[F] WebVTT support

[1aurend]

No description provided.

[1aurend]

Dana, Alexa, question for both of you — this adds support for only one text track per video resource. Should we expand it to handle captions in multiple languages/other types of text tracks?

And for Dana— the package we're using for rendering the video player is deprecated and hasn't been updated in years. Should we take the opportunity to replace it? If so, do you have a preferred implementation?

[scryptmouse]

I can't really say. If it's easy, we should support as many as makes sense. If it's hard to model or expose it in the UI, I think it depends on how likely it is folks are going to need multiple tracks.

[dananjohnson]

Yeah. I think the known use case we have is for a single track. Support for multiple sounds good, but only if it's not too difficult to do. On the frontend, I guess we'd check for a track in the currently set language and fall back to "en"?

We should upgrade the package, but lets put a pin in it and revisit when we work on dependency upgrades next quarter.

[1aurend]

Sound good re: package upgrade.

And probably makes sense to wait on multiple languages/tracks until we have a use case. The video player package displays a dropdown list of languages so the users can choose which to show, if it's supplied with more than one text track of kind="captions". So, that part could be pretty trivial. As for modeling it, we could do something like what we did with alt_text to associate a text_track_lang and/or text_track_kind field with the relevant attachment. Anyhow... thoughts for if we get there.