WikiSym features research on Wikipedia

WikiSym (short for the International Symposium on Wikis) is an annual international conference about wikis and wiki technology. This year's conference was the fifth annual WikiSym, and was held in Orlando, Florida on Oct. 25-27. It was co-located with OOPSLA, the major ACM conference on object-oriented programming.

Compared with Wikimania, the annual Wikimedia Foundation conference, the conference has more of an academic focus, with papers getting published in the ACM digital library, and is broader in scope, with papers about all aspects of wikis. It is also much smaller than Wikimania, with three main tracks: presented papers, workshops/tutorials, and open space.

The opening conference keynote was given by Fernanda Viégas and Martin Wattenberg, and was called "Visualizing the Inner Lives of Texts" (pdf). In their talk, Viégas and Wattenberg discussed various tools for visualizing the history and content of texts (including wiki articles) that they have developed, including History Flow, Chromogram, and the Many Eyes project.

There were many other papers and posters about researching Wikipedia in the program as well. These included work on the slowing growth of Wikipedia, the formation and roles of groups and WikiProjects, the lifecycles of articles, searching Wikipedia, user interface extensions, bots and assisted editing tools, as well as various ways of measuring quality, credibility, collaboration, and conflict.

Papers that featured research based on Wikipedia included:

• "Bipartite Networks of Wikipedia's Articles and Authors: a Meso-level Approach" (pdf) by Rut Jesus, Martin Schwartz, and Sune Lehmann, about how networks of related articles and editors behave;
• "Herding the Cats: The Influence of Groups in Coordinating Peer Production?" (pdf) by Aniket Kittur, Bryan Pendleton, and Robert E. Kraut, about editor behavior before and after joining WikiProjects on Wikipedia, and the impact of WikiProjects (the researchers found that edits to WikiProject pages account for around 1% of all Wikipedia edits total);
• "The Singularity is Not Near: Slowing Growth of Wikipedia(pdf) by Bongwon Suh, Gregorio Convertino, Ed H. Chi, and Peter Pirolli, about Wikipedia editing activity by various levels of editors (by edit count), the slowing growth of mid-level editors, and the growing number of new edits that get reverted;
• "A Jury of Your Peers: Quality, Experience and Ownership in Wikipedia" (pdf) by Aaron Halfaker, Aniket Kittur, Robert Kraut, and John Riedl;
• "Assessing the Quality of Wikipedia Articles with Lifecycle Based Metrics" (pdf) by Thomas Wohner and Ralf Peters;
• "Organizing the Vision for Web 2.0: A Study of the Evolution of the Concept in Wikipedia" (pdf) by Arnaud Gorgeon and E. Burton Swanson, which looked at the evolution of a single Wikipedia article, Web 2.0, over time.

The best paper award went to Michael D. Ekstrand and John T. Riedl at the University of Minnesota for their paper "rv you’re dumb: Identifying Discarded Work in Wiki Article History" (pdf), which provides a new way of visualizing an article's history and revision flow, as well as whether diffs between revisions discard previous work, with an overlay over the current MediaWiki history page display.

Workshops and technical demos were also held, including a demonstration called "ProveIt: A New Tool for Supporting Citation in MediaWiki" (pdf) by Kurt Luther, Matthew Flaschen, Andrea Forte, Christopher Jordan, and Amy Bruckman.

Other papers presented at the conference focussed on various aspects of wikis, including wiki search, mapping the universe of wikis, and more. Workshops included topics such as "Wikis for software engineering." Tom Malone of the MIT Sloan school gave the conference keynote.

The conference also made use of Open Space Technology to hold many ad-hoc sessions on a wide range of topics. Open space sessions included discussions on how academics can better research Wikipedia, parsers and search functions, a possible Wikimedia Commons for references, how Wikipedia compares to other large Internet websites, and the future of wikis.

Finally, Brion Vibber gave the closing conference keynote on "Community Performance Optimization: Making Your People Run as Smoothly as Your Site" (pdf).

The next WikiSym will be co-located with Wikimania 2010 in Gdańsk, Poland.

• 2009 WikiSym conference website
• Conference Proceedings

In this issue

2 November 2009 (all comments)

Article contest

Conference report

Election report

Audit Subcommittee

Dispatches

News and notes

Discussion report

WikiProject report

Features and admins

Arbitration report

Technology report

+ Add a comment

Discuss this story

These comments are automatically transcluded from this article's talk page. To follow comments, add the page to your watchlist. If your comment has not appeared here, you can try purging the cache.

• The fact that editing of Wikipedia is slowing is not surprising. It takes more expertise / dedication to write greater detail than to provide a brief overview. As the paper states "the low fruit have been picked".Doc James (talk · contribs · email) 18:10, 3 November 2009 (UTC)[reply]

• Right - Aren't we entering a phase where people now are working on improving existing articles rather than starting articles on less and less notable topics? Can anyone do an overview on what all the research over the past year or two has shown re: quality? -- Ssilvers (talk) 20:41, 3 November 2009 (UTC)[reply]

• I'll offer another theory, partly because I think it has a lot of validity, and partly to demonstrate that it may be very difficult to ever figure out why edits (and editors) peaked in early 2007 and have declined significantly since then. My theory is that since editing Wikipedia competes with other things that people can do on the Web, participation here has been impacted by the ever-increasing number of interesting things to do. One obvious example is the rise of social media such as Facebook. As importantly, all the other user-content-generated websites (Flickr, YouTube, etc.) have an obvious incentive to make it as easy as possible to participate (the more the participants, the more the audience, and thus more potential revenues), while Wikipedia does not. So, for example, in the 2007-2009 period, prior to usability initiative, almost nothing was done to improve ease of editing; in fact, as infobox and template usage continued to expand, the initial user editing experience almost certainly became more difficult. -- John Broughton (♫♫) 20:20, 4 November 2009 (UTC)[reply]

• A very interesting theory, which takes as a premise that people edit Wikipedia because they are bored. Now certainly our top editors get more out of it, but boredom (or escape from real life work) is certainly a gateway to editing.HereToHelp ^{(talk to me)} 18:16, 7 November 2009 (UTC)[reply]

• One fact seems to have been missed by students of Wikipedia so far, & I confess I only thought of it a few days ago: if we compare the numbers of Wikipedia users against editors, the share of users who became editors has been shrinking for some time -- even more dramatically than any other yardstick. (Even if the number of editors remain steady or increased slightly, the share obviously must have shrunk as our Alexa rank increased.) However, the actual curve remains to be determined, & whether the inflection points on this curve have any value. -- llywrch (talk) 19:18, 3 November 2009 (UTC)[reply]