search

w3c / imsc

TTML Profiles for Internet Media Subtitles and Captions (IMSC)
https://w3c.github.io/imsc/
Other
31 stars 17 forks source link

APA WG comment: semantic layers #524

Open michael-n-cooper opened 4 years ago

michael-n-cooper commented 4 years ago

Section C. Forced content This seems like a temporary solution to us. Mightn't it be better to define semantic layers of information so that each layer could be made visible and invisible at runtime as appropriate for the user? For example, the user might want to see either speech-only (subtitles), (parts of subtitles) or some combination of these.

css-meeting-bot commented 4 years ago

The Timed Text Working Group just discussed APA WG comment: semantic layers, and agreed to the following:

The full IRC log of that discussion <cyril_> Topic: APA WG comment: semantic layers
<cyril_> github: https://github.com/w3c/imsc/issues/524
<cyril_> nigel: TTML and IMSC permit metadata description to be specified on particular bit of information
<cyril_> ... there is no formal requirement to do anything on that
<cyril_> ... the facility to have layers exist already
<cyril_> ... by using e.g. ttm:role
<cyril_> ... but there is no normative requirement on processor to use it
<cyril_> ... so force content provides a clear mechanism for authors and processors to define a interoperable presentation behavior
<cyril_> pal: "Forced" is a very specific tool for a very specific use case
<cyril_> ... the broader question is how to indicate the semantics of timed text
<cyril_> ... how to get the consistency across the ecosystem but that's beyond the scope of IMSC
<nigel> q?
<cyril_> q?
<nigel> ack cy
<cyril_> ack
<cyril_> SUMMARY: this is a really interesting topic, but we don't think we can make any useful change to IMSC in response to this comment
<cyril_> SUMMARY: TTWG suggests this should be the beginning of a conversion with APA and other interested parties
gzimmermann commented 4 years ago

I think engaging into a conversation with TTWG makes sense. We need to develop meaningful use cases and clearly state the purpose of a semantic layering.

css-meeting-bot commented 4 years ago

The Timed Text Working Group just discussed APA WG comment: semantic layers imsc#524, and agreed to the following:

The full IRC log of that discussion <nigel> Topic: APA WG comment: semantic layers imsc#524
<nigel> github: https://github.com/w3c/imsc/issues/524
<nigel> Nigel: I think the proposal here is that we agree to hold a joint meeting with whoever
<nigel> .. wants to attend, and that we will make no change to IMSC 1.2 to address this.
<nigel> Pierre: I guess they're not objecting to deferring this?
<nigel> Nigel: Agreed, that's my reading anyway.
<nigel> SUMMARY: TTWG would like to participate in a joint meeting to progress this, and will make no changes in IMSC 1.2 to try to address this.
gzimmermann commented 4 years ago

I agree. This issue needs more time than we have currently available for IMSC 1.2. We should arrange for a joint meeting after IMSC 1.2 was published.

michael-n-cooper commented 4 years ago

APA agrees to defer this to future version.

michael-n-cooper commented 4 years ago

Re-adding a11y-needs-resolution label because of new information about how the horizontal review tracker works, but this is still signed off from APA.

nigelmegitt commented 1 month ago

Discussed in joint call with APA, TTWG and MEIG today 2024-09-27. Further discussion needed; in the meantime, a note pointing to the potential use of DAPT metadata for represents and language source would be helpful.

palemieux commented 1 month ago

I am not excited to introduce additional metadata without a matching track selection algorithm, such as the one specified in Common Manifest.

nigelmegitt commented 1 month ago

Same, as a normative requirement. The idea of the note is to cross-pollinate knowledge of metadata that's available that would be permitted in IMSC, but not required. It formally already is permitted I believe, so this would be informative only.