Re: WebVTT feedback

On Mon, Dec 5, 2011 at 7:38 PM, Ian Hickson <ian@hixie.ch> wrote:

> > If wrapping to similar line lengths is what's wanted, then the VTT
> > renderer (and so CSS) should support that.
>
> Possibly. It's not currently. It would require a new 'white-space' value.
> Given that, it would be easy to support in VTT; just have authors not use
> line breaks at all, and let the renderer do it.
>

I think this is much better than having people do it by hand.

I guess I'd have to see evidence that anyone actually wants to include
> inline comments like this. I don't recall seeing any in the files I
> examined when researching captions.
>

Just to be clear, I meant that I'm fine with handling my use case (which
may be rare) by using a class with an associated display: none style.

I know that this is a discouraged way of using CSS (since disabling the
stylesheet will break the rendering, by causing comments to always be
shown), but I can live with that.  If you want to give people a way of
supporting inline comments without abusing CSS like this (eg. @hidden),
that's fine too, but I don't know if it's worth it either.

On Sat, 3 Dec 2011, Philip Jägenstedt wrote:
> > We're going to be doing the same script detection heuristics that we do
> > on web pages. Differentiating between simplified Chinese, traditional
> > Chinese and Japanese isn't particularly hard.
>
> Can we define these for interoperability, or are they proprietary? (I
> don't imagine people writing their own small WebVTT implementations are
> going to know how to do this if we don't have a spec.)
>

It's like character set guessing (character frequency analysis, TLD
matching), which I don't think can really be specified, because it's
inherently fuzzy and improvable and in some cases may use the user's locale
as part of the decision.  I'd be happy to be wrong here, though.

There's another wrinkle that makes it a bit more complicated than charsets,
though: you can put any character in any language.  What does <span
lang=en>漢字</span> do?  That probably also falls back on heuristics, which
means that you can hit the heuristics even when a language is explicitly
specified, a problem that charsets don't have.

These are all really HTML questions.  As far as WebVTT is concerned, I
think it should just pass the language off to @lang.


(I find the formatting of the following quote particularly apropos to the
topic...)

On Mon, Dec 5, 2011 at 9:00 PM, John Foliot <jfoliot@stanford.edu> wrote:

> > Possibly. It's not currently. It would require a new 'white-space'
>  > value.
> > Given that, it would be easy to support in VTT; just have authors not
> > use
> > line breaks at all, and let the renderer do it.
>
>         [CC-8] Allow the author to specify line breaks.
> http://www.w3.org/WAI/PF/media-a11y-reqs/#captioning
>

Allow, not require; nobody's suggesting that authors shouldn't be *able* to
break lines.  Line breaks should be explicit, like <br>, not caused by
newlines.

-- 
Glenn Maynard