[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [cam] RDF for CAM?
Serm, Excellent thoughts. Yes - the RDF would need to be customized to provide inheritance into "the semantics". I think this mechanism would be quite different from datatyping or UID lookups which provide that data binding type "stuff", so the fact that RDF is limited may not be an inhibitor. The RDF has potential to have a CAM template do more for agent directed stuff. So the RDF semantics would be purposed to tell an agent "about" the content in some abstract way. This would be lightweight and simple so the RDF should work quite nicely for this. Particularly on the discovery side - if an agent has choices of two or three similar CAM templates - which one does it pick? It can look at the context variables to see what values the CAM context reacts too - but if it can also "see" the RDF about content - it may make choices based around that too... Thanks, DW. ======================================================= Message text written by "Boonserm (Serm) Kulvatunyou" > David, Sounds like an interesting topic here. My opinion is RDF has quite limited expressiveness to express semantics. We may want to consider OWL. However, OWL needs special parser to validate as it interpret things (such as xml schema) a bit differentlly. My immediate thought is since all of this semantic annotation language use the notion of "resource" which by defintion should be able to point to any peice of information in the web space (as the whole file or specific element), one should be able to provide semantics for each CAM element by referencing. -serm<
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]