Ray Nigel Title: Becoming Wikipedian: Transformation of Participation in a Collaborative Online Encyclopedia Authors: Bryant, Forte, and Bruckman This paper is about users of Wikipedia. Wikipedia is an open encyclopedia available to the public. Any user is able to read and extend the encyclopedia. In this paper, experienced editors are interviewed in an attempt to learn about the transition from a novice to an expert Wikipedian. This paper also focuses on applying activity theory (AT) and legitimate peripheral participation (LPP) to Wikipedia. In this sense, Wikipedia is treated as a social collaboration and users are peripheral observers before they are accepted into the group. At the heart of this study, is a survey of nine active Wikipedia editors. This paper did a good job of describing the process by which new Wikipedians become advanced users. I believe applying the legitimate peripheral participation model to Wikipedia is a novel idea. The apprentice butcher analogy seemed out of place though. The barrier to entry into a group may discourage new participants, but this does not seem to occur in Wikipedia. The paper shows how new users are slowly assimilated by the core active Wikipedia participants. Registering, discussion boards, and talk pages all help new users evolve from a peripheral user to a full-blown active Wikipedian. I believe the activity theory model ties in well with how editors of Wikipedia evolve their internal model of the structure of Wikipedia. The anecdotal evidence of the participants in the study shows a clear change over time in how the users edit Wikipedia entries. This is evidenced by their more sophisticated use of tools and a change in what kinds of contributions they perform. Initially, the participants in the study would make slight changes only to articles which they happen to be reading and which they consider themselves to have advanced knowledge. As their confidence increased, the Wikipedians would expand their edits. Eventually, they learned more about the organization of Wikipedia and some of the underlying rules and conventions. This shift coincides with the active theory model. Initially, the users had a very narrow and skewed view of the object, tools, rules, and community. As they experience more of Wikipedia, their model of Wikipedia as a social structure also shifted. I have a couple of concerns with how this study was performed. Only nine people were interviewed. This seems like a small number of people. These people all had varying degrees of experience with Wikipedia, but all of them were considered advanced editors. This could give them a skewed perspective of the path to joining the Wikipedia community. The only people surveyed, were people who were able to figure out how Wikipedia works and how to be a valuable participant in the community. If joining Wikipedia is like becoming a butcher, with difficult hurdles to overcome, this study would not discover this fact. Only successes are interviewed. If people try and fail to join the Wikipedia community, they are excluded from this study. My second concern involves the length of experience one of the participants had. Participant 9 has been active for only 2 months, yet s/he has performed 3,664 edits in November. I find it unusual that Participant 9 was able to completely figure out Wikipedias internal workings in such a short period of time. According to my understanding of the table, Participant 9 averaged over 120 edits a day in the month of November. I am not sure how someone is able to make that many edits. Three of the participants each made over 10,000 edits in November. That is an enormous amount of editing work being done. I would like to know how much time these users spent doing this work.
Nigel Ray Title: The Dynamics of Mass Interaction Authors: Whittaker, Terveen, Hill, and Cherney This paper seeks to learn about mass interaction in Usenet forums and apply the common ground model to these social interactions. Three characteristics are used in this paper: size of newsgroup, conversation strategies, and threading. They wish to test several hypotheses by fitting these three characteristics into the common ground model. From this, they wish to extrapolate how well the common ground model applies to mass interactions. Great care goes into the methodologies of this paper. The paper goes into depth of how they randomly chose the 500 newsgroups to study. An explanation was given for each newsgroup excluded from the study. The paper also explains how they objectively defined familiarity, interactivity, and conversation strategies. Reasoning is also provided for their characterizations. Once major concern I have with this paper is why the average threading depth of a newsgroup corresponds to interactivity. This idea makes sense in face-to-face conversations, but I do not feel it is a reasonable measure newsgroups. This idea may make sense in some social newsgroups, but fails to make sense when users are asking simple technical questions. If a user asks a specific question and a user provides the answer as a reply then post, the conversation is usually done. This paper would classify this is low interaction. I wonder if a more applicable measure would be average response time (i.e. the amount of time to receive a reply to a post). Another point, brought up at the end of the paper, is using the above measure for interactivity in the presence of flame wars. In a flame war a post may receive many threads, which would result in a large interactivity. Since some newsgroups are prone to more flame wars then others, there is a bias in concluding more belligerent newsgroups have more interactivity. One result of the papers research indicated that larger newsgroups tended to have smaller message sizes. This same conclusion was reached in Cherneys MUD work from last week. It makes sense that people more familiar with a newsgroup are less likely to cross-post. Cross-posting seems like something a person would do when they wish to discuss a topic, but do not know the proper forum. Since the papers results showed a correlation between familiarity and decreased cross-posts, I believe this goes more to show that the methodology for deciding familiarity was accurate. Many of the hypotheses involving interactivity were invalidated. The paper concludes that the common ground model fared poorly in predicting interactivity. They believe this has to do with the weak ties found in Usenet. I wonder if the methodology for determining the interactivity of a newsgroup is inaccurate. Common ground could provide an adequate model if a different measure of interactivity is used. I believe further research is required to prove the weak link between the common ground model and interactivity.
Nigel Ray Title: Studying Cooperation and Conflict between Authors with history flow Visualizations Authors: Viegas, Wattenberg, and Dave This paper is about a visualization tool called history flow. This visualization is meant to visualize Wikipedia articles over time. In Wikipedia, many users are allowed to edit a page and all of the revisions are tracked. The paper cites a lack of visualization for the progression of an article. The only view before history flow is a text based listing of who did what edits and when. After describing their program, the paper describes a few patterns they found. The paper describes how visualizing Wikipedia edits could help shed light on usage patterns of vandals. If a proper pattern can be gleaned, then automated systems could be created to discover vandalized pages. I found this aspect of the research to be particularly useful. Perhaps if a visualization was created centered around this idea, then more patterns could be discovered. The visualization itself is simple and expressive. Once the viewer understands how to interpret the visualization, it conveys the flow of the Wikipedia article. I like that the more colorful an articles visualization is, the more contributors have edited its content. My only concern with this project is that the visualization is not intuitive. If the visualization is only meant for internal resea...