A Partnership between Structured Data and Ontotext to address weaknesses in linked data framed it this way:
Volumes of linked data on the Web are growing. This growth is exposing three key weaknesses:
- inadequate semantics for how to link disparate information together that recognizes inherently different contexts and viewpoints and (often) approximate mappings
- misapplication of many linking predicates, such as owl:sameAs, and
- a lack of coherent reference concepts by which to aggregate and organize this linkable content.
The amount of linked data is trivial compared to the total volume of digital data.
Makes me wonder about the “only the web will scale argument.”
Questions:
- How do these three “key weaknesses” compared to current barriers to semantic integration? (3-5 pages, no citations)
- “inadequate semantics?” What’s wrong with the semantics we have now? Or is the point that formal semantics are inadequate? (discussion)
- “coherent reference concepts?” How would you recognize one if you saw it? (3-5 pages, no citations)