The Signpost
Single-page Edition
WP:POST/1
10 October 2011

Opinion essay
The conservatism of Wikimedians
News and notes
Largest ever donation to WMF, final findings of editor survey released, 'Terms of use' heavily revised
In the news
Uproar over Italian shutdown, the varying reception of BLP mischief, and Wikipedia's doctor-evangelist
WikiProject report
The World's Oldest People
Featured content
The weird and the disgusting
Arbitration report
Senkaku Islands closes; administrators authorized to place articles on discretionary sanctions
Technology report
1.18 deployment and HTTPS switchover completed, but developer help now needed with new projects
 


2011-10-10

The conservatism of Wikimedians

Contribute  —  
Share this
By Ironholds

Oliver Keyes (User:Ironholds) is an administrator on the English Wikipedia. The following article on the conservatism of the Wikimedia movement was adapted from an August 27 2011 post on his website, Quominus.org. Oliver previously spoke about related issues in his address to the 2011 Wikimania conference "Hippies with Guns: how ideological conflict shapes Wikipedia and what we can learn from it".

The views expressed are those of the author only. Responses and critical commentary are invited in the comments section. The Signpost welcomes proposals for op-eds. If you have one in mind, please leave a message at the opinion desk.


A word cloud generated from the skills statements from the Strategy wiki's call for participation.

The status quo

Above is one of my favourite images at the moment – a graphical representation of all the key words found in taskforce planning for the Wikimedia Strategy project, in which the community got together and basically crowdsourced a long-term strategic plan. The reason I find it interesting isn’t because of the words which are displayed prominently, but the words that aren’t; specifically, those looking to move the community forward neglected the word “community” (slightly to the right of “work”) and the word “social” (slightly above “working”, and, in comparative terms, dwarfed by the word “php”. Sigh.)

I’m not going to lie, this doesn’t surprise me. Wikimedians actually tend to put a fairly small amount of stock in changing things to boost the community or the social aspects of the movement. Whether it’s WikiLove, help reform or any other project to ameliorate the less pleasant aspects of the projects, the same refrain comes from an annoyingly large chunk of the community – “I managed to edit and contribute to the projects without [whatever is being proposed], so other people can too”.

The source of this is fairly clear – people don’t like change, and because existing editors are largely comfortable with the current situation (after all, they built it, either accidentally or deliberately, and since they’ve stayed around we can conclude they don’t mind it that much) they don’t really see that there’s so much of a problem. In a way, I reckon this is a result of our successes more than our failures; while new user numbers are dropping, we maintain an enviably high retention rate for existing editors. As a result, experienced users don’t see a problem. Why would they? Oh, sure, they hear rumours that user numbers are dropping, but all their friends are still here, so it can’t be that bad. Cries that “yes, all your friends are still here, but you just missed out on a thousand new ones” aren’t met or internalised well.

The problem

My problem with all of this isn’t just that there is genuinely a lot of stuff we need to do to Fix the Wiki that people don’t get, it’s that this position is based on a pair of fallacies that the individuals in question could, with a bit of effort, think through and avoid. When people say “I managed to edit and contribute to the projects without [whatever is being proposed], so other people can too”, what they’re saying is:

  1. The on-wiki system has remained the same
  2. The people we’re trying to recruit are the same as the people we already have
The on-wiki system has remained the same

People drop ideas for reform on the grounds that they managed and therefore anyone else can. This is based on the fallacy that the system has remained the same since they started, which is of course not the case – except, possibly, for our newest users. Even users who are a year old are dealing with a completely different environment from the one they started off with; policy bloats, our tolerance for expansion has been reduced, and the community has become increasingly harsh and automated in its dealings with new users over time. The fact that the system in 2006 when Exampleexperienceduser joined was nice and characterised by decent, friendly communication and a million opportunities to expand the wiki does not mean that the current one is.

The people we’re trying to recruit are the same as the people we already have

In other words – for the aforementioned statement to be true, the potential new editors now have to be largely identical in terms of motivation, interests and enthusiasm to the existing users. We can’t say that our current system works for new users because it works for established users unless there are actually similarities between the two groups.

Why this matters

The problem is that there aren’t. Early adopters of Wikipedia were, by-and-large, nerds. Utter nerds. I don’t mean that in a bad way, I mean that in a “it was 2001 and they were on the Internet” way. Longhairs from the 1960s and 1970s counterculture movement (cf. Fred Turner’s From Counterculture to Cyberculture, recommended), early computer and Internet proponents, FOSS addicts, codejunkies, we had them all. Early adopters were people who saw a nascent project – or later on, an established project that everyone with any academic or political value ripped the hell out of – as something it was worth contributing to, for free. Early adopters were happy to use a pseudo-HTML markup scheme to contribute and delve deep into the raw guts of an article or talkpage to make their views or facts known.

What we’re dealing with now in terms of potential volunteers is largely different. People bandy around the term “Facebook generation”, but it’s true; we’re dealing with a completely distinct group of people. 2011 users are likely to be people who have grown up with the Internet and with Wikipedia. If you’re at university now, Wikipedia will have been around since you entered secondary education, and possibly before that. It is no longer new and exciting, it is no longer a step into the unknown, and so the motivations of people who do try to contribute are likely to be pretty damn different. Moreover, the people we’re dealing with don’t want to have to deal with markup, or raw pagetext in this fashion; they aren’t used to looking under the hood. If they’re WordPress users they live in a world in which looking under the hood is not mandatory; if they’re Facebook users they live in a world in which looking under the hood isn’t even possible.

So we’re not dealing with the same system, or even the same people; we’ve got a bloated and increasingly unfriendly wiki on the one hand and a pool of potential recruits with vastly different motives and expectations from our norms on the other. This means that we can’t simply sit in our ivory tower and make judgements about whether Wikimedia is failing, or succeeding, or what our potential users expect; we have to go to those potential users and ask them. The problem is that the Foundation has been doing exactly that in a lot of places, trying to get information on what people want from the horse’s mouth, and when acting on the data (as with WikiLove) has largely been met with a big stubborn sign saying Do Not Want.

What we can do about it

Wikimedia is a fantastic movement. It’s the Whole Earth Catalogue for the 21st century, a smorgasbord of educational material and trivia that for the first time offers the potential for people to truly take their education into their own hands – after a fashion, anyway. But if we want to live up to the expectations, if we want these projects to continue, we have to start accepting the data people bring us and accepting that we might need to make some changes. We need to throw out our assumptions, throw out our fallacies and innate resistance to change, and seek to build a movement that people want to contribute to rather than one people don’t mind contributing to.

Reader comments

2011-10-10

Largest ever donation to WMF, final findings of editor survey released, 'Terms of use' heavily revised

Stanton Foundation give $3.6 million grant

The Stanton Foundation is the philanthropic legacy of the late Frank Stanton, president of CBS from 1946 to 1971 and who organised the first televised presidential debate in the United States.

The Wikimedia Foundation announced this week that they have received a $3.6 million grant from the Stanton Foundation. According to the press release from the Foundation, the funding will go to support development of the long-awaited visual editor and other technical improvements including the addition of editing facilities to the new Wikimedia mobile site. The grant can also be put towards any project that "make[s] Wikimedia a friendlier and more understandable environment for new editors".


The $1.2 million grant in 2010 from the Stanton Foundation helped fund the Public Policy Initiative and a 2008 grant of $890,000 helped fund usability improvements. This week's $3.6 million donation is the largest ever to have been received by the Wikimedia Foundation.

Brief notes

Screenshot of the winning entry of WikiViz 2011.

2011-10-10

Uproar over Italian shutdown, the varying reception of BLP mischief, and Wikipedia's doctor-evangelist

Media storm for Italian Wikipedia shutdown

Following the announcement and discussion covered in last week's Signpost, Italian Wikipedia was shut down for three days in protest against a proposed law that could have held bloggers, web site owners and possibly even Wikipedians liable for posting corrections to content within 48 hours or facing a stiff fine.

The Wikimedia Foundation rapidly endorsed the shutdown with a blog post entitled Regarding recent events on Italian Wikipedia; the endorsement was reported on by Business Insider. The shutdown was widely covered by the media with articles posted on the websites of BBC News, The Independent, The Atlantic, Euronews, The Register, TechCrunch, Ars Technica, Techdirt and CNETA News. An Associated Press report was published on a number of sites including Forbes.

Cynthia Wong posted an entry on the blog for the Center for Democracy and Technology claimed that the Wikipedia shut down shows that if passed the law "would, in effect, shut Italy out of the global community of Wikipedia users".

During the shutdown, Jimmy Wales weighed in through an interview posted on the International Journalism Festival website. In the interview, Wales claimed: "The decision was taken by the Italian community in part because they felt that there was no genuine avenue for protest in the mainstream media without a bold action."

As the Nieman Journalism Lab blog reports, during the shutdown a hashtag appeared on social media website Twitter called #graziewikipedia, with users of the microblogging service showing support and solidarity for the shutdown. Cory Doctorow blogged about the story for BoingBoing describing the proposed law as "punitive" and "batshit". News aggregation sites Reddit and Hacker News both had lively discussions of the protest.

On Thursday (October 6), as AFP reports, Italian Wikipedia reopened following an announcement that the law would be amended to only cover traditional news media sites. The Wall Street Journal's TechEurope blog reported the amendment following the strike as "in the short-term at least... a victory for Wikipedia".

Canadian doctor-editor recruits colleagues in the lecture theatre

Wikimedia Canada president Dr. James Heilman (User:Jmh649), physician and Wikipedia evangelist

TheSpec and Physorg both covered the Wikipedia evangelism of Wikimedia Canada president Dr James Heilman (better-known hereabouts as User:Jmh649) at McMaster University. Heilman was there to beseech the academic community to engage with the encyclopaedia, entreating that "We know it is extensively read and we know the quality of information is hit and miss. Some is good. Some is poor. We want to provide high quality medical information to everyone around the world and to do that we are going to need the help of our colleagues." Citing the high usage rate of Wikipedia's medical articles by both patients and physicians (of whom between 60% and 70% consult the site according to surveys cited by Heilman) as evidence for the importance of improving coverage of medical-related topics, and recounting his own experience as a contributor, Heilman called on faculty to incorporate Wikipedia editing into university assignments. It was, he asserted, a "huge missed opportunity when you have these very intelligent people doing research papers and learning all about an issue just to have the paper sit on a shelf ... we can leverage the knowledge to improve everyone's knowledge about health-care subjects. We all have an interest in making sure good information is available". The event attracted 30 participants. Last month, Heilman had published an article in a medical journal explaining to his colleagues "Why we should all edit Wikipedia" (Signpost coverage), and Wikimedia Canada is currently offering a C$1000 scholarship to the Canadian student who makes "the most significant contribution to Wikipedia's medical content."

In brief

  • Kindle Touches only Wikipedia: In an upset to Amazon Kindle fans used to availing themselves of its free boundless Internet – but perhaps a marker of Wikipedia's commercially friendly ubiquity – PC Mag reports that the new Kindle Touch 3G eReader will permit users to access only the Kindle Store and the online encyclopaedia without charge. Time's Techland column derisively remarked "Where can the Kindle Touch go? Try the Kindle Store, no surprise, and—wait for it—Wikipedia, that fount of immaculate, always-reliable and scholarly information, and encyclopedic repository for whoever your favorite celebrities are dating."
Shiju Alex, prominent Malayalam Wikipedian joins Wikimedia Foundation in India as Consultant, Indic languages
  • Indian outreach projects advance: The Deccan Chronicle and Malayalam newspaper Mathrubhumi hailed the appointment of Malayalam Wikipedian Shiju Alex as consultant for the Wikimedia Foundation's India Programs, whose office opened in South Delhi – the first outside the United States. In acknowledging his appointment, Alex emphasised the importance of getting Wikipedia into Indian schools, and getting Indian students into Wikipedia: "The most significant project would be Wikipedia's entry into school curriculum. A project would be developed in association with education departments and students will be asked to create Wikipedia articles as part of the internal tests." Elsewhere, the Indian Daily News & Analysis website hailed the "rise of the Wikimedians", in reference to the forthcoming WikiConference India 2011, scheduled for 18–20 November in Mumbai.
  • Congressman and hockey star BLPs targeted: The LA Weekly picked up on unseemly editing related to the race for the redrawn 30th congressional district of California. Sitting Democratic congressman Brad Sherman's entry was altered to include allegations that he was a recovering alcoholic and aficionado of the company of men, who had abandoned his Jewish faith to become a Scientologist. The paper found that the IP address responsible for the edits geolocated, mysteriously, to Narberth, Pennsylvania. The Chicago Sun-Times unearthed mischievous edits to the biography of ice hockey star Dave Bolland which contended among other things that "if he gets an unsportsmanlike-conduct penalty, he grins maliciously to the sin bin". Unlike most targets of unencyclopaedic editing, Bolland appeared appreciative of the efforts, remarking of them that "Whoever does them is pretty good ... It's funny. It's awesome seeing that stuff."
  • Science wars rage on: Evolution News, a project run by the Discovery Institute (advocates of intelligent design and critics of Darwinism), took issue with the article on evidence of common descent, offering a point-by-point critique of the page's assertions. Meanwhile, USA Today's ScienceFair column reported accusations by the Deep Climate blog against prominent statisticians Edward Wegman and Yasmin Said of plagiarising Wikipedia in their academic papers. Last year, several uncredited passages in another text by Wegman, Said and one other author—the influential Wegman report for the U.S. Congress that questioned the validity of climate change research—were found to have likely been plagiarised from Wikipedia (Signpost coverage: "Report on climate change for the U.S. Congress plagiarised Wikipedia and other sources").
  • Foundation acquires new servers: The Comms-Express noted the acquisition this week by the Wikimedia Foundation of three new server racks, thereby doubling its application-server capacity.
  • Reliability of Wikipedia explained to kids: Teen Kids News, a US TV news magazine aimed at young viewers, asked: "Is Wikipedia Trustworthy?" The four minute segment features a young Wikipedia admin and User:Pharos from Wikimedia New York City, and quotes the 2005 Nature study that had found the reliability of Wikipedia to be comparable to that of Encyclopædia Britannica.

    Reader comments

2011-10-10

The World's Oldest People

WikiProject news
News in brief
Submit your project's news and announcements for next week's WikiProject Report at the Signpost's WikiProject Desk.
Jeanne Calment had the longest confirmed human life span in history, reaching the age of 122
Henry Allingham, a British supercentenarian who was the world's oldest living man for one month, cited "cigarettes, whisky and wild, wild women" as the reasons for his longevity
Frank Buckles, pictured with then-president Bush, was the last surviving American veteran of World War I

This week, we spent some time with WikiProject World's Oldest People, a project dedicated to improving biographies and lists of notable centenarians and supercentenarians. The project got off to a rocky start in March 2008 when its founder, an enthusiast from Yahoo's World's Oldest People discussion board, attempted to affiliate the WikiProject with the Gerontology Research Group. After years of edit warring, hostile debates, deleted page histories, removed images, AfDs, and arbitration, the project finally found peace and has accumulated six Good Articles and one Featured Article. The project supplements the guidelines of its parent, WikiProject Biography, with some notes on notability, sourcing, and the treatment of biographies of living people within the project's scope. We interviewed two project members, SiameseTurtle and David in DC, along with independent contributor The Blade of the Northern Lights.

What motivated you to join WikiProject World's Oldest People?

SiameseTurtle: I was involved in researching the World's oldest people before joining Wikipedia. I got involved because I wanted to help spread the information to the public, as there are many longevity myths that appear even in the most reputable news sources. Wikipedia has a wonderful way of a lot of information in a simple way.
David in DC: I was troubled by the poor sourcing, and dubious notability, of many articles in the longevity suite. It was also - fairly ubiquitously - ignoring, flouting or openly deprecating common Wikipedia norms (only rare edit summaries, WP:WALLEDGARDEN, WP:OWN). I felt it needed input from wikieditors who were not longevity experts and hobbyists, to try to keep the project from turning Wikipedia into a webhosting service for the Gerontology Research Group and the World's Oldest People Yahoo group. As I got more involved, I became fascinated by the topic of human longevity and how the normal human lifespan has increased so radically. Notable long-lived people in Shakespeare's time would have been 50 or 60 years old. Today, that lifespan is remarkable for being short.
The Blade of the Northern Lights: Though I'm not actually a member, I have been very involved with the WikiProject from around November. I saw accusations of a walled garden mentality, and I thought I could help out with them. As I got a little further in, I too became extremely interested in longevity. The main catalyst came when in real life, I started taking a senior-level college class on World War I, and Frank Buckles, the last American veteran from WWI, was close to turning 110; this spurred me on to stay involved in the topic, and I have to this day.

How does the project handle articles about disputed or missing records of a person's longevity?

SiameseTurtle: In the case of missing records, the people's ages will not be validated by an international organisation, such as the Gerontology Research Group. They're unlikely to have an article themselves without substantial news coverage (eg. Sahan Dosova). Without proof, they are treated as claims to longevity. In terms of people who are disputed, this is slightly different. Disputed in this context means that a person's age was originally thought to be valid, but new information casts some level of doubt on that age. Or it may be that the specific date of birth is unknown, perhaps due to lack of historical records (eg. Moses Hardy). Obviously these are sensitive areas, so we state the facts. However to be scientific, we do have to go with the validated date of birth. Shigechiyo Izumi was originally thought to have lived to 120 - the oldest man ever. However Japanese researchers later discovered that he was probably named after an elder sibling who died young. The case was retracted last year.
David in DC: It relies on "experts" who make these decisions off-wiki and self-publish them on a set of Gerontology Research Group web pages.
SiameseTurtle: These tables are also published on a regular basis in the peer-reviewed journal Rejuvenation Research. Due to the dynamic nature of the subject, with birthdays and deaths happening constantly, the tables are also published online so they can be updated more frequently, as David in DC mentions.

How is notability determined for articles about the world's oldest people? How frequently does the project deal with disagreements and misunderstandings about notability?

SiameseTurtle: Notability is usually determined using multiple birthday reports. But it often depends on how much information those articles have. So it can take many years to gather enough information to write an article. And of course, they have to stay alive, which is a big obstacle since around 50% die each year. There are often disputes over what is deemed notable. Some articles might be borderline and might be merged into national lists of supercentenarians (eg. List of British supercentenarians). However, we recently sorted through many articles to see whether they met the general notability guideline or not. In short, it really depends. Sometimes there are many disagreements at once, and at other times very few at all.
David in DC: The sorting is a long-term project. Notability is one issue, but reliable sourcing problems still abound.
The Blade of the Northern Lights: To expand upon what David in DC says, there are a few people who straddle the line of notability. It tends to be a bigger problem with people from smaller countries where less attention is given to longevity, so while an American like Walter Breuning will have a lot written about him, someone from Belgium will probably have less, which has led to a few very contentious AfDs. Last year around November/December, some of the members of the Wikiproject (with some outside assistance) went to cut through the large walled garden that had accumulated, and while most of the debates came to a definitive conclusion there are still a few on which no consensus has developed; these are primarily on people in countries with less consistent coverage on this topic.

The project is home to one Featured Article, one A-class Article, and five Good Articles. Have you contributed to any of these articles? What are some challenges to getting an article on a supercentenarian up to FA or GA status?

SiameseTurtle: I have contributed to Henry Allingham and Jeanne Calment's articles. The main challenge is that most supercentenarians are not famous before 110. All the information about them has to be garnered from news articles, which are often quite similar year on year. They generally give patchy information about their past life (where they were born, what their job was, where they lived), and what they do now, and what they believe to be the reason for their longevity. Most of the best articles are of people who served in World War One, because of the amount of press attention they received.
David in DC: I've contributed to many articles. I've dealth with Jeanne Calment on a few lists but I don't know if any stand-alone articles I've contributed to are FA, A or GAs. The longevity tables are shaded with legends for categories like "Verified", "Unverified", and "Disputed". These categories largely mirror the information on the GRG websites.

Are images of the world's oldest people difficult to obtain? What lessons have you learned from acquiring the images currently used in articles?

SiameseTurtle: It's very difficult to obtain uncopyrighted images. The main reason is that very few people get to visit these people. They don't really get out very often, so they are rarely seen publicly. In addition, most photographs come from birthdays and it's mostly just family and the press who are invited. So nearly all of the images of supercentenarians are copyright (and also very recent), which is an issue when trying to get images for Wikipedia.

What are the project's most pressing needs? How can a new member help today?

SiameseTurtle: I would say the biggest issue is a lack of detail in Wikipedia articles. The newspaper articles are out there (which may or may not be in another language), but many members do not have the time, or may spend their time on Wikipedia updating the many lists we have. Thanks to the digitisation of many newspapers, it has been possible to lengthen articles such as Margaret Ann Neve and John Mosely Turner, who both died long ago. It really shows to me that articles can be rescued, it often just need a little bit of effort. I would also say that images are a real problem. Studying the World's Oldest People is a very human thing, and it's a shame that we do not have as many images as we would hope.
David in DC: The project needs many more generalist, wikipolicy-wise editors to help it reach consensus about sources and notability.
The Blade of the Northern Lights: I think the project would do good to encourage some more dispassionate editors to get involved in the topic area; notability to extremely passionate people may not jive with Wikipedia's notability guidelines, and the more acrimonious debates tend to occur when the two intersect. Sourcing is also an issue for some (though certainly not all) articles as well. Part of that is simply a lack of diversity in third-party sources reviewing such material (the Gerontology Research Group is the one most frequently used), but at the moment there are too many articles with justified refimprove tags in the topic area.

Anything else you'd like to add?

SiameseTurtle: The WikiProject does not encompass all people who research the World's Oldest People, just those who are regulars on Wikipedia. We do have many people from countries all over the world who help contribute to the work we do here. Fortunately that also means that we have a very broad view on our project as we use articles from everywhere - not just the English-speaking world. I think that makes our group very encompassing and thorough in what we do.


Next week, we'll see where some of the world's oldest people might have lived. Until then, continue to read about the good old days in the archive.

Reader comments

2011-10-10

The weird and the disgusting



Reader comments

2011-10-10

Senkaku Islands closes; administrators authorized to place articles on discretionary sanctions

This week by the numbers; edits and page views.

The Arbitration Committee closed one case this week, Senkaku Islands. Another case, Abortion, is still pending before the Committee.

Senkaku Islands

Senkaku Islands has closed, a month and a half after it was originally filed by Qwyrxian. Two parties were topic banned, Tenmei indefinitely and Bobthefish2 for a year, and the former was also banned from Wikipedia for a year; Arbitrators cited the editor's past Arbitration sanctions in another topic area in the decision. Unusually, the Committee also considered an indefinite ban for the editor; but while five Arbitrators supported it, they did so only as an less-favored alternative to the 1 year ban. STSC was given a warning.

Standard discretionary sanctions were also authorized for the Senkaku Islands topic area, marking the sixth time in the past twelve months that the Committee has authorized discretionary sanctions. In contrast, from October 10, 2009 to October 10, 2010, the Committee only placed a topic on discretionary sanctions three times. The case also marked the first time the Arbitration Committee has devolved the authority to place discretionary sanctions; any "uninvolved administrator may, after a warning given a month prior, place any set of pages relating to a territorial dispute of islands in East Asia, broadly interpreted, under standard discretionary sanctions for six months if the editing community is unable to reach consensus on the proper names to be used to refer to the disputed islands."[nb 1]

  1. ^ Although not specifically stated in the final decision, it appears that at least two Arbitrators' intentions were to exclude the island of Taiwan from this remedy.

Deliberation on checkuser and oversight candidates

Public commenting on the candidates for checkuser and oversight functionaries (see previous Signpost coverage) has concluded, and the Committee is expected to release its decision on who it is appointing by October 15. This deadline has been pushed back several times in previous stages of the appointment process; the decision date was originally to be October 5. The list of candidates that ArbCom is considering is:

Clarifications and amendments

The Committee also received two requests for clarification, one on WikiProject Conservatism (which was filed as a case request)[1] and the other on the propriety of administrators overturning unblocks by the Arbitration Committee. It closed two other clarification requests, one on the naming of Ireland articles and the other on Eastern Europe. The Committee also continued to consider a request for amendment by William M. Connolley, who had filed a request lifting removal of sanctions imposed on him in Climate change. As of publication, an unusually high number of 24 other editors had commented on the request, most of whom had been previously involved with William M. Connolley or the topic area in some form. The Committee received an additional amendment request from mediator Steven Zhang, who requested that two editors recently topic banned at Arbitration Enforcement be allowed to continue editing a mediation discussion page.

Reader comments

2011-10-10

1.18 deployment and HTTPS switchover completed, but developer help now needed with new projects

1.18 deployed to remaining wikis

With the dust settled from the first two phases of the deployment of MediaWiki version 1.18, on October 4 the full rollout programme began. Since small but technically interesting wikis had been the focus of the first two phases, only 2% of traffic had been routed through MediaWiki 1.18; now, all visitors and editors will be able to take advantage of its new features, which include support for gender-specific user pages and better directionality support for RTL languages. According to a post on the Wikimedia blog, progress with the final phase was slightly slower than anticipated but still good, with the French, Polish and English Wikipedias, along with Wikimedia Commons, having their version of MediaWiki updated within the first four hour window. Other wikis were then transferred during secondary windows on October 5 and 6. The deployment did not go perfectly, however, and a number of bugs have since been discovered with 1.18; as of time of writing, approximately 40 open ones are currently being tracked under the auspices of bugmeister Mark Hershberger, although few are serious. The list of bugs reported but not yet fixed includes problems with the watchlist API (bug #31526), the localisation update system (bug #31559) and the display of <math>...</math> (bug #31442). Responding to the relatively high volume of bugs found despite a recent emphasis on improving Wikimedia and MediaWiki's pre-release test infrastructure, Hershberger appealed for help in writing unit and/or parser tests for key bug fixes to ensure that regressions are spotted more quickly in the future. If this is the case when 1.19 nears deployment (currently scheduled for late this year), the whole process would be likely to pass off "a *lot* smoother", wrote Hershberger (wikitech-l mailing list).

Improved https support comes to Wikimedia wikis

Also announced this week was the https switchover. Writing for the Wikimedia Foundation blog, operations engineer Ryan Lane said that the secure.wikimedia.org domain had been officially deprecated. Lane advised security-conscious visitors to simply change http to https in their URL instead to take advantage of new functionality which has taken months of planning to achieve (including the introduction of protocol-relative URLs). In addition to a noticeable speed improvement over its secure.wikimedia.org forerunner, Lane was clear on the benefits of the switchover to full https functionality:

The deployment of new https functionality was commended by many commentators; one wrote that "the lack of proper HTTPS" support had in his eyes been a long-standing issue with the site. At time of writing, support for secure browsing on the mobile site has not yet been enabled (bug #31333) – there are no plans to enable mobile support until the mobile and non-mobile sites are fully merged – but the core support for non-mobile devices outlined by Lane in his blogpost at least appears to be working correctly. HTTPS Everywhere, a popular browser add-on designed to make it easier to use the secure version of a website where it exists, is in the process of being updated to take advantage of the new format (wikitech-l mailing list).

In brief

Not all fixes may have gone live to WMF sites at the time of writing; some may not be scheduled to go live for many weeks.

How you can help
Call for developers

This week, Tomasz Finc, the WMF's Director of Mobile and Special Projects, issued a call for potential app developers for the Google Android operating system, whilst developer Jack Phoenix appealed for help with the Video extension.

  • Quick fix turnaround for serious bug: A bug in the JavaScript component of Wikimedia wikis that caused the browser windows of Internet Explorer 8 users to crash has now been fixed. The bug (#31424) was fixed only a day after it was filed, tracked down to a fault in the jQuery library version that MediaWiki 1.18 came packaged with (wikitech-l mailing list). The version of jQuery packaged with MediaWiki has since been upgraded to a version that does not cause the problem and Wikimedia wikis have been updated to reflect the change. In unrelated news, WMF Internalisation team member Gerard Meijssen blogged about bug #31310 as a paradigm for good volunteer involvement in making sure bugs are fixed quickly (it has since been re-opened pending further investigation).
  • Discussion over conversion to Git continues: The finer details of a plan floated late last month to convert the MediaWiki repository to Git began this week on the wikitech-l mailing list. Discussion so far has focussed on the most widely acknowledged point of contention, whether MediaWiki should remain as one repository or spin-off extensions and/or localisation files into their own repositories.
  • Wikimedia Blog mobile friendly: A special mobile edition of the combined Wikimedia blog was released this week, according to an announcement on the blog itself. Previously, mobile devices had been forced to display the full fixed-width layout that had been designed solely with larger screens in mind.
  • $3.6 million grant announced: As reported in more detail in this week's "News and Notes", the Stanton Foundation have announced that they have awarded Wikimedia a $3.6 million grant to support technical projects aimed at making its wikis more newcomer friendly, including work on enabling editing from the new mobile site and the long-awaited visual editor project.
  • Bot tasks open: Bot requests for approval are currently open on the English Wikipedia for helping automated Mediation Cabal pages and mass updating instances of a biological infobox. In addition, a BRFA focussed on article creation by Rich Farmbrough is also open.

    Reader comments
If articles have been updated, you may need to refresh the single-page edition.



       

The Signpost · written by many · served by Sinepost V0.9 · 🄯 CC-BY-SA 4.0