(Just a quick note to capture an idea that cropped up over lunch with Marja) Our XML/RDF annotation vocabularies should capture the language primarily used in any textual content within the annotation. For eg., as a dc:language property of the annotation. This would allow consumer applications to filter out a particular natural languages view of some annotation resource. For example use case, Spanish or chinese or .... annotations of http://www.w3.org/2001/sw/ could be pulled out via an RDF query. For users browsing through content in a language that isn't their first, this could be a really useful service: one might view a page overlaid with all the annotations of it that have been made in some selected natural language. (I'd expect this to be one of a vast number of ways we'll select annotations of interest using RDF queries). Related idea: annotation servers implemented over HTTP might be able to pick up language of the annotator from HTTP headers, rather than needing special case support. Dan -- (semantic web geek) http://www.w3.org/People/Danbri/Received on Tuesday, 1 May 2001 06:00:10 GMT
This archive was generated by hypermail 2.2.0 + w3c-0.30 : Friday, 25 March 2005 11:19:17 GMT