From: Sigi Reich (sr_at_ecs.soton.ac.uk)
Date: Tue 21 Jul 1998 - 08:09:54 CDT
thanks for your input. I quite like the approach of defining XPointer as an OHP locspec. This would provide a 'link' to the Web which I agree with Jacco would be a nice thing to have - and it would just demonstrate how open OHSs are.
I also thought about Jacco's statement asking whether there is something wrong with XPointer and if yes, what it is and how could we have an influence on a new version of the specification. What I believe the issue is there that in order to allow addressing in arbitrary data one would basically end up with location specifications à la HyTime - which were too 'heavy' for OHP and probably also for the current specification of XPointer. Also, Steve DeRose is one of the editors anyway so the ideas of HyTime should be well reflected in the specification process. Perhaps it's (still?) too early for the Web but this doesn't mean that a fancy OHP Picture Viewer could demonstrate how addressing in graphics files might work.
This archive was generated by hypermail 2.1.5 : Tue 13 Aug 2002 - 07:21:01 CDT