- From: David Singer <singer@apple.com>
- Date: Thu, 22 Dec 2011 10:35:57 -0800
- To: Ralph Giles <giles@mozilla.com>
- Cc: public-texttracks@w3.org
On Dec 22, 2011, at 10:33 , Ralph Giles wrote: > On Thu Dec 22 10:18:24 2011, David Singer wrote: > >> These categories probably overlap. I'm thinking of examples like >> -- "looks better if these style-sheets are used" >> -- "has an overall language of en-US" >> -- "is intended for this 'kind' in HTML" >> -- "when used with the expected mpeg-2 transport stream, <this> VTT timestamp is (maps to) <that> MPEG-2 clock reference" >> ... >> Do we have proposals, or are we at the use-case and data-collection stage? > > I hadn't thought of your last two examples, so I'd say we're in the > data-collection phase. > > Metadata is always open-ended, so I think we should define an > extensible mechanism and pick some simple categories to standardize for > initial interoperability. > > I would add title, author, date, translator, etc. labels. > > I really like the timecode offset suggestion. How would software handle > this? Does the CMS Re-timestamp the webvtt file when it's deployed in > conjunction with that video? Are videos referenced by string? SHA-1? > Would user agents parse and apply this on the fly? I am not sure. But with both VTT and TS timestamps not zero-based, and the possibility of 'tuning ing' to a TS stream without a clear or known beginning, it seems worthy of thought. > > -r > David Singer Multimedia and Software Standards, Apple Inc.
Received on Thursday, 22 December 2011 18:36:42 UTC