- From: Sander Tekelenburg <st@isoc.nl>
- Date: Sun, 23 Sep 2007 19:30:13 +0200
- To: public-html@w3.org
At 13:08 +0200 UTC, on 2007-09-22, Charles McCathieNevile wrote: > On Sat, 22 Sep 2007 07:52:33 +0200, Sander Tekelenburg <st@isoc.nl> wrote: [...] >> Sounds awfully semantic to me. Wouldn't it be better if HTML would allow >> authors to mark up such *meanings*, and leave it up to the UA to provide >> the appropriate keyboard shortcut?... > > This is indeed the sort of thing that @role is designed for - and in > particular the extensibility of @role was meant to cope with such a case. Thanks. OK, so @role, or something similar, then seems to be something to seriously consider for inclusion in HTML5. (But I seem to remember there are objections?) > If the UA has no idea what the role is, then it is helpful if the author > has requested a shortcut (or 6) that the UA can use as a mnemonic if it is > available in some way. Agreed. >> The main problem with this would be that with a predefined set of such >> "roles", there could still be cases where authors wish to provide >> keyboard access to non-predefined "roles". > > The current design of role allows for this with extensibility. But there > is a proposal on the table with a lot of support that in order to have > @role in HTML 5 it would lose the xtensibility Sorry, I can't follow that last sentence. What proposal, on which table, and what "support"? (By parties?) >> Possibly the old @accesskey can be upgraded > > Well, it could just be used. It doesn't provide much useful info, but it > does provide some. Right. I meant "upgraded" just in the sense that it be specced more clearly (than HTML 4.01 does), so that authors can know how to use it, and that gives interoperability. >> it just isn't a good idea at all to let authors define key combos. > > I think everyone who has though about this recognises that the author's > suggestion should not be automatically followed, since it may introduce > conflicts. Exactly. -- Sander Tekelenburg The Web Repair Initiative: <http://webrepair.org/>
Received on Sunday, 23 September 2007 17:33:49 UTC