- From: Florian Rivoal <florian@rivoal.net>
- Date: Fri, 4 Dec 2015 11:08:13 +0900
- To: "Cramer, Dave" <Dave.Cramer@hbgusa.com>
- Cc: Nick Ruffilo <nickruffilo@gmail.com>, Leonard Rosenthol <lrosenth@adobe.com>, Ivan Herman <ivan@w3.org>, Luc Audrain <LAUDRAIN@hachette-livre.fr>, W3C Digital Publishing IG <public-digipub-ig@w3.org>, Robin Berjon <robin@berjon.com>
> On 04 Dec 2015, at 00:39, Cramer, Dave <Dave.Cramer@hbgusa.com> wrote: > > On Dec 3, 2015, at 10:28 AM, Nick Ruffilo <nickruffilo@gmail.com> wrote: > >> Dave, >> >> Would this suggest that there is not a global navigation that defines the order of all pages, and simply a relative positioning of next/prev in each document? Or would this be as as-well-as? > > That's an interesting question. In the spirit of progressive enhancement, I think the prev/next links could be a foundation that would work with hardly any effort--it's conceptually simple and exists today. Other mechanisms, such as using nav as a default ordering, require more work from the rendering environment but could provide fancy stuff like multiple sequences. prev/next exist today in theory, but in practice most UAs don't do anything with them. While a <nav> section in index.html would work just fine in todays browsers. Even if they don't understand that this particular <nav> is special, it's still full of links that the user can follow. Also, the semantics of prev/next aren't that straightforward, as they let you create non linear reading orders (loops, forks, going back to something else than where you can from...). Maybe this is a good thing, as it is quite expressive, but I think it can just as easily be considered a footgun. Maybe we could enforce sanity of the prev/next relationships at the validator level. - Florian
Received on Friday, 4 December 2015 02:08:41 UTC