Re: Validity constraints on <section>

On Fri, Mar 22, 2013 at 8:39 AM, Jirka Kosek <jirka@kosek.cz> wrote:

> On 21.3.2013 22:00, Silvia Pfeiffer wrote:
>
> > I look at it like a book's table of contents. It could have its own
> section
> > or be used before any sections start. In books, the table of contents
> often
> > has its own section and is listed in the table of contents. It would
> create
> > a shadow dom with links to all the sections. In books they would also
> have
> > page number - when printing such a Web page, page numbers could be added,
> > but for online viewing they would just be links.
>
> I think that Robin's idea of having just memory structure of outline
> which can be then processed in any way is more useful. In my experience
> there are too much ways in which ToC can be presented -- with/without
> numbers, with/without inline formatting inherited from headers, just
> some levels of headers, with more information then just headers (e.g.
> also names of authors of each section), ...
>
> If I understand to your proposal correctly shadow DOM for outline will
> be constructed in one particular way which will not work for everyone.
> Of course, as always there is trade off between complexity and
> flexibility of approach used.
>

That's no different to other complex controls in the Web platform: e.g. the
video controls are rendered by every browser, but if you don't like them
and want them to look different, you can run your own.

Silvia.