- From: Craig Northway <craign@cisra.canon.com.au>
- Date: Sat, 30 Apr 2005 08:51:59 +1000
- To: Bjoern Hoehrmann <derhoermi@gmx.net>
- Cc: www-svg@w3.org
Bjoern, Bjoern Hoehrmann wrote: >* Craig Northway wrote: > > >>The working group have discussed this again. This will be clearly >>dissallowed. The UA must not reinsert elements when the document seeks >>backwards. If you are satisfied with this the text will now read: >> >> Seeking backwards in the timeline must not re-insert the discarded >> elements. Discarded elements are intended to be completely removed >> from memory. So, authors are encouraged to set the 'playbackOrder' >> attribute to forwardOnly when using the 'discard' element. >> >> > >This change alone does not satisfy me, the draft would then contradict >the definition of the element which states that discarding an element >is equivalent to calling removeChild() which would no longer be true as >user agents are not prohibed from re-inserting nodes removed via that >method when seeking. > Excuse me? They are explicitly prohibited from re-inserting nodes. This method now works exactly like calling removeChild/Node() from a script. If the element is removed and you seek backwards past its removal time, it does not reappear. I quote from my earlier reply: *Seeking backwards in the timeline must not re-insert the discarded elements.* How clear can this be! > I suggest to add a new section or sub-section that >clearly defines requirements for seeking which would then be linked from >the discard element definition (as I requested). > > Yes, we could add a reference to SMIL from the discard element prose. Regards, Craig
Received on Friday, 29 April 2005 22:52:12 UTC