Social knowledge production and services

In taking up this week’s assignment, I found the Duguid piece particularly inspiring. His research method seemed to resonate with our task at hand. His comparison and contrast (via cross-referencing) between “traditional” knowledge production (what he calls “most reliable sources” [15] or “traditional and professional equivalents,” to borrow Dr. Gazan’s words) and social knowledge production generated by various applications of Web 2.0 sheds valuable light on how the two forms of knowledge could inform one another in complementary ways. Hoping to make a similar case as Duguid’s, especially in his section on Wikipedia, I immediately picked the first of the four general areas of comparison: Peer production in online environments vs. in-person collaboration. And I attempted to make some observations about how “traditional” and “social” forms of knowledge could complement one another. My example comes from one event in Indonesian history known as the “Madiun Affaire,” which took place on 18 September 1948 in Madiun, a provincial town on the island of Java. This is also a topic I have written a paper about several years ago by relying solely on books and articles in print.

So, what do we learn about the Madiun Affair by using Web 2.0? Not surprisingly, I first turned to Wikipedia (http://en.wikipedia.org/wiki/Madiun_Affair):

The entry informs us that it was “a communist uprising in 1948.” Then, I went on to look into “Revision history of Madiun Affair.” There are seven different versions of the Madiun entry, including the current rendition. One thing that caught my eye was that the entry was originally created on 14 July 1948 “at the Indonesian Revolution” (http://en.wikipedia.org/wiki/Indonesian_National_Revolution) and later (on the same day!) included into a broader category, “History of Indonesia” (http://en.wikipedia.org/wiki/History_of_Indonesia).

The next step I took was to compare the Wikipedia entry to its equivalent in Encyclopedia Britannica, a peer-reviewed and reputable encyclopedia (i.e. Carol Tenpir’s “Quality Still Matters” succinctly compares errors in Wikipedia and Encyclopedia Britannica). But instead of turning to Encyclopedia Britannica via UHM Library (login required), I continued to use open-access resources online. The entry on “Madiun Affair” is about the same length as the Wikipedia counterpart, and the content, too, is not so different from the Wikipedia entry (http://www.britannica.com/EBchecked/topic/355889/Madiun-Affair). For instance, the first sentence reads: “communist rebellion against the Hatta-Sukarno government of Indonesia, which originated in Madiun, a town in eastern Java, in September 1948.”

Having reviewed both entries, we learn that the Madiun Affair was a Communist uprising/rebellion against the Indonesian government in September 1948. This perspective is not so far-fetched from what you find in “traditional” resources in print. For instance, George Kahin, a leading political scientist and historian of Indonesia in 1950s-1990s, wrote a chapter on the Madiun Affair (256-303) in his Nationalism and Revolution in Indonesia (Ithaca: Cornell University Press, 1952). Today many scholars in the field consider the book as a classic on Indonesian nationalism. And Kahin titled the aforementioned chapter, “The Internal Struggle for Power from Renville through the Communist Rebellion.”

At the same time, given its sheer volume, the chapter naturally offers more in-depth coverage of the turn of events, and we meet those individuals who were closely linked to the incidence. For instance, on the outbreak of the affair, Kahin writes:
“After consolidating its control over the villages, the Pesindo [Socialist Youth of Indonesia] and other PKI [Indonesian Communist Party] troops seized control of the towns and the city of Madiun. . . . This militant phase began at 3 A.M. on September 18 in Madiun with Sumarsono and Djokosujono leading the operation. . . .” (291).

Then, can we learn more about Sumarsono and Djokosujono through social computing? I turned to Google as most Internet users would do in this type of situation and conducted a plain keyword search, “sumarsono madiun.” Interestingly (and to my excitement), the search led me to a serendipitous encounter with two columns by Ibrahim Isa (
http://www.polarhome.com/pipermail/nasional-a/2002-October/000051.html; http://www.polarhome.com/pipermail/nasional-a/2002-October/000052.html). In the posts, Isa talks about a meeting held in the Netherlands in 2002, where dozens of participants, mainly Indonesians based in Europe, sat down with Sumarsono and shared his eyewitness accounts of the Madiun Affair. In my earlier reading of published sources, I have tracked a couple of his personal accounts about his involvement in the Madiun Affair (i.e. Hersri Setiawan, 2002). But this was the first time I have heard of the meeting, and his view expressed at the meeting seems to be in line with what he has said about Madiun in the published equivalents.

To turn to a technological side of the story, Isa’s posts are archived in a list serve called National-a Archives maintained by four moderators (http://www.polarhome.com/mailman/listinfo/nasional-a). The National-a Archive is a closed and a hidden list, which means that the list administrators screen our online “applications” and decide our status (approved/rejected). The list of members is available only to the administrators. The National-a Archives list is also included in www.polarhome.com Mailing Lists (http://www.polarhome.com/mailman/listinfo). I was able to review Isa’s posts because they were archived as “prior postings,” which were released at some unknown point for public viewing. I hope that with my somewhat haphazard attempt above, I was able to give some specifics to how traditional and social forms of knowledge could possibly inform one another.

At the same time, we are also reminded about limitations of social knowledge production and services, as Leibenluft writes about Yahoo! Answers:
“While Answers is a valuable window into how people look for information online, it looks like a complete disaster as a traditional reference tool. It encourages bad research habits, rewards people who post things that aren’t true, and frequently labels factual errors as correct information.” (1)

On the other hand, Geisler and Burns are more optimistic about the usefulness of the growing web-based social tagging systems, particularly for digital video and moving image. In analyzing the mechanism of web-based social tagging systems, Lerman refers to Digg, one of the most successful social news aggregators, and suggests that the site’s success owes much to its reliable information filtering moderated by networks of “friends.” When it comes to issues of trust, the Haythornthwaite piece gives the topic a new twist by introducing three key dimensions of trust management; recognition, reputation, and reward. Finally, as Geisler, Burns, and Lerman do in their studies on social tagging systems, Gazan and Dempsey explore aspects of current innovations in social knowledge production, social annotations in digital library collections and the integration of mobile communication into library services, respectively.

Week 4 Session 1

After completing the readings, I explored two online communities that to the best of my knowledge appear to implement different social trust mechanisms. The sites I visited are: Slashdot.com and Citizendium. I chose these communities to explore aspects of information control in Web 2.0, particularly the role of experts, and to tackle a broad question of “what is quality information.” I also plan to discuss these issues in my final project.

Since “trust” is a keyword here, I would like to briefly outline what I picked up about the concept from this week’s readings. I found the Massa piece particularly relevant. Massa employs a broad notion of “trust” and uses the term to indicate “different types of social relationships between two users, such as friendship, appreciation, and interest” (51). Moreover, "[T]hese trust relationships are used by the systems to infer some measure of importance about the different users and influence their visibility on the system.” (51)

Having called the current community of Internet users a “global village,” where users can make contact and communicate with one another from distant (and local) locations, this new situation necessitates “a decentralized collaborative assessment of the quality of the other unknown people, that is to share the burden to evaluate them” (53). The reliability of information is commonly assessed and measured by what Massa calls “trust statements,” in which “a user expresses her opinions about every other user, asking how much she finds her interesting and worth her attention” (53). Not surprisingly, more and more Web sites are giving “prominence to content created by trusted users” (53).

Then, Massa goes on to classify online systems into seven broad categories, each one of which implements different (and to some extent overlapping) trust mechanisms. These categories include E-marketplaces; Opinions and activity sharing sites; Business/job networking sites; Social/entertainment sites; News sites; The Web, the Blogsphere, and the Semantic Web; and Peer-to-Peer (P2P) networks (55). According to Massa, Slashdot.com is one of the representative “News sites,” where “users can write and submit stories and news they want to share with the community.” In this type of Web sites, “the users can rate other users’ activities (posted news and comments) and these ratings are used to give more visibility to posts and comments the other users appreciate and value” (56). As for Citizendium, although it is not discussed in the Massa article, on a general level, not every user can rate other users’ activities but those who are identified as “experts” can (I will discuss information control in Citizendium in more detail below).

To go back to my final project proposal, I got the initial inspiration in Carol Tenopir’s “Web 2.0: Our Cultural Downfall?” from week 1, where she discusses (by referring to The Cult of the Amateur written by a Web 2.0 skeptic, Andrew Keen) how Web 2.0 could potentially undermine the quality of professional research and scholarship. As Web 2.0 continues to grow and elaborate its applications, Keen fears that more professional researchers would act unprofessionally and start to rely on readily accessible online resources such as Wikipedia. Massa also mentions the same phenomenon (p. 52, referring to Coates, “The Mass Amateurisation of Everything,” 2003).
Although Tenopir’s concise piece does not discuss in great detail what evidence Keen used to support his skepticism about Wikipedia and other Web 2.0-generated resources (we basically need to read the book), some of the readings for this week bring up interesting insights. Eryilmaz et al note that “Wikipedia’s technical implementation does not provide a gate keeping function to ensure quality material is being contributed” (2).

The low barriers to entry on Wikipedia is in large part a product of what Gleave et al call “a collective concern for the quality of the resulting articles” (7). Gealve et al suggest that “social roles” on Wikipedia are defined by “informal, collectively defined roles” that are “organized around practices like fighting vandalism, copy editing, enforcing standard formats, welcoming new users, evaluating article quality, and writing tools to help the community” (7). In other words, on Wikipedia, “any reader may edit or contribute” (7) whether under real name or alias. The question of transparency about sources of information is one yardstick that many authors have used to differentiate Wikipedia from other wiki projects such as Citizendium (I will get back to this point below).

Pessimism notwithstanding, Keen, as Tenopir mentions, also sees a light in the future of information control in Web 2.0. He sees that Citizendium, the wiki encyclopedia launched by a co-founder of Wikipedia, could be a potential source of more reliable and higher-quality information (than Wikipedia) because the site implements a tighter information filtering system that “combines public participation with the guidance of experts” (Tenopir, 2007).

On the differences in information control in Wikipedia and Citizendium, Sundin and Haider write in “Debating Information Control in Web 2.0”:

“While WP [Wikipedia] claims not to attribute special status to any of its contributors, CZ [Citizendium] . . . intends to assign a decision-making role to subject experts. It is not CZ’s intention to change the collaborative approach, yet here contributors have to register and write under their real name. Also, there are meant to be different categories of users which differ in terms of their expert status. In contrast, in WP anyone can add a subject entry or edit an existing one anonymously” (2).
[O. Sundin & J. Haider, “Debating Information Control in Web 2.0: The Case of “Wikipedia vs. Citizendium. Short paper to ASIS&T 07,” Joining Research and Practices: Social Computing and Information Science, October 18-25, Hyatt Regency Milwaukee, Wisconsin]

Citizendium’s quest for reliable and high-quality information is highlighted in the “About” page [http://en.citizendium.org/wiki/CZ:About].

Then what about Slashdot.com, the second online community I explored for this assignment? The Sundin and Haider article I mentioned above makes references to two discussion threads on Slashdot.com, “A Look Inside Citizendium” (June 2006) and “Co-Founder Forks Wikipedia” (October 2006), each of which assesses strengths and weaknesses of Citizendium (and Wikipedia) in the area of information control. One post in the latter caught my eye [see screenshot].



Not many of us would disagree with what “nlitement” says here in the last sentence: "In the end, it's really up to the end-user to weed out bad information." Web 2.0 is a user-oriented and user-generated communication tool. It is users who decide on what constitutes “good” or “bad” information. We reach out the community of Internet users to satisfy our information needs regardless of whether the needs are driven by our areas of interests, as we experimented on Week 2, or defined by impending tasks at hand like our experiences on Answerbag.

Insofar as my experience in Citizendium is concerned, I am not too impressed with the quality of information presented on the site. For instance, I looked up “Southeast Asia,” a topic I am familiar with, in both Wikipedia and Citizendium to test the water. In this particular instance, Wikipedia happened to offer more current and accurate information [see screenshots].

Wikipedia

Citizendium

Note: East Timor, the newest nation in the region (since May 2002) is not included here.

The Wikipedia entry was last modified on 1 March 2009, whereas the Citizendium entry was last modified on 3 February 2009. I will see if information gets updated (corrected) in the coming weeks.

Are we being to critical of Wikipedia? We’ve been hearing a lot of criticism about the inaccuracy of information on Wikipedia. But what about proponents of utilizing Web 2.0 in professional research and what do they have to say? In evaluating the quality of information on Web sites, how do we take into account of what might be called advisory opinions expressed in an open-access forum like Slashdot.com and a scholarly commentary by an expert like Carol Tenopir (i.e. “Quality Still Matters”)? How do they compare? This is just a random list of (unanswerable) questions I am hoping to address in my final project.

Finally, just a random reflection on Williams, Allen et al and Ellison et al. Allen et al made a creative analysis of how “trust” is negotiated in a particular (and paradoxical) setting, where individuals collaborate to accomplish a task while playing against each other. The Williams article is a nice introductory piece on “social capital,” which the author sees as “a contentious and slippery term” (2). But a main idea behind “social capital” is, just like in financial world, “using it creates more of it” (2). The article nicely outlines two types of social capital. “Bridging” tends to be more inclusive and provide little emotional support, whereas “bonding” can be (mutually) exclusive and provide “emotional or substantive support for one another” (5). Ellison et al also focus on aspects of “social capital.” The authors suggest that participants in their survey (college students) use Facebook to “keep in touch with old friends and to maintain or intensify relationships characterized by some form of offline connection such as dormitory proximity or a shared class” (21). The use of Facebook in these manners forms a type of “social capital,” which the authors characterize as “weak ties” (21). I wish the authors elaborated a little more about social psychology of non-users: What motivates them to not use Facebook? Is lack of time a factor here?