AoIR 2010

Mapping Australian User-Created Content: Methodological, Technological and Ethical Challenges

Axel Bruns, Jean Burgess, Thomas Nicolai, and Lars Kirchhoff

• 22 Oct. 2010 – Association of Internet Researchers conference, Gothenburg

This paper reports on a three-year (2010-12) research program that has developed new methodologies for mapping the Australian blogosphere (Bruns et al. 2008a/b/c, 2009a/b). We improve on conventional Web crawling methodologies in a number of significant ways: first, we track blogging activity as it occurs, by scraping new blog posts when such posts are announced through RSS feeds, rather than by crawling existing content in the blogosphere after the fact. Second, we utilise custom-made tools that distinguish between the different types of content and thus allow us to analyse only the salient discursive content provided by bloggers, without contaminating our data with static links and ancillary material. Finally, we are able to examine these better-quality data by using both link network mapping and textual analysis tools, to produce both cumulative longer-term maps of interlinkages and themes across the blogosphere, and specific shorter-term snapshots of current activity which indicate clusters of heavy interlinkage and highlight key themes and topics being discussed within these clusters in the wider network.

In this paper, we discuss the methods and findings from our observation of the Australian blogosphere through 2010, exploring differences between how Australian bloggers and Australian mainstream media engage with current affairs. First, we provide an overview of our approach to identifying what constitutes 'Australian' blogging - in the absence of comprehensive national blog directories for Australia, we employed a combination of methods to move out from a snowball sample of known and active Australian blogs to a compile a more encompassing, living masterlist of blogs in the country. Second, we document the methods used to process and evaluate the data gathered from these blogs, and present the results to date: patterns of activity through the year, networks of interlinkage between blogs, and topical interests and trends over time. Emerging from this work are clear differences in interests and activities between different sections of the Australian blogosphere, as well as clear distinctions, in turn, from the focus of mainstream media coverage.

Beyond this first-stage focus on the Australian blogosphere, our project will extend this approach during 2011-12 to an examination of other social networks used by Australians, including Twitter, YouTube, and Flickr. This adaptation of our methodology moves away further from narrow models of political communication, and towards an investigation of everyday and popular communication, providing a more inclusive and detailed picture of the Australian networked public sphere. In previewing these further developments, we also address the ethical challenges inherent in our examination of content which may be public in availability, but private in intent, and could be used in aggregate to form a detailed profile of individual participants.