundation's end", she said: "the articles were decided between the professors and the students. The non-translation classes were in subjects like mathematics, literature, language, drama, and communications—course-related topics. There was a bit more leeway in the translation classes, where the very first article posted on the Arabic Wikipedia was a translation of the French Wikipedia article on civil disobedience.""According to our survey, 92% indicated they'd like to continue. Of course, other responsibilities often get in the way of good intentions, but we are hopeful that many will continue to contribute in some form to the Arabic Wikipedia. Students, professors, and ambassadors all spoke extensively about how important it is to have information available in Arabic on the internet, and how it's their civic duty to improve the Arabic Wikipedia. I'm really looking forward to seeing this program mature and what kind of difference we can make on the Arabic WIkipedia."
The only technical hitch was apparently that participants found the left-to-right wikimarkup confusing and unintuitive (almost all semitic languages are written from right to left). LiAnna says that the foundation technical department has been informed of this issue.
Reader comments
In news that has come in just as we publish this week's edition, the Russian Wikipedia has been blacked out for 24 hours until 20:00 UTC Tuesday as a protest against a bill currently before the Duma (the Russian parliament) that proposes mechanisms to block IP addresses and DNS records. Visitors to the Russian Wikipedia are confronted by the sign above in protest at a bill before the parliament that appears to enable significant internet censorship. The Russian word for Wikipedia is crossed out in this banner, and the text says: "Imagine a world without free knowledge. The State Duma is currently conducting the second reading of a bill to amend the "Law on Information", which has the potential to lead to the creation of extra-judicial censorship of the Internet in Russia, including the closure of access to the Russian Wikipedia. Today, the Wikipedia community protests against censorship as a threat to free knowledge that is open to all mankind. We ask that you oppose this bill."
The action by the Russian Wikipedia echoes the worldwide blacking-out of the English Wikipedia on 18 January as a protest against two bills before the US Congress (see previous Signpost coverage and Wikipedia initiative). The head of Wikimedia Russia, Vladimir Medeyko (User:Drbug), told the Signpost that in practice this will probably give the government authority to take action against internet websites—potentially including Wikipedia—and it could be delegated to police officers and intelligence operatives without due diligence. "Although the draft legislation allows for judicial appeal, in Russia this is typically a very lengthy process that is likely to lead to significant harm to internet freedom."
A post on Wikimedia Russia's Meta blog states (translated by the Signpost):
“ | The Russian Wikipedia will be closed on 10 July because the Russian parliament is debating amendments to information law that could create real censorship of the internet—a blacklisting and filtering of internet sites. Supporters of the law say it is aimed only at widely prohibited content such as child pornography, ... but we believe the conditions for determining where content falls under this law will create something akin to the Chinese firewall. The existing Russian legal environment suggests that the worst-case scenario is not unlikely, in which access to Wikipedia will soon be closed throughout the country. | ” |
Medeyko says the second reading of the bill is scheduled for tomorrow, Wednesday, Moscow time. "What makes it so dangerous is that the decision to ban an address may be issued by any 'duly authorised' agency, with unexplained procedures for providing evidence to the 'registry'. The very vagueness of this wording is most concerning. The Russian Wikipedia community moved quickly with a proposal for the black-out—it was voted on in only one day—and implemented with 292 in support, 22 for posting a banner only, and 74 against any action. The decision was that no one would edit the project during the blackout, although administrators are technically still able to do so."
The Russian minister for information, Nikolay Nikiforov, has made several tweets that suggest the government may not force the law through in the short term, including: "I am sure that this law will still be accepted, to the 2nd of reading only a few managed to avoid controversial positions and extend up to 1 Nov." and "the idea of combatting child pornography on the web is correct. But in general, the Internet should remain a free environment".
The Wikimedia Foundation's efforts regarding editor engagement having recently focused on achieving three results: the rollout of version 5 of the article feedback tool; research into the effectiveness of the existing MoodBar extension; and the creation of the LastModified extension, representing the first feature experiment completed by the new editor engagement experiments (E3) team, founded in April 2012.
The fifth incarnation of the article feedback tool, AFT5, started in August 2010 as part of the public policy initiative. The feature empowers readers, who otherwise wouldn't participate in the discourse on an article, leaving publicly displayed feedback at the bottom. A prominent example is Higgs boson, on a topic that is at the center of public attention over the CERN experimental results. The article feedback tool received 400 responses, ranging from praise to editorial suggestions and trolling, so far. While SPat, an author of the article, praised the tool, there remains the question of how to handle the display or removal of inappropriate input. The tool is due for deployment in some 10% of English Wikipedia articles by July 17, and issues surrounding its use are still open for community debate.
Another project inherited by E3, the MoodBar, started in June 2011; the tool aims to improve the interaction between new editors and the established community. Results published over the past month suggest that while new users who use MoodBar tend to be more active than the average new editor, "there is only mild and circumstantial evidence that the reported mood is associated with a higher or lower edit count".
In May 2012, E3 undertook a timestamp experiment to test whether making the article history more prominent would result in its wider usage, and thereby an article more transparent to its readers (see also this week's Technology report). Results based on a sample of about 20,000 English Wikipedia articles show that IP editors and readers clicked on the normal "View history" tab more than twice as often when the prominent timestamp was present.
While the timestamp experiment has now closed and the AFT and MoodBar will continue to be tested (see the main research page for more information), the specifications of the new post-edit feedback tool and the long-awaited new RC tool, under development since March 2012 as a monitor of newly created pages, have yet to be finalized.
The Wikitravel proposal, launched in April 2012 (Signpost coverage on April 16 and 23) is still live after nearly three months. On July 6, the RfC entered its second round having been promoted in a site notice from April 18 after weeks of debate on Meta as to whether Wikimedia should host a travel guide project.
The move came after weeks without substantial progress on Meta, while the German NGO that hosts Wikivoyage, a second project that forked away from WikiTravel in 2006, is applying to join the proposal, in effect using a migration to the WMF umbrella to merge back with WikiTravel. Wikimedian Doc James, who is strongly supportive of the Wikitravel proposal, forwarded the Wikivoyage proposal to WMF board trustee Samuel Klein and WMF executive director Sue Gardner on May 27.
It has been claimed that the stalling, despite wider community support for the experiment in the RfC, suggests a broken innovation process on Meta and the responsible committee has yet to sort out its own formal framework, the WMF board is scheduled to look at the application during its meeting at Wikimania in Washington DC this week. Volunteers interested in the proposal are due to meet at the conference on July 12.
The Signpost notes that the WikiTravel site currently hosts sidebars of direct Google travel-related advertising, although WikiVoyage site does not. While there would be no direct advertising for information from a travel site migrated to the foundation, the nature of consumer tourism does expose the WMF's narrative text to unseen pressure to mention and describe commercial operators, which will in itself represent a highly effective form of product placement under the foundation's authoritative logo.
This week, we spent some time with WikiProject Football, which focuses on the sport also known as association football or soccer. WikiProject Football is by far the largest sport project and one of the most active projects on Wikipedia in terms of the number of articles covered, edits to articles, and talk page watchers. In the seven years since the project was founded, nearly 400 active members and an equally large number of former members have contributed to the project. There are a wealth of football-related subprojects and taskforces. To learn how things have changed since our previous interview in March 2008, we turned to project members BigDom, League Octopus, WaitingForConnection (WFC), Thumperward (Chris Cunningham), and Cloudz679.
What motivated you to join WikiProject Football? What team(s) do you support? Have you ever played in a football/soccer league?
When we first interviewed WikiProject Football four years ago, the project was home to 34 Featured Articles, 42 Featured Lists, and 56 Good Articles. Since then, the project's collection has exploded and now encompasses 65 Featured Articles, 146 Featured Lists, and 323 Good Articles. How did this happen? Why have Good Articles grown so much faster than the Featured materials?
How often do you encounter regional differences in football rules, terminology, and tradition? How are conflicts resolved?
In addition to English, versions of WikiProject Football exist in 31 other languages. Have you had any contact with editors from these other versions of Wikipedia? Has there been any sharing of content or resources among WikiProject Football's counterparts?
How difficult has it been to acquire images for football articles? Are some countries or time periods more difficult than others to find appropriately licensed images?
WikiProject Football is by far the most active sport-related project on Wikipedia and claims a very large membership. What has attracted so many editors to this project? How can other WikiProjects emulate WikiProject Football's success? Would WikiProject Football be open to collaborating with editors from less-active sports projects?
What are WikiProject Football's most pressing needs? How can a new contributor help today?
For those with some interest in the sport, one good way to contribute is to periodically update the prose of articles on people that interest you. Given that a lot of football articles do not seem to be developing, I think all editors – regardless of their knowledge about the sport – could help by providing feedback on whether our notability guidelines for football seem in line with the general notability guideline, and if not, how we might change them. Internal discussions about guidelines have at times been contentious – some fresh input could provide inspiration.
Next week's article will be a tour de force. Until then, ride over to the archive to read our previous reports.
Reader comments
Eight featured articles were promoted this week:
Eight featured lists were promoted this week:
One featured picture was promoted this week:
No cases were closed or opened, leaving the number of open cases at three.
The case concerns alleged misconduct with regards to aggressive responses and harassment by Fæ toward users who question his actions. The case was brought before the committee by MBisanz. The other parties are Michaeldsuarez and Delicious carbuncle. A decision was expected on 6 July.
In response to a workshop proposal calling for the removal of his adminship, Fæ's administrator rights were removed at his request on 18 June; he has declared he will not pursue RfA until June 2013, and that should another user nominate him and he feels confident to run, he will launch a reconfirmation RfA rather than requesting the tools back without community process.
The case concerns behavioural issues related to Ohconfucius, Colipon, and Shrigley. The accused parties deny TheSoundAndTheFury's claims and decried his alleged "POV-pushing". According to TheSoundAndTheFury, the problem lies not with "these editors' points of view per se"; rather, it is "fundamentally about behaviour".
In the proposed principles, drafting arbitrator Hersfold reaffirms that Wikipedia is a reference work built through consensus and written neutrally, that it is not a battleground, nor a soapbox for propaganda, that editors are expected to act civilly and respectfully and pending further cases where long-term disputes cannot be resolved, the committee may "adopt draconian measures as a last resort for preventing further damage to the encyclopedia."
In view of the dispute, the committee has observed that certain involved parties have pushed their own points of view; in particular Homunculus has edited to enhance the Falun Gong movement and discredit the Communist Party of China, whereas Ohconfucius and Colipon have been found editing in a manner which enhances the party and discredits the movement. The former was found engaging in uncivil conduct and the latter failing to assume good faith. Further, Ohconfucius and Homunculus have engaged in edit-warring on topics related to the movement.
Given these findings it has been proposed that Colipon, Homunculus and Ohconfucius be topic-banned from articles related to the movement and "prosecution thereof". Mandated external review by uninvolved administrators was also proposed; editors placed on review would be subject to the following restrictions:
The case, filed by P.T. Aufrette, concerns wheel-warring on the Perth article after a contentious requested move discussion (initiated by the filer) was closed as successful by JHunterJ. The close was a matter of much contention, with allegations that the move was not supported by consensus. After a series of reverts by Deacon of Pndapetzim, Kwamikagami and Gnangarra, the partiality of JHunterJ's decision was discussed, as was the intensity of Deacon of Pndapetzim's academic interests in the topic. Questions were also raised about the suitability of the new move review forum.
In a workshop proposal, uninvolved user Ncmvocalist outlined in proposed principles the need for administrators to lead by example, behave respectfully and civilly in their interactions with other users, learn from experience, and avoid wheel-warring irrespective of the circumstances or nature of the dispute; and that WikiProjects are not platforms for point-of-view pushing or the pushing of one's own agenda and where consensus cannot be reached other venues of discussion should be sought out. Proposed decisions are due on 12 July.
The committee resolved unanimously to revoke Carnildo's administrative tools for "long-term poor judgement" in his use of the tools, particularly in view of the recent block he issued to Itsmejudith. Carnildo may regain the tools via a successful request for adminship.
Reader comments
The results from last month's trial of the LastModified extension were published this week on the Wikimedia blog. The first analyses have indicated a significant positive impact, suggesting that the extension – which makes the time since a page's last edit much more prominent in the interface – could eventually find its way onto Wikimedia wikis.
This more prominent display (see right) was added to some 20,000 English Wikipedia articles, linking directly to the full revision history. The results of this trial were nevertheless surprising: rather than click on the new timestamp, visitors preferred to click directly on the history tab, indicating salience of its location. "The increase was particularly strong for anonymous editors and readers, who landed on the history page more than twice as often (+120.6%) [as the control sample]", explained Steven Walling, on behalf of the Foundation's editor engagement experiments (E3) team. "For registered users, there was a smaller but still significant increase in article history views (+42%). This result was seen even when we controlled for repeat clicks on either link".
Despite this apparently positive result, Foundation developers (perhaps feeling the effect of recent controversies) seem wary of pushing the extension onto communities for the moment. Anonymised data has been released, but thereafter the E3 team will simply move onto new experiments – such as "transforming this timestamp into a more direct call to edit articles that are severely outdated, though clearly the point at which an article becomes out of date is somewhat subjective" – rather than handing over the extension to a deployment focused team.
Also published this week was an extensive analysis of the impact of the already-deployed MoodBar extension. That research, also suggestive of a broadly positive impact, is expected to be followed up with a further study to eliminate the possibility of selection bias.
"At midnight UTC on July 1, Wikimedia’s search cluster stopped working" (Wikimedia blog). The proximal cause was surprising: the insertion of a single leap second, so that June 30 officially had 86,401 seconds rather than the usual 86,400. This caused significant problems, not just on Wikimedia wikis but across the web, affecting sites such as Reddit, Foursquare and LinkedIn. Wikimedia's search services were restored in slightly less than two hours.
As Lead Platform Architect Tim Starling explained, "leap seconds are added to our clocks once every few years so that the sun will be directly overhead of the Royal Observatory in Greenwich at precisely 12:00. Some people believe that the desire to keep these two time standards synchronised is anachronistic, and that it would be better to let them drift apart for 600 years and then add a single “leap hour”. I’m sure many computer engineers would breathe a sigh of relief if such a change were implemented."
A knock on, more general problem was finally resolved on July 2 following technical work by the WMF's network operations team (wikitech-l mailing list). Given that those problems seem to have affected the Foundation's servers hosted only in Tampa, Florida (and not their Ashburn, Virginia counterparts), the implication is that ageing hardware that is expensive to replace could have exacerbated the problem. The Signpost hopes to publish an interview with a member of the operations team about this and other recent issues in the near future.
Not all fixes may have gone live to WMF sites at the time of writing; some may not be scheduled to go live for several weeks.