Connection, Fragmentation, and Intentionality: Social Software and the Changing Nature of Expertise

Connection, Fragmentation, and Intentionality: Social Software and the Changing Nature of Expertise

Christopher Watts (St. Lawrence University, USA)
DOI: 10.4018/978-1-4666-4301-7.ch042
OnDemand PDF Download:


Social software forms new kinds of collectives and expands the means of producing and disseminating knowledge. Yet the combination of persistent connection and fragmented communication can undermine intentionality. Philosophies of technology that privilege data over users exacerbate this danger; more humanistic approaches to software design are now required. Through personal observation and an examination of recent literature (largely drawn from the popular press), the author examines the philosophies that underlie social software designs, explores ways they affect interaction, describes potential pitfalls, and theorizes a reimagining of expertise in this context—complementing, rather than replacing, the scholarly traditions of the academy.
Chapter Preview


The use of social software in almost every aspect of daily living has exploded in recent years in a way that the earliest adopters of the Web could never have anticipated. It has changed how human beings interact with one another in ways both positive and negative. Perhaps the chief benefit of social software is most succinctly described by Shirky (2008): organizing without organizations. Social software allows users to send and receive information in targeted ways, and can potentially be a powerful way of creating and organizing new knowledge. In higher education, the realization of the epistemological function of digital media technologies (including social software) has been somewhat slow. That is, it would seem that many college professors have been reluctant to see such technologies not only as means of disseminating knowledge, but also as sites of knowledge creation. Making a short video about topic x, for example, is not simply a means of showing someone else what one has learned about that topic; making the video is also a way of knowing about the topic. Uploading that video to a video-sharing site and participating in an online dialogue about it furthers both the creation and dissemination of knowledge about topic x. For a student, this experience can also help to make the concept of joining an ongoing scholarly conversation more concrete—something that does not always happen when the end product is a paper that only a professor will read.

What do educational institutions know about the intersection of relatively new social software and the age-old academy? Some of the benefits are clear enough. The most coordinated ways in which academic institutions have adopted social software, it seems, have been related to admissions and public relations. “Some colleges are treading the new territory with specific strategies—to recruit students or engage alumni—while others are showing up and feeling their way” (Lipka, 2009: para. 7). Social software has crept into the classroom in a more decentralized manner. (The same trend may now be seen with mobile computing; see Keller, 2011.) Many instructors have experimented with using popular social software platforms in place of institutional learning management systems to coordinate communication and resources for class members. There are many different ways to do this, and many different reasons why it might be done. Social software is conducive to ad hoc activities by design, and perhaps there is no need to coordinate such activities in broad ways. However, it is, of course, extremely useful to share best practices, and some organizations, such as the New Media Consortium and EDUCAUSE, are doing so (New Media Consortium and EDUCAUSE Learning Initiative, 2007, 2008).

As with any relatively new technology, the current crop of social software has been the target of some criticism, much of which has been dismissed as Luddite whining. In truth, a number of very smart, thoughtful, and technologically sophisticated individuals have offered serious, considered criticism, and their arguments are well worth considering by technology enthusiasts and Luddites alike.

In the last few years, there has been a great deal of writing about technology in general, and social software in particular, by technology luminaries from across the disciplines (using the quaint technology of the book). The questions raised in this chapter have now become relevant for a general readership. Interestingly, much of the material that I have found most relevant appears in the popular press, and it is from the popular press, then, that I will draw most of the context for this chapter. In 2010 alone, a number of compelling and contradictory books were published that have profoundly affected my thinking. I am indebted to several recent books in particular that have helped shape my ideas: Kelly’s What technology wants (2010), Lanier’s You are not a gadget (2010b), Rushkoff’s Program or be programmed (2010), Siegel’s Against the machine (2008), and Turkle’s Alone together (2011).

The objectives of this chapter are fourfold: (a) to examine briefly some of the philosophies that underlie social software designs; (b) to explore the ways that these designs subsequently affect interaction among individuals and groups; (c) to describe pitfalls of this arrangement that are particularly relevant to higher education; and, finally, (d) to call for a reimagining of expertise in a way that builds on (rather than replaces) the scholarly tradition of the academy.

Complete Chapter List

Search this Book: