- From: Dave Singer <singer@apple.com>
- Date: Fri, 21 Feb 2003 13:42:41 -0800
- To: Jean-Claude Dufourd <Jean-Claude.Dufourd@enst.fr>, public-tt@w3.org
At 11:57 +0100 2/21/03, Jean-Claude Dufourd wrote: >Dear Dave, > >Dave Singer wrote: >>Well, that appears to be what happens, but alas, is not. What >>you'd like to know is "play from 0 to 10 minutes", "play from 12 to >>15 minutes", and so on. What you actually get is a continuous >>program where those edits have already occurred. >> >>Happily, the timecodes get edited with the video and audio; that's >>why I say that it is a stream. If you look at the timecodes going >>by, you can work out what edits happened. If there is a text tag >>with a time-code value of "13 minutes" as its trigger, that will >>play 12 minutes in (minute 11 being missing). If the trigger said >>"11 minutes" then that text element will never display as that >>timecode was edited out. > >You are describing not requirements, but a solution/implementation. >The fact that all current broadcast implementations look the same >should not force us to adopt the very same solution. We should go >back to requirements, in order to be able to unify this with the >needs of the PC world, as well as the different needs of the mobile >world. No, I am describing requirements. The broadcast world needs to be able to express the link of the start time of an element in a composition to a specific time-code in another stream of that composition. -- David Singer Apple Computer/QuickTime
Received on Friday, 21 February 2003 16:53:44 UTC