- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Fri, 19 Oct 2012 11:02:40 -0700
- To: "L. David Baron" <dbaron@dbaron.org>
- Cc: Brad Kemper <brad.kemper@gmail.com>, Yves Lafon <ylafon@w3.org>, "www-style@w3.org" <www-style@w3.org>
On Thu, Oct 18, 2012 at 6:38 PM, L. David Baron <dbaron@dbaron.org> wrote: > On Thursday 2012-10-18 10:24 -0700, Tab Atkins Jr. wrote: >> I agree, actually. However, then we should modify more of the >> property. We can't make it "positive numbers", because that's an open >> range, and we avoid those when possible. There's a legitimate reason, >> for this property, to keep a "non-negative" restriction, though - >> ideally, you shouldn't have to scan the entire document to find out >> what the first element in tab order is, at least in the common case. >> So, having a minimum value (either 0 or 1) that's the default makes it >> possible to just jump to the first element with that value. It's >> unfortunate that this means you can't put an element in front without >> moving *all* the other elements, but that's a tradeoff. > > I'm not convinced by that argument. Tabbing order requires being > able to go both forwards and backwards, and going backwards requires > finding the last item quickly. > > I'd rather just allow negative values. Ah, good point. I forgot that you can tab backwards. :/ In that case, I withdraw my statements. Let's go for the full number range, -Inf to +Inf. > (I think the more important thing we need to address with nav-index > is adding the ability to establish scopes (rather than requiring all > the indices to be global), though.) Yeah. ~TJ
Received on Friday, 19 October 2012 18:03:28 UTC