OK, I'm up next at AoIR 2009, as part of a blog concept mapping double-header with my brilliant PhD student Tim Highfield. Here's the Powerpoint - hope the audio recording works out, too... and the audio is attached now, too.
The next speakers in the blogging session at AoIR 2009 are Muhammad Abdul-Mageed and Priscilla Ringrose, whose focus is on war blogging. Such blogging addresses the exceptional communication demands during war situations, and war bloggers in warzones can meet these needs speedily and with authority. This also reflects a continuing shift in the media overall. The focus of this paper is on the 2006 war between Israel and Lebanon, where western media profiled (English-language) Israeli and Lebanese blogs.
So, the bloggers here belonged to two oppossing, warring nations,and espoused different ideological positions; how were they chosen and what positions do they reflect? What demographics, structural features, thematic, regional, and political positioning do they exhibit? According to which parameters were they selected? The study analysed all posts from 40 blogs (20 Israeli, 20 Lebanese) during the 34-day war in June and August 2006, which were found using search engines, media outlets, and blogs. Blogs had to be based in Lebanon or Israel, had to have at least five posts during the 34 days, had to be in English, had to have at least one hit in the global media, and had to be single- or group-authored rather than blog fora.
The final speaker in this first session at AoIR 2009 is Carol Soon, who shifts our focus to Singaporean political bloggers. Political blogging and related forms challenge conventional top-down communication flows, of course, and in doing so also undermines established entities' authority in information dissemination. What follows is a diversification of political participation in the networked public sphere - and in the Singaporean context, then, who are the key players here?
The networked public sphere can be seen as an autopoietic system,in which flows of communication and relationships are self-organising, move from the bottom up, freely within clusters and in a self-determined fashion. This challenges systems which traditionally hold more powerful positions - and hyperlink analysis can be utilised to examine the flows of information in this changing environment. Such flows may involve conventional political parties, but also civil society groups (which in Singapore particularly challenges the established system).
This paper provides an update on an ongoing research project which maps and investigates the Australian political blogosphere, and expands on work presented at IR9.0 in Copenhagen (Bruns et al. 2008). The project is situated in a growing tradition of quantitative and mixed-method research into the shape and structure of national and international blogospheres (cf. e.g. Adamic & Glance, 2005; Kelly & Etling, 2008; and a number of the studies collected in Russell & Echchaibi, 2009), which utilise a combination of link crawling, data scraping, and network visualisation tools to map interconnections between blogs and analyse their contents. However, our work also addresses some of the limitations of these studies.
This paper examines the observable patterns of content creation by Australian political bloggers during the 2007 election and its aftermath, thereby providing insight into the level and nature of activity in the Australian political blogosphere during that time. The performance indicators which are identified through this process enable us to target for further indepth research, to be reported in subsequent papers, those individual blogs and blog clusters showing especially high or unusual activity as compared to the overall baseline. This research forms the first stage in a larger project to investigate the shape and internal dynamics of the Australian political blogosphere. In this first stage, we tracked the activities of some 230 political blogs and related Websites in Australia from 2 November 2007 (the final month of the federal election campaign, with the election itself taking place on 24 November) to 24 January 2008. We harvested more than 65,000 articles for this study.
Not only are social media a major online phenomenon: they are also producing a vast amount of data and metadata about cultural practices, most of which are shared openly and deliberately - blogging, social bookmarking, social networking, and other practices would be impossible to imagine without RSS feeds, open APIs, and other sources of detailed and up-to-date information about what users are doing. This provides researchers with significant new opportunities to track, analyse, and interpret online cultural practices on an unprecedented scale, and virtually in real time: we can see Twitter traffic spike in response to major events, we can track the viral distribution of YouTube videos, we can map the social graphs of the blogosphere, etc. At present, in fact, we are in the unusual position of having more research-ready data available than we have research questions to ask of these data - and we are only developing the tools and methodologies to engage with this resource. This seminar will outline some of the opportunities, and point to the methodological and interpretive challenges we face in confronting them.
The third speaker in this session at ANZCA 2009 is Tim Highfield, who works on a comparative study of political blogs in Australia and France (and is one of my PhD students). He focusses here on the Australian side and its reaction to the inauguration of Barack Obama. The project tracks some 245 blogs and news Websites in Australia, and extracts from these each post (and its links) as they become available online. These data are then quantitatively analysed for keyword and link patterns.
The Obama inauguration was a major political event, of course, and provided a useful case study for this work; other such samples could be the swine flu epidemic or the 'utegate' controversy storm in a teacup. Interestingly, only about 50 blogs in the population published a post or more during the two weeks surrounding the inauguration (possibly due to the fact that January is a major holiday month in Australia). There was no major spike on inauguration day itself, either.
I'm the first speaker of the next session of ANZCA 2009, presenting some baseline data from our first test run of our blog mapping methodology during the Australian federal election in November 2007. The Powerpoint is below (with audio to follow soonish also online now), and the full paper is online as well. Links to more information are in the final slide of the Powerpoint.
After the rather unruly cultural panel, WebSci '09 has now moved on to the next keynote, by Noshir Contractor. His theme is the application of Web science to social networks, and he begins by noting some of the experimental mobile tools now available for social networking. The Web in general enables us to communicate and collaborate with any one at any time, but what is necessary are tools that enable us to identify who it is that we should be or want to be collaborating with. This is where social network analysis and Web science comes in.
The second and last day of WebSci '09 starts with a session on social networking, although the first paper in this session, by Yorgos Amanatidis, has the somewhat technical title "Kernels for Complex Networks" - we'll see what that's all about... Visual network graph models, apparently, for graphs which represent relational data in an abstract way. Such graophs can be used in the analysis, simulation, and prediction of network topologies, focussing especially on aspects like scaling, clustering, and node centrality.
What can be observed in real networks is the degree distribution: as the Web grows, the average degree is constant, but there is huge variance and no concentration around the average; indeed, we see the small world phenomenon which produces networks with small diameter and strong clustering tendencies (the friend of my friend is likely also to be my friend). Kleinberg, for example, modelled the fact that in small world networks there are not only short paths between nodes, but that these nodes can find such paths effectively using local information.
