- From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
- Date: Sat, 31 Jan 2009 19:23:01 +0100
- To: www-svg@w3.org, birtles@gmail.com
> <animate id="a" begin="5s"... /> > <animate id="b" begin="a.begin+2s" ... /> > <animate id="a" begin="click"... /> > <animate id="b" begin="a.begin+2s" ... /> Concerning the timing, this is similar, in both cases the begin from b is not resolved before a starts as that for a in the second case before or without a click-event (or an activation per hyperlink). But then it is 2s after the begin of a (note that negative offets can generate funny results for some combinations). In the first example before 5s the 7s is only a possible begin of b (everything might be possible and is not really relevant for SMIL animation), this gets 7s at 5s, but not before. However for implementations it is somehow simpler as to try to predict something, because they have to correct the predictions less often with this method. And those times are not really predictable before the active duration has started, because they may change, therefore any information about the 'next' begin can be only an estimate and is not a really reliable information. Whatever is received from this DOM method, this is useful in general only for the current already active animation or maybe for some very simple cases. If the end attribute is specified too (and depends on events or syncbases), the situation gets even less predictable.
Received on Saturday, 31 January 2009 18:40:56 UTC