- From: Erik Hodge <ehodge@real.com>
- Date: Thu, 07 Aug 2003 10:33:20 -0700
- To: "Glenn A. Adams" <glenn@xfsi.com>, Johnb@screen.subtitling.com
- Cc: public-tt@w3.org
Received on Thursday, 7 August 2003 13:32:02 UTC
Just a quick comment related to the timing semantics of John's example: At 10:00 AM 8/7/2003 -0400, Glenn A. Adams wrote: >I.e. - using example (crudely hatcheted from GA examples) > ><seq> ><ref id="T27" dur="3s"/> ><ref id="T28" begin="T27.end " dur="3s"/> ></seq> > >If we support this idiom, then we would introduce a new element type cue >instead of using the SMIL ref element type. It would take a select >attribute that would use XPointer or XPath to select the content. Also, >your use of the begin attribute above is redundant, since the semantics of >seq imply that subsequent children follow immediately in time. It's not only redundant, but syncbase-value begins are disallowed (among others) in children of a seq. (See: http://www.w3.org/TR/smil20/smil-timing.html#Timing-SeqSyntax )
Received on Thursday, 7 August 2003 13:32:02 UTC