sociosemantic_web

toc

The sociosemantic web vision
This page http://tagunity.wikispaces.com/sociosemantic_web DRAFT 1161149822

Peter Morville's approach to the structuring content (in online databases) incorporates both formal and informal notions of content classification. Morville's vision for a more "sociosemantic Web" privileges both social and technical aspects of content classification. Morville argues of the politics of community needs to be considered, and points to the need to consider the impact of social factors on community members’ classification practices.

The notion of the "sociosemantic Web" is, in part, Morville's attempt to overcome what he says Tim Converse calls "Semantic Web Disease," a disease that begins as a mild irritation, that if left untreated can progress "to a full blown belief that content creators everywhere will work together in harmony, and speak with one (meta-) voice," ([|Tim Converse], quoted Morville, 2005).

Morville suggests that metadata (data about data) has a social life (2005, pp. 125-126), since, "It is metadata's ability to help people find what they need that has driven a resurgence of interest [content] classification in databases on the World Wide Web. These approaches are both formal and informal, and range] from the ontologies of the Semantic Web to the folksonomies of social software," (Morville, 2005 p. 125).

Morville on boundary objects
The notion of "boundary object" is central to Morville's vision of "sociosemantic Web". To Morville a "boundary object" is an artefact or idea that is shared but "understood differently by multiple communities." That is to say, individuals may be members of different groups and attach "different meaning to [a single] boundary object, [which] serves as a common point of reference and means of translation," (2005, p. 119). (See boundary object, a term coined by Susan Leigh Star in 1988.)

Morville sees that while folksonomists and ontologists "face similar problems ... they speak in different tongues," (p. 126). He tentatively offers: "Hopefully, we can use metadata as a boundary object, to foster translation, build shared understanding, and encourage real social progress,” (p. 126).

Content classification and knowledge sharing in communities
Morville is justifiably tentative in his proposal of boundary objects as metadata enablers of social knowledge sharing. In his summary of the merits of folksonomies versus ontologies, Morville highlights the point that both take a techno-centric view of how classification schema can be created by individuals using online application(s). This, in turn, highlights the fact that neither folksonomies nor ontologies provide a framework for understanding knowledge sharing within communities. Yet Morville appears eager to find some way of including social elements in his notion of the sociosemantic Web. The sociosemantic Web Morville suggests will need to incorporate notions of human choice, and to incorporate social factors that will win users’ confidence. As Morville argues, "Our ability to make informed decisions will depend on how we allocate attention and trust, [and] how we define authority," (2005, p. 154). He suggests our journey will offer many paths, from which to choose, "As we build our Internet of objects, the permutations of sociosemantic metadata will create new avenues of findability ... as every object and location [i.e. unique URI] sprouts tags: social and semantic, embedded and unembedded, controlled and uncontrollable,” (2005, p 153). Morville continues: "To manage complexity we must embrace faceted classification, polyhierarchy, pluralistic aboutness, and pace layering. And to succeed, we must collaborate across categories, using boundary objects to negotiate, translate, and forge shared understanding," (2005, p 153-154).

Yet, as Bowker and Star (1999) point out, content classification is a political process.

Morville’s vision for a more sociosementic Web hints strongly at the need for a better understanding of //what people do// when they contribute and classify content in databases on the World Wide Web. While Morville is emphatic: "The magic of the boundary objects lies in its ability to build shared understandings across social categories," (2005, p. 119), the sentence deserves to be rewritten to incorporate ideas of practice and community: "The [potential benefits of understanding the classification of] boundary objects [in database on the World Wide Web] lies in [a better understanding the role of social processes in boundary objects'] ability to build shared understandings across social categories." That is to emphasise that social knowledge creation and sharing is a large jigsaw, of which content classification in social software application is but one small piece.

The role of communities
A key aspect of tagging in online applications is that taggers can follow the links behind each tag to "discover more about [the] topics and the people behind them. And [thus] learn about who else is interested in a [particular topic], [because in] a sense, the object serves as a seed for emergent community, (Morville, 2005, p137). An online application can produce a list of people who have tagged a particular object, but this does not make the list of taggers into a community. It is merely a list.

Sociosemantics beyond the techno-centric
The challenge in moving beyond Morville's mostly techno-centric vision for the development of shared semantics (e.g. tags), is to conceptualise users not merely as individuals but as groups. Users of online systems are not only isolated individuals connected electronically via their networked computers. They are also members of dynamic communities with complex social and political relations. They are people who create and share knowledge through not only their use of online database applications like those support various types of content classification, but who speak to each other and develop and share knowledge through their collective activities and shared practices.

1153051424