- From: David Ronca <dronca@netflix.com>
- Date: Thu, 31 Jan 2013 01:33:09 +0000
- To: "public-tt@w3.org" <public-tt@w3.org>
> ok, but now i'm back to why not just use tts:extent, since the AR of the > root container is nothing more than extent(width) / extent(height)? Because tts:extent specifies exact pixel container area that the client must use. If a video starts at 480p, and the tts:extent="640,480", then when the video adaptive-switches to 720p, and/or 1080p, the caption area (as a percentage of the video) will be reduced, and move towards the top-left. We need a caption presentation model that is completely independent of the video. All positioning must be relative, and the client must know whether to present the captions in a 4:3 area of the video or a 16:9 area. David
Received on Thursday, 31 January 2013 01:33:37 UTC