Re: STYLE section inside VTT file


> On May 17, 2016, at 8:24 AM, Simon Pieters <simonp@opera.com> wrote:
> 
> On Mon, 16 May 2016 15:04:22 +0200, Bradley Botkin <bradley_botkin@wgbh.org> wrote:
> 
>> Sorry for the newbie question… I’m confused about using the STYLE section in a WebVTT file.
>> 
>> There’s quite a bit of documentation and postings about the STYLE section inside a VTT file. It boils down to something like this (for example):
>> 
>> WEBVTT
>> 
>> STYLE
>> ::cue(.red){ color: red; }
>> 
>> 
>> 1
>> 00:00:00.100 --> 00:00:50.000
>> <c.red>This should be a red caption</c>
>> 
>> My tests with Safari/Chrome/Firefox (all Mac) all fail to render the red caption.  However, putting that same style markup in the CSS within the HTML all succeed in rendering the red caption. I’m sure I’m missing something obvious.  Caption vendors will typically be able to deliver a WebVTT file to a client, but have no access to underlying HTML. It should never be necessary for the Webmaster to modify HTML in order to display captions properly.
>> 
>> 
>> For a demonstration of my problem, point a browser at:
>> 
>> https://s3-us-west-2.amazonaws.com/ncamftp.wgbh.org/brad/webvtt/capswithstyles.html

>> 
>> 
>> What am I doing wrong?
>> 
>> Thanks.
> 
> Hello Brad! STYLE is not yet implemented, that's why it doesn't work. :-) Please file issues for browsers asking to implement it. If it isn't implemented it will eventually be removed from the specification.
> 
> cheers
> -- 
> Simon Pieters
> Opera Software

I’ll do that, but… I don’t see how VTT can become adopted for any use-cases involving delivery of caption/subtitle data by caption vendors from *outside* a particular web video environment. So the current implementation is fine for a dedicated caption editor within a targeted web ecosystem (e.g. YouTube, or other similar) where the ecosystem has complete control over all HTML/CSS pieces and the caption author only enters text and styles it using build-in widgets.  But for delivery *into* such closed ecosystems by caption authors from outside those environments using 3rd party tools, lack of STYLE support inside VTT files is a non-starter.  I’m surprised that working groups and implementations and got this far without eliciting input from actual caption/subtitle vendors.  It seems a glaring omission.

Cheers,
—Brad

Received on Thursday, 19 May 2016 12:49:39 UTC