Re: Roll-up captions in WebVTT

I *think* future/past deals with text that is *in* the current cue, but which lies in the future/past because of extra timing inside the cue itself.


On Nov 30, 2011, at 1:17 , Christopher Giffard wrote:

> Correct me if I'm wrong, but could the :future / :past pseudo-classes enable this kind of styling?
> 
> It's possible that we'd need a :current pseudo-class in order to select the last two lines which went past, but maybe this would work too:
> 
> ::cue:past:last-child, ::cue:past:nth-last-child(2) { … }
> 
> If we were moving the implementation of roll-up captions to authors though, we need to evaluate this, as the allowed properties don't currently enable enough flexibility to implement roll-up without jumping: http://dev.w3.org/html5/webvtt/#the-'::cue'-pseudo-element
> 
> Kind Regards,
> 
> Christopher Giffard
> 
> (Sorry about the <strike>double</strike> triple message David - missed the mailing list with the last ones!)
> 
> On 30/11/2011, at 4:05 AM, David Singer wrote:
> 
>> At the moment, I can't think of a way of doing even jump-scroll without repeating text, unless (as is very likely) I am being dense.
>> 
>> Maybe someone could correct me if the best practice on jump-scroll is something other than repeat?
>> 
>> 
>> On Nov 29, 2011, at 17:40 , Christian Vogler wrote:
>> 
>>> This is not quite true if the text is duplicated in the WebVTT file.
>>> This would present serious accessibility issues for people who use
>>> screenreaders (and that includes the deaf-blind population). It also
>>> would make searching and indexing harder if text were duplicated.
>>> 
>>> Christian
>>> 
>>> On Tue, Nov 29, 2011 at 11:33 AM, David Singer <singer@apple.com> wrote:
>>>> I think a fundamental question that needs addressing is whether we expect roll-up to be (a) 'part of' the core VTT vocabulary or (b) a presentational issue that is 'optional'?
>>>> 
>>>> I tend to think the latter.  Yes, maybe smooth roll-up is easier on the eye than jump-scroll, but the same information is presented.
>>>> 
>>>> If that is the case, I think using CSS transitions is a good candidate.  Alas, we don't use CSS positioning (for obvious reasons -- positioning DOES have to be part of the core of VTT).  But that doesn't hose us.
>>>> 
>>>> Consider a stream that has Line1, Line2, …. LineN, showing in a three-line area.  A simple way to jump scroll that is simply to repeat each line, every time it 'moves' in the 3-line display.
>>>> 
>>>> <blank>
>>>> <blank>
>>>> Line1
>>>> 
>>>> then
>>>> 
>>>> <blank>
>>>> Line1
>>>> Line2
>>>> 
>>>> Line1
>>>> Line2
>>>> Line3
>>>> 
>>>> Line2
>>>> Line3
>>>> Line4
>>>> 
>>>> displayed simply, the lines appear to jump.
>>>> 
>>>> If we can convince the CSS engine that Line1 in the first cue is the same as Line1 in the second, we could ask for a transition on, say, 'y'.  That's pretty easy to do with markup -- if a line of text in two consecutive cues has the same marked-id, then it's the same, and any CS transitions declared on its class apply, if you use CSS.
>>>> 
>>>> Saying CSS transitions can be used to smooth out display transitions also allows for fade-in/fade-out (transition on color), and so on.
>>>> 
>>>> 
>>>> David Singer
>>>> Multimedia and Software Standards, Apple Inc.
>>>> 
>>>> 
>>> 
>>> 
>>> 
>>> -- 
>>> Christian Vogler, PhD
>>> Director, Technology Access Program
>>> Department of Communication Studies
>>> SLCC 1116
>>> Gallaudet University
>>> http://tap.gallaudet.edu/
>>> VP: 202-250-2795
>> 
>> David Singer
>> Multimedia and Software Standards, Apple Inc.
>> 
>> 
> 

David Singer
Multimedia and Software Standards, Apple Inc.