Re: WebVTT updates

The spreadsheet just says 'bidi_ruby.html' with no further details.  Where
can I find the details of the rendering test vectors?

Thanks,
David

On Fri, Mar 8, 2019 at 8:02 PM Silvia Pfeiffer <silviapfeiffer1@gmail.com>
wrote:

> Hi David,
>
> The spreadsheet [4] that Gary attached had all the answers about
> implementations.
> With the Japanese requirements, WebVTT relies on the browser support of
> those features.
>
> Hope this helps.
>
> Cheers,
> Silvia.
>
> On Fri., 8 Mar. 2019, 9:14 am David Ronca, <dronca@netflix.com> wrote:
>
>> Gary,
>>
>> We are interested in understanding where WebVTT is in terms of global
>> subtitle support, so that we can determine our WebVTT roadmap.  Is there a
>> WebVTT implementation that supports the essential Japanese Subtitle
>> Features (as listed below) that we can work with?
>>
>> Essential Features
>> ● Vertical text
>> ● Ruby annotations
>> ● Side marks (‘bou-ten’)
>> ● Horizontal-in-vertical (tate chu yoko)
>> ● Side lines (‘bou-sen’)
>> ● Oblique (Slanted) (‘nana-me’)
>>
>>
>> Thanks,
>> David
>>
>> On Wed, Mar 6, 2019 at 8:32 AM Gary Katsevman <me@gkatsev.com> wrote:
>>
>>> Hello everyone,
>>>
>>> I wanted to send out an update regarding WebVTT and to start the ball
>>> rolling on the next steps.
>>>
>>> This week I was at FOMS and was speaking with Crunchyroll and they are
>>> investing heavily in WebVTT, which I thought was very cool and wanted to
>>> share.
>>>
>>> I've made a pull request[1] against the WebVTT repo to mark the
>>> following as at risk:
>>> - collision avoidance with snap-to-lines false,
>>> - ::cue-region pseudo-element
>>> - :past and :future pseudo-classes
>>>
>>> STYLE blocks are not marked as at risk because I have a prototype[2]
>>> that will be fully implemented by the time the 28-day wait period of
>>> updating the CR has finished.
>>>
>>> I have also made a branch[3] with test removals to match the at-risk
>>> features. With the test removals of the at-risk features, there is only one
>>> test that fails (see the rendering 2/27/19 sheet on the WebVTT web platform
>>> tests spreadsheet[4]). This one test is because browsers have not yet
>>> implemented `text-wrap: balance`[5] and the WebVTT defines that cues should
>>> have that property set.
>>>
>>> I think we should proceed with marking the above features as at-risk,
>>> then working on a new draft that removes the text to get it moved to PR.
>>> Afterwards, we can create a new Draft or even CR with the at-risk features
>>> put back in and work towards a new version of WebVTT.
>>>
>>> Thanks,
>>> Gary.
>>>
>>> [1]: https://github.com/w3c/webvtt/pull/449
>>> [2]: https://github.com/videojs/vtt.js/pull/35
>>> [3]:
>>> https://github.com/web-platform-tests/wpt/compare/master...gkatsev:webvtt-at-risk?expand=1
>>> [4]:
>>> https://docs.google.com/spreadsheets/d/e/2PACX-1vSHzJtwi_2ijdV7hRUH0Fe8qZeSwMaCLErK7w7U_IFzhzNmW9YwPRWrER1eHR9-4Eo0kqyXJ4G35b1b/pubhtml
>>> [5]: https://www.w3.org/TR/css-text-4/#valdef-text-wrap-balance
>>>
>>