_02.jpg)
The community-written press release on Commons noted that "the 5 millionth uploaded file was attained on September 9, 2009, so the project has doubled in size in only eighteen months." The Wikimedia Foundation's press release mentioned planned technical improvements that should increase the growth rate further: The new upload tool (Signpost coverage: "Upload Wizard release expected shortly") and better support for video.
As part of the ongoing discussions about improving participation and newbie friendliness in Wikimedia communities (also topic of last week's IRC Office hour with WMF Executive Director Sue Gardner), specifically the WikiGuides project, the Foundation's Community Department tried to answer the question "How much do new editors actually improve Wikipedia?", by analyzing a random sample of "155 new registered users on the English Wikipedia who made at least one edit in mid-April of this year" and assessing the quality of their first edit on a five-level scale from "vandalism" to "excellent". They found that
| “ | most new editors made contributions worth retaining in some way, even if they weren’t perfect. More than half of these first edits needed no reworking to be acceptable based on current Wikipedia policy. Another 19% made good faith edits but needed additional help to meet standards defined in policy or guideline. | ” |
A comparison with an analogous sample from 2004 indicated that the ratio of vandalism has multiplied since then, while the ratio of acceptable or better quality edits has declined. However, the blog posting focuses on a positive aspect of the result: "The key thing to note in comparing the two samples is that the percent of acceptable edits made by newbies did not dramatically decrease from 2004 to 2011."
The German Wikimedia chapter has set aside €200,000 for a "community project budget", which will fund "the realization of ideas from the Wikimedia communities for the Wikimedia projects", according to the announcement in the "Kurier" (the Signpost's sister publication on the German Wikipedia). The selection of projects to fund, with a minimum budget of €5000, will be based on the recommendation of a seven-member budget committee, consisting of the chapter's treasurer, three members elected by chapter (Verein) members, and three members elected by the community.
The idea is a reaction to a blog posting by longtime Wikipedian Southpark. In January, during the run-up to a non-regular general meeting of the chapter that had been prompted by a motion of distrust against the chapter's board (signed by more than 60 members, based on criticism of the board's unilateral decision process while restructuring the chapter – Signpost coverage -, but ultimately unsuccessful), he observed:
| “ | The Problem: Wikimedia Deutschland is sitting on a whole lot of donation money. No one really knows what to do with it, and the opinions about spending it efficiently are diverging, to put it mildly. Particularly few money is flowing directly back into the projects. Part of the reason is that no one knows where to spend it there in a sensible way. | ” |
To solve this part of the problem, Southpark started an open call for ideas in his user space, which received a lot of suggestions, some of which might now get a chance of being realized.
Some months ago, the Wikimedia Foundation's grants process, funding mission-related projects from US$500 upwards, was expanded from chapters "in a limited way, to volunteers and like-minded organizations."
In September, the German chapter had announced a "contest of ideas" dedicated to promoting free knowledge, aiming to include non-Wikimedia projects (Signpost coverage). Eight winning projects (each funded with a sum between €500 and €5000) were announced in December, among them the purchase of high-resolution aerial images for 500 German cities for use by the OpenStreetMap community, and a "motivational video for Creative Commons", encouraging artists, authors and others who create content to release it under a free license.
Adam Brown, a political scientist from Brigham Young University has concluded that Wikipedia has become a reliable source of political information. He published a peer-reviewed study, titled "Wikipedia as a Data Source for Political Scientists: Accuracy and Completeness of Coverage" in Political Science and Politics 44:339–43, in which he reviewed "thousands of Wikipedia articles about candidates, elections, and office holders to assess both the accuracy and the thoroughness of Wikipedia’s coverage." He found "that Wikipedia is almost always accurate when a relevant article exists, but errors of omission are extremely frequent", following "a predictable pattern: coverage is best on topics that are more recent or prominent."
In the paper's introduction Brown observes that "studies of Wikipedia’s accuracy have generally found worries about its credibility to be overblown." He notes that most of them, including the famous 2005 Nature study, have used what he calls "the 'small-n, every-detail' approach": "The reviewers select a small number of seminal topics within a field and then check the accuracy of every statement in those articles". He argues that
| “ | [This] approach is flawed in both its 'small-n' and its 'every-detail' aspects. First, when reviewers create their sample, they inevitably choose those articles that deal with the most important issues in their respective fields—but because of their importance, these articles are likely to be the most read, most edited, and therefore most accurate articles in Wikipedia. Second, the “every-detail” approach tends to focus on minor rather than major inaccuracies. Reviewers adopting this approach check every word in their sampled articles for errors, no matter how inconsequential. | ” |
In his own study, he tried to avoid these biases with "a 'large-n, specific-fact' approach. I identify a specific fact that every article in a category ought to contain and then check every article’s accuracy on that fact." Specifically, he examined all 230 Wikipedia biographical articles about major party US governorship candidates who ran for office between 1998 and 2008 and "checked a specific fact: whether Wikipedia accurately characterized the candidate’s previous political experience. I found no errors in these articles at all." Separately, he examined all yearly articles about United States gubernatorial elections back to 1976 to see whether the given results of the major political parties' candidates agreed with the official results, finding that "In only four (2.6%) gubernatorial elections was Wikipedia’s margin off by more than one percentage point."
While acknowledging that Wikipedia is not a quotable source for academic publications, the study concludes by cautiously suggesting that "for political scientists with limited time and research assistance, Wikipedia may be just accurate enough to permit its use in preliminary work."
The study was covered by The Salt Lake Tribune ("BYU scholar: Wikipedia’s political content is reliable"), UPI Science News and Trebuchet magazine.
Recently, several US law blogs noted a February court decision where the judge admonished one party for plagiarizing from the Wikipedia article Strickland v. Washington: "The court notes here that defense counsel appears to have cobbled much of his statement of [a law relevant to the case] by cutting and pasting, without citation, from the Wikipedia web site. Compare Supplemental to Motion for New Trial (DN 199) at 18–19 with http://en.wikipedia.org/wiki/Strickland_v._Washington (last visited Feb. 9, 2011). The court reminds counsel that such cutting and pasting, without attribution, is plagiarism. The court also brings to counsel’s attention Rule 8.4 of the Kentucky Rules of Professional Conduct, which states that it is professional misconduct for an attorney to “engage in conduct involving dishonesty, fraud, deceit or misrepresentation.” ... Finally, the court reminds counsel that Wikipedia is not an acceptable source of legal authority in the United States District Courts." On the "Minnesota Lawyer" blog, an attorney (coincidentally named Michael Goodwin) commented that "Issues of plagiarism aside, this lawyer isn’t the first to use Wikipedia in a legal document. There probably are appropriate uses of Wikipedia in legal writing, and in fact, there is a Wikipedia article about courts that cite to Wikipedia. Most courts that do so use Wikipedia for illustrative purposes, such as this recent copyright case that cites to Wikipedia’s entry about fantasy football as a means of illuminating the context out of which the case arises. Using Wikipedia to establish anything of significance, however, can be troublesome ..."
An academic paper titled "From Castalia to Wikipedia: openness and closure in knowledge communities" explores parallels and differences between Wikipedia and the fictional academic world depicted in the 1943 novel The Glass Bead Game (which earned its author Hermann Hesse a Nobel Prize in literature), regarding "knowledge, decision-making and social organization". (E-Learning and Digital Media, Volume 8 Number 1 2011, paywalled, abstract; by Peter Roberts and Michael A. Peters, professors of education at the University of Canterbury in New Zealand and the University of Illinois at Urbana-Champaign/US, respectively.)
The authors assert that "of the novels that might be considered when addressing the idea of openness, particularly as this applies to knowledge communities, none is more helpful" than The Glass Bead Game. Hesse's book is set in the twenty-third century and centers around an intellectual system (whose "rules" are never laid out in detail) that evolved from a literal game involving actual glass beads into a kind of universal language encompassing and connecting all of the arts, sciences and religion and "emerged as the supreme form of cultural engagement. Castalia, a dedicated knowledge community, grew from the ruins of the twentieth century and became the home of the Game", as the paper says in its first part, a summary of the novel. It continues with a one page overview of Wikipedia, mentioning the Wikimedia Foundation and the Five Pillars.
The comparative analysis in the third part starts with differences: "Castalia is premised on the principle of intellectual and cultural elitism, with a privileged community of scholars and only the very best making it through to the most advanced schools; Wikipedia, on the other hand, is built on the notion of popular participation ... Far from being separated from the rest of the world, Wikipedia is the world. People who read and edit Wikipedia come from all corners of the earth", but that there still exist "definite patterns of participation, with a small group of dedicated contributors dominating the editing process" and moreover that there is "a clear hierarchy in Wikipedia, just as there is in Castalia, although in the case of the former this is more fluid and less formal." As the second common property, the authors identify the fact that "Neither Castalia nor Wikipedia focuses on the creation of new knowledge. Castalians assume that the ‘manuals, pedals and stops’ of the Glass Bead Game are now fixed, with nothing further to add to the vast stock of human knowledge on which exponents of the Game play", while Wikipedia has its Verifiability and No Original Research policies. "The ambitions that underpin Wikipedia as an enterprise bear a resemblance to the conceptual architecture of the Glass Bead Game. The Glass Bead is, the narrator informs us, capable of reproducing the entire intellectual content of the universe. The aim of Wikipedia [actually, the Wikimedia Foundation] is no less than ‘a world in which every single human being can freely share in the sum of all knowledge’. The authors also find common traits in the attitudes towards individuality and collectivism: "In both Castalia and Wikipedia there is, as an ideal, a sense of participating in a process that transcends the thoughts, feelings and ideas of any individual – a commitment to working with knowledge in a certain way. Unlike the academic world, with its ‘star’ researchers, in both Castalia and Wikipedia the rewards are meant to be more intrinsic."
Returning to differences, it is noted that in the centuries-old Castalia, changes are occuring slowly, and it is "self-consciously conservative in its social organisation", while Wikipedia is characterized by rapid reactions to events, and promotes boldness. Both try to ensure quality by self-regulation, but in different ways – in Castalia "through a process of elite selection (with only the most able students making it to the top schools in the pedagogical province)", in Wikipedia by "a more populist form of collective monitoring and adjustment". Finally, the authors claim that Hesse's book, through its main character Josef Knecht, "problematises universalist constructs of truth", while Wikipedia "is explicitly wedded to an ideal of truth emerging through consensus. There is a faint echo, in the Wikipedia pages, of the Habermasian notion of dialogue through consensual communication between rational subjects".
This week, we take our first in-depth look at the WikiProject Council. Started in August 2006 by Kirill Lokshin, the Council is a group of Wikipedians gathered to encourage and assist with the development of active and new WikiProjects, and to act as a central point for inter-WikiProject discussion and collaboration. The project does not maintain a formal list of members, but there are 128 editors with the project's userbox. The Signpost interviewed project members John Carter, WhatamIdoing and Kirill Lokshin.
John is an administrator, and has been on Wikipedia since January 2007. He describes himself as "a reviewer-type editor who helped establish a lot of the existing WikiProjects". WhatamIdoing has been a Wikipedian since March 2007, and joined the project because of her interest in clarifying the rights of WikiProjects, "Who decides the group's scope? Who decides which articles to tag? Who gets to write advice pages?" Kirill is an engineer who designs and builds ground systems for satellites and spacecrafts. He has been on Wikipedia since June 2005, and is also an administrator, as well as a current member of the Arbitration Committee. During 2005–07, he was particularly involved with WikiProject Military history (MILHIST), and was one of the main players in the consolidation of different projects in that topic area into a single entity, later serving as the lead coordinator of the amalgamated project for a number of years. Kirill said, "The idea of the WikiProject Council came about primarily because of a number of editors who approached myself and the other MILHIST coordinators for advice on setting up or reviving other WikiProjects; the number of such requests was sufficient to suggest that creating a central place to share ideas and recommendations would be beneficial. The initial concept of the Council was somewhat more formal than the current one. One of the models originally considered was that of an assembly composed of elected representatives from different projects; but the key objectives of documenting best practices and encouraging inter-project dialogue and collaboration were identified at the outset, and have continued to drive the direction of the Council in the time since."
Many Wikipedians have come across the WikiProject Council, but few know what it actually does. So, what are the most important functions of the project? According to Kirill, the project has three major functions. The first is the maintenance of the WikiProject Guide: "The initial drafts of the WikiProject Guide were written in mid-2006... Since then, the guide has seen extensive revision and updating. Some sections are, admittedly, rather dated at this point and a number of more recent approaches to WikiProject infrastructure and organization are not described, but the guide remains a useful resource overall." The second is maintenance of the WikiProject directory and the WikiProject proposal process. "A third function, which has always been an objective, but which has generally seen little activity, is serving as a central discussion point for matters concerning WikiProjects in general. While many of the editors most active in WikiProject maintenance do monitor the Council's discussion pages, many do not; and so there has traditionally been a tendency for discussions on WikiProject matters to be split among a number of venues (such as the Council, the various Village Pumps, and a variety of other pages), without participants at any particular venue necessarily following, or even being aware of the others," added Kirill.
We asked what their advice is for someone wanting to propose a new WikiProject or Task force, and the main reasons proposals fail. John's advice is to check whether an active or inactive project which deals with the subject of the proposal already exists: "Many projects, particularly those related to pop culture in some form, become inactive when the artist, TV show, movies, etc., become inactive. Also, I think it makes a great deal of sense to in most cases check to see if there is an existing project with which the new proposal can work some sort of subproject status. Regarding why projects fail, that can be for any number of reasons. Sometimes, the proposal seems to be promoting POV of some sort – if that is the case, the project will likely be eventually merged with something else so our existing POV guidelines are followed. Other times, it may simply be that the subject is not one which is so interesting to others that they feel they are likely to devote much time to it. And, of course, if there aren't many materials available on the subject, however much interest there might be, there might not be sources enough to develop content." For Kirill, "the first, and most important piece of advice, is to be quite sure that you're willing to put in the time and effort in making a project successful before creating one. While WikiProjects might have many members, their infrastructure and processes are typically maintained by a much smaller group, even in the largest projects; and newer or smaller projects are particularly reliant on a small group or even a single editor, to keep the project "running" during periods of lower activity. The behind-the-scenes work involved is typically neither particularly exciting nor particularly enjoyable; but it must be done if a project is to function as an entity in its own right, and many projects fail because no individual member is able or willing to take responsibility for doing it."
WhatamIdoing believes that most people forget that a WikiProject is a group of people – not a subject area, a project page or a talk page banner: "There's a sort of 'build it and they will come' mentality. If you just make a pretty enough project page, then surely a dozen editors will show up and collaborate with you. It actually works the other way around. If you've got a dozen editors working with you, then it's probably time to create a WikiProject page. If you're on your own, creating a project page is likely to be isolating, discouraging, and ineffective. Almost all "WikiProjects" with only a couple of members go {{Inactive}}, often within a few months. You need a critical mass of people to make a project successful. Proposals by inexperienced editors are highly likely to die—and the English Wikipedia usually loses those enthusiastic editors entirely in the process."
What are the biggest challenges for the project? John says that keeping track of all the projects that get created is a big challenge, "Considering several are made without consultation from anyone, that is a bigger problem than may be obvious. Another situation we face is trying to bring editors who are affiliated with individual projects to cooperate with other projects. Kirill says that while the Council has some level of visibility into the creation of projects that go through the proposal process, there is no requirement that projects be formally proposed before creation, and many new projects bypass the proposal process entirely. "With deletions, the situation is even more challenging; as a general rule, the Council has no visibility into deletion nominations beyond individual participants who might want to keep an eye on MFD. In addition, because many of the projects nominated for deletion are inactive, there is nobody with any real reason to inform anyone about the nomination," he added.
Does the Council collaborate with other WikiProjects? John is keen to see more of this, "We would like to help encourage interaction among editors from related projects to help develop content which may be only peripherally related to their topic, and the proposed 2011 religion-philosophy meeting is one way we are trying to help encourage such interaction. Results, sadly, aren't in yet on how effective they are. Kirill says, "There has always been a vision, albeit a rather vague one, of the Council serving as a central forum for discussions that involve multiple WikiProjects. To date, we've had only limited success in actually bringing this vision to life. In large part, this is because the Council is a lower-traffic discussion area than most of the projects who might otherwise be motivated to participate in such discussions; most projects will therefore prefer to hold discussions "in-house" to attract more participation. We are, however, continuing to explore ways in which we can promote increased dialogue and collaboration."
Finally, we wanted to know what the most pressing needs for the WikiProject Council are, and how a new contributor can help. John says that keeping track of all the extant groups is a priority, "Once that is done, it will help a lot to encourage and promote inter-project collaboration. In general, maybe one of the easiest ways to help it achieve some of its purposes is help promote the interaction of extant projects and task forces, and helping make the pages of such groups more useful to new editors interested in working on related content." Kirill sees the most pressing need as the area of directory maintenance, "...updating the WikiProject directory to reflect new projects, and changes in existing ones, is something that requires little experience, and can be readily done by new participants. The directory underlies much of what the Council does and might wish to do in the future. Without a clear picture of what WikiProjects exist, it's difficult to organize anything involving them, so this task is not an unimportant one. Beyond that, the Council discussed a number of ideas for improvement earlier this year. Many of them have yet to be implemented, and assistance with any of them would certainly be appreciated." WhatamIdoing says that anyone who feels responsible for any WikiProject would likely benefit from watching the Council's pages, "Seeing the "mistakes" that other groups make can help you avoid them in your own group. Make a particular effort to respond to messages that are posted. Tell your fellow members what you're doing and pass along any news that might interest them."
Next week, we visit the home of the Stanton Drew stone circles and the Glastonbury Festival. Until then, do take a stroll through our archives.
The Signpost welcomes Bahamut0013 (nom) as our newest admin. Bob has been with the US Marine Corps since 2003 and with Wikipedia since 2006. He contributes mainly to articles that deal with the Marine Corps, military life and equipment, and video games, movies, anime, and TV shows. He is helping to construct the battleship portal with other members of Operation Majestic Titan. He has just been appointed to the Audit Subcommittee.
At the time of publication there is one live RfA: RHM22, due to finish Tuesday.
.jpg)
Eighteen articles were promoted to featured status:
One list was promoted:
One topic was promoted:
Five images were promoted. Medium-sized images can be viewed by clicking on "nom":
The Arbitration Committee opened no new cases but closed one case. One case is currently open.
During the week, several arbitrators submitted proposals, some of which will form part of the final decision. Other proposals are likely to be submitted in the coming week.
This case was opened after a number of divisive noticeboard discussions involving allegations of misrepresentation of sources, as well as tendentious and antisemitic editing. 19 editors submitted on-wiki evidence, and several users submitted proposals in the workshop. Drafter Newyorkbrad also submitted a full proposed decision in the workshop which attracted input from arbitrators, parties and others. A total of 14 arbitrators voted in the case before the case came to a close today.
On Tuesday (12 April), MediaWiki version 1.16.3 was officially released to external sites (Wikimedia Techblog). It included a group of three security fixes that had already gone live to Wikimedia sites, which are running a pre-release version of 1.17:
| “ |
|
” |
After the release, however, it soon became clear that the first of the three issues had not been entirely cleared up, prompting the second release of the week, MediaWiki 1.16.4, on Friday (15 April) (Wikimedia Techblog, wikitech-l mailing list). The updates also took advantage of recent localisation efforts in order to provide users with an interface translated into their own language.
In related news, no official date has yet been set for a release candidate of MediaWiki 1.17, the version WMF wikis are currently running. A beta version is expected "probably next week", however, according to developer Tim Starling, who is overseeing the release effort (also wikitech-l). The accompanying discussion also included calls to "branch" version 1.18 within the next fortnight. Branching would separate a snapshot of the software from the developmental bleeding edge version of the MediaWiki software (also known as "trunk"), allowing it to be stabilised, tested and released in the next few months.
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.
_02.jpg#filehistory){kind=link}
