No, that’s not my idea. That’s a Jet Beetle made in Stanford University. Photo by Wending Tang
As mentioned in the intro, the core of this PhD or Die Trying series is to come up with interesting dissertation ideas. With my advisor backing me up, here’s my first belated cluster of brainstorms this week. Please excuse any spelling or grammatical errors. If you wish to use any of these ideas, let me know and perhaps we can work out something together.
1. How RSS spells Death to Online Serendipity
Last year, Dan Li mentioned how she had stopped surfing web pages and instead saw the web through the lens of RSS feeds. This brought about the question of whether she was truly discovering new things for herself online, or if she had confined her agenda to those of the popularist (e.g. Del.icio.us/Popular, Digg). Perhaps she was concerned about how whether she was turning into a follower (early adopter) more so than an original source / first discoverer (innovator). She explained how our early use of the web might be likened to exploring a dark cave where you might find things you intended to find, but make accidental yet wonderful discoveries along the way (serendipity). With RSS feeds and popularity web services feeding us information, we seem to be entering a pre-explored cave, complete with a guided tour. In fact, blogger GeekSpin asks: “Are we being lobotimized by memetrackers?” Since the release of memetrackers, we use memeorandum and tailrank to find the cream of the crop. “Are we missing the plot with the memetrackers, are we to focused on the opinion of other people? Are we losing our individuality? RSS subscriptions will differ from me to you, unique, but if we follow memetrackers are we not just like the zombies from “Shawn of the dead”? Have we been digitally lobotomized?” In essence, are we learning less than we realize?
Keywords: popularity, blogging, collective intelligence, education, mass media
2. Building Social Capital through Blogs
Taking our knowledge of social capital and the participatory nature of blogs, there seems to be an initial fit between both spaces. Both are democratic in nature and recognize various types of network connections, including bonding, bridging and linking. Could blogs facilitate social capital building for particular communities (e.g. academic, NGOs)? Some questions to explore might include how well principles of social capital apply to blogs as a form of social glue? Elements to look at might include types of network relationships, defining measures for such connections, mapping network connections to intended outcomes.
Keywords: social capital, blogging, collective intelligence, virtual community, collaborative
3. Can Collective Intelligence be bought out?
The collective intelligence which powers Wikipedia explains Collective Intelligence (CI) as a working form of intelligence which overcomes “groupthink” and individual cognitive bias in order to allow a collective to cooperate on one process—while maintaining reliable intellectual performance. George Pór, author of The Quest for Collective Intelligence (1995), defined this phenomenon as “the capacity of a human community to evolve toward higher order complexity thought, problem-solving and integration through collaboration and innovation.” Deconstructing CI, I see three things, namely cognition, cooperation and coordination. We see CI in real life more so then ever before. Popular news sites such as Digg and Slashdot depend on the collective intelligence of users to attach value to news submissions. Yet, if users of highest influence (e.g. Top Digg Users, Most Friends, ++ Karma Points, etc) were to behave differently, would the collective system be resistant enough to heal or replace itself naturally? On that note, what conditions are required to allow for a less exploitable democratic architecture?
4. Visualizing Emergent Opinions in the Blogosphere
The current technology for aggregating the blogosphere is limited to tracking conversations based on links, trackbacks and tags between blogs. Perhaps it would be more fruitful if we could focus on content of blogs in specific domains (e.g. by geography, by nationality, by interest) and to visualize the emergent clusters of opinion based on topics. For instance, we could use our content analysis package to compare cross-border sentiments through the blogs from countries in conflict (e.g. middle-east). This could be technically achieved by collecting the RSS feeds of a few hundred blogs in each country and processing these xml files to compare the different use of similar terms on each side. Such would include frequency and semantics of the names of countries in conflict, their concepts of war/peace, ethics: right/wrong, hate, use of profanity, just to name a few. Since our content analysis package is able to process non-roman characters as well, we could study countries where English isn’t used. For example, we could compare India & Parkistan (Sanskrit), China & Taiwan (Hanzi). Once this study is done, we would be able to decern the hotspot issues between neighboring countries which may possibly help in improving bilateral relations. More importantly, we would have a established a “mechanism” or methodology for assessing the relationship between countries. Future research may allow us to adapt this for more localized studies once we can accurately determine the geographical position of bloggers (based on Halavais’ prior research). It is hoped that instead of an instance where the Indian Govt. blocks bloggers entirely (perhaps the govt. doesn’t know where to start?), they can respond more accurately and directly to particular bloggers given such a Zeitgeist tool.
5. Mapping Your Blogosphere’s Influencers
The blogosphere is made up of conversations by bloggers where their individual participation combines into larger social movements. As seen in various studies including the recent PEW Internet Report on “Who are the Bloggers“, there has been difficulty determining the particular kind of individuals who blog (e.g. 14yr old girl who has a cat). This problem could stem from the confusing identification of a blog (no comments still = blog?) and the sheer array of blog platforms (which is most popular?). Perhaps the most prominent reason is because we’ve come to realize how there was no centrality in the blogosphere to begin with, but instead many centers or blogospheres (as Halavais first mentioned). This makes tracking influencial blogger difficult since readership (measured by FeedCount, web traffic stats, inbound links) might not be available, especially for non-tech related groups. Alternative methods of measuring influence and reputation other than hyperlinks may be required, which may include a communication study of a particular social network (via survey). By comparing a blog community’s hyperlink connections with real-world connections (e.g. phone calls, emails, face to face), we can discover more characteristics of the particular social network. As such, it may be useful to know the key influencers in particular blogospheres (e.g. knitting, cat owners, Phd students who procrastinate) for various purposes (e.g. marketing).
Okay, there are my five thoughts… I should put out more since it’s a brainstorm, but I feel this need to make everything relevant before I publish it. If you are doing anything related, do share your thoughts in the comments. Maybe we can trade anecdotes and data in that order.
UPDATE: For more ideas, I just remembered some academic trendwatching I did a while ago…