- From: Ian Hickson <ian@hixie.ch>
- Date: Mon, 5 Nov 2007 17:22:12 +0000 (UTC)
On Tue, 5 Dec 2006, fantasai wrote: > > I believe the 'index' link type definition isn't quite right, considering > prior definitions of the term[1]. It seems to me that 'index' would be more > appropriate, given its prior definitions, for linking to things like CSS2.1's > list of properties (Appendix F). > http://www.w3.org/TR/CSS21/propidx.html > > I think that for the hierarchical root, 'top' (which was defined in HTML 3.2) > is more appropriate, both historically[2] and by virtue of its English > semantics. > > I also think that 'contents' and 'toc' should not be equated with 'top' > because > they are not always the same thing. I would expect 'toc' or 'contents' from > http://developer.mozilla.org/en/docs/Gecko_DOM_Reference > to link to > http://developer.mozilla.org/en/docs/Gecko_DOM_Reference > but 'top' to link to > http://developer.mozilla.org/ , i.e. the homepage of the website. > > [1] http://fantasai.tripod.com/qref/Appendix/LinkTypes/ltdef.html#index > [2] http://fantasai.tripod.com/qref/Appendix/LinkTypes/ltdef.html#top While I agree with the theoretical distinction, I don't really see that any practical problem would be solved by separating these into different semantics. The spec just uses it to define document hierarchy. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Monday, 5 November 2007 09:22:12 UTC