Breakout II.b (Bren 1520): "Fluid Ontologies for Social and Historical Research"
- Facilitators
- Participants
One challenge for social computing today that is philosophically, socially, and technically challenging arises in the intersection between ontologies (systems for classifying people, relations, ideas, objects, etc.) and metadata (ways of making such classifications computationally-tractable through standard tagging or markup protocols). How does a Native-American community, for example, represent the complex relationality of its social relations and socially-inflected knowledge in global information systems whose metadata schemes presume a different pattern of how things fit together with other things (as codified ultimately in relational databases)? A simple illustration: where does a name like Little Running Bear (with any additional modern American cognate name) go in a database or TEI text-encoding scheme for "first name" and "last name"? A mainstream aspect of the problem: what are the social and technological solutions to mediating between bottom-up ("folksonomical tagging") and top-down ("controlled vocabulary") approaches on Web 2.0?
The historical analog is also compelling: how do we translate between historical and contemporary knowledge-scapes? What do we do, for instance, with several thousand "anonymous" authors in 1600, or book "peddlers" in 1700 (not exactly equivalent to any social role today), or "apprentice" as a social role?
Are there "fluid" ontology and metadata schemes that can negotiate? And, if so, can they be made computationally tractable? (See the work of Transliteracies faculty member Ramesh Srinivasan, e.g., "Fluid Ontologies for Digital Museums.")
- What are the top-ten research problems/opportunities in this area?
- And is there a recommendation for a particular issue for RoSe to pursue?
Comments (0)
You don't have permission to comment on this page.