- From: Daniel Rubin <rubin@med.stanford.edu>
- Date: Wed, 21 Mar 2007 21:34:22 -0700
- To: Dan Brickley <danbri@danbri.org>, "Miles, AJ (Alistair)" <A.J.Miles@rl.ac.uk>
- Cc: SWD WG <public-swd-wg@w3.org>,public-esw-thes@w3.org
Yes, what you are describing is that links between objects and concepts are always bidirectional with the corresponding inverse relation. I think that's an important generalization of this discussion Daniel At 04:59 AM 3/21/2007, Dan Brickley wrote: >Miles, AJ (Alistair) wrote: >>Hi Daniel, >>Actually I meant using multimedia objects to annotate concepts, rather >>than the other way around. >>Sorry, reading my email again, it's pretty ambiguous. Which is not to >>say that use cases for annotating images and image regions aren't >>equally important :) > >Ah, I misunderstood too. Annotation images, image regions, videos, >and video portions, audio, and audio portions --- definitely very >much of interest, even if that wasn't what you were asking. > >Perhaps the difference is minimal: providing typed links between >things in effect provides information about both entities. If I say >that you work for CCLRC, ... that tells the world a little about >you, and a little about CCLRC. So same with concepts? If I say that >avalanche.jpg indicates some-skos-scheme-the-concept-of-disaster, >... in some ways I've also annotated the concept is being >illustrated by the media object too? > >Dan >
Received on Thursday, 22 March 2007 20:49:31 UTC