And we're on to the last session of the day, which continues the blogging theme. I should probably note that not all papers at this conference are concerned with blogs - there are some eight or nine sessions running simultaneously here, themed around many other topics; as always, the topics I cover here are only the ones I was most interested in and do not necessarily represent the overall intellectual thrust of the conference accurately. (I'm sure the other bloggers here will present a very different version of AoIR 2004 in their blogs, even if admittedly many of them have also attended most of the blogging-related sessions…)
Alex Halavais is the first speaker in this session; he's interested in macro hyperlink analysis to make assumptions about the micro nature of links between pages. Hyperlinks possibly imply higher levels of similarity, interest, reputation, and computer networks also serve as social networks or friends linking friends. But, how are hyperlinks used, and what do they mean? Alex offers an idealised view of neighbourhoods interconnected by affinity, clustered and webbed, and communities of interest are defined by hypertext. The presence of celebrities amongst bloggers and the nonconnectedness of people along the edges speaks against this view, however - many people aren't linked to at all, and the 'blog ecosystem' study data supports this as well.
A Graph of Connections between Scholarly Blogs
Alex studied the scholar-blog community, which he calls 'chunky'; half of these link to no others in the list (we're talking about links on the main index page, not throughout the blog, by the way), and there are some interesting small- and large-scale chunks. The distribution is still very logarithmical - with a high concentration of links in a small number of blogs. Alex also suggests a typology of blog linking: fans (relatively few inbound links, many outbound links); stars (relatively few outbound links, many inbound links), and friends (many reciprocal links); perhaps also bridges (relatively even number of inbound and outbound, but few of the reciprocal).
Links are used differently: are reciprocal links of a certain class? Can we detect something of a page-text relationship which differs in reciprocal links? Alex is coding links into different categories: identification links, sourcing links, masthead links, blogroll, and reciproll (and trackbacks).
There are also important qualitative issues here: not all blogs are the same; in fact there are very many different varieties of blogs which make overall comparisons difficult if not impossible. Girilinks (giri is Japanese for 'feeling like you have to do something') skew the data since some bloggers who are linked to feel obliged to link back even though they have no real interest in doing so. There are different types of linking (news, people as links, etc.). And there are link references which are done as a kind of shorthand, automating the process of explanation by linking to someone else's explanation (Alex calls this 'standing on the shoulders of midgets').
Kaye Trammell is next, speaking of blogging from a pedagogical perspective (co-authored with Richard Ferdig; Kaye also has a list of useful blog research on Blog Research). She follows Lev Vygotsky's social constructivist approach to learning, moving beyond a recitation script towards a zone of proximal development (suggesting the linkage of the zone of everything the student can learn on their own, and the zone of everything the student can learn with the help of a teacher). Blogging can work well in this context: teachers could assess the blog goal and develop their project, and implement either a laissez-faire or a boot camp approach. Laissez-faire would involve relatively little instruction and allow students to explore blogging on their own, while the boot camp approach would set some very strict guidelines for how blogs are to be used (and would therefore involve more instruction in this).
The benefits of blogging might be that it creates subject matter experts amongst students, increases their interest in learning, and shares assignments with the outside world (which adds to students' self-policing - they're less likely to produce poor work if they know it might be seen widely). Practical suggestions for implementing blogs are to publicise this effort in the blog community, to create ground rules for doing this, to explain the potential reach of online assignments (e.g. to prospective employers), and to discuss the ethics of blogging (potentially, incidences of plagiarism might go down in a blogging context). Practical concerns include the use of technology (hosting, services, etc.), the question of content (what you ask for isn't necessarily what you get), and the question of student security as they are becoming more visible.