- From: Marja Koivunen <marja@annotea.org>
- Date: Thu, 05 Jan 2006 13:53:31 -0500
- To: Marja Koivunen <marja@annotea.org>
- CC: iris garden <iris2000sa@yahoo.com>, semanticweb@yahoogroups.com, semantic-web@w3.org
I found the examples and it seems that the idea is to not do collaborative tagging but actually add the tags in the documents with class definitions. This is actually what I and then some others started to do in Web Accessibility. For instance, to have images and corresponding captions marked with class="figure" and caption with class="caption". The idea is that if more people use the same classes it is easier to find certain parts of the documents with screenreaders and by other tools benefitting from the structure. Unfortunately we cannot rely on people always using the "figure" for the image. To make this work better in the Semantic Web would be to attach a URI that describes the "figure" class by using Semantic Web metadata. Then it would not matter so much what class names were used. Maybe we could use the embedded RDF for this? Marja Marja Koivunen wrote: > > iris garden wrote: > >> Hi >> >> There is a new specification called microformats (microformats.org), >> that use special tags to annotate part of an XHTML file, my question >> is to what extent does this new initiative going to overlap with >> semantic web grow? >> >> Thanks >> Iris >> >> ------------------------------------------------------------------------ >> Yahoo! Photos >> Ring in the New Year with Photo Calendars >> <http://us.rd.yahoo.com/mail_us/taglines/photos/*http://pg.photos.yahoo.com/ph//page?.file=calendar_splash.html&.dir=>. >> Add photos, events, holidays, whatever. > > > Annotea > (http://kmi.open.ac.uk/events/usersweb/papers/01_koivunen_final.pdf) > is based on Semantic Web and it has done annotations on parts of any > XML resources by using XPointers to point to the exact context in the > document for a long time (see slides 19-21 in > http://www.w3.org/Talks/2001/1021-KCAP/slide19-0.html) . Additional > checksums, using ids and CVS information etc. has also been discussed > to make these pointers more robust when the documents change. > > Based on discussions with users we started Annotea bookmarks and > topics (a form of tagging, social bookmarks, and folksonomies based on > Semantic Web) by just using the resource URI. However, from the > beginning we also discussed and nothing really prevents from adding > similar context property to bookmarks except at the time we felt the > world was not yet ready for that. The context would tell the location > in the XML document but for a long time we also have had plans to > extend the context for time codes for video and areas for bitmap > images etc. Some of these extensions have been experimented by others > as our resources have always been scarce. > > So I do think that tagging and microformats can very well live in > Semantic Web if done right. Annotea has all the basic means already. > > Marja
Received on Thursday, 5 January 2006 18:54:08 UTC