- From: Nick Kew <nick@webthing.com>
- Date: Tue, 19 Mar 2002 01:29:46 +0000 (GMT)
- To: Jim Ley <jim@jibbering.com>
- cc: <www-annotation@w3.org>
On Tue, 19 Mar 2002, Jim Ley wrote: > > Not a problem. A hash on a string comprising the (lowercase) names of > all > > elements in a DOM - separated (eg) by space - will do nicely. > > All HTML 4.01 elements? perhaps to cope with elements which different > parsers may ignore, for example IE places a <moomin> element into the DOM, > and I don't believe Mozilla would. Erm, you lost me. Surely you can set the parser NOT to generate extraneous gunk? > > Text content and attributes should be ignored: changes to these might > > slightly shife a pointer, but won't invalidate it to the point of > > becoming totally wrong and/or misleading. > > I still think it might be worth adding id's or href's or something as > changes of these do indicate changes of substance of the document, > although that's perhaps something more for other areas. I don't think so. They indicate change which may affect the annotation, but won't invalidate the XPointer to it. As for links, flagging when they've changed is a separate issue. Maybe we can use annotea for that, but it's not what the present proposal is about. -- Nick Kew Site Valet - the mark of Quality on the Web. <URL:http://valet.webthing.com/>
Received on Monday, 18 March 2002 20:30:12 UTC