RE: SVG 1.2 Animation question

This time with the attachment.

> Hello WG,
>
> A question from my colleague...
>
> After close review of some of the W3C SVG test slides, we believe there
> might be some inconsistency in the interpretation of a particular clause
> of the SMIL specification.
>
> The slide media-audio-207-t.svg seems to expect that an animation that has
> more "begin" times than "end" times should not restart if a begin time
> comes later than the latest end time. This behavior seems valid, as it is
> described by the SMIL spec in this algorithm :
>
> http://www.w3.org/TR/2005/REC-SMIL2-20051213/smil-timing.html#Timing-
> BeginEnd-LC-End
>
> However, this rule contradicts the expected behavior of the slide udom-
> svg-209-t.svg. In this slide (of which a simplified version is attached),
> scripting is used to start/stop an animation by adding begin and end times
> to it. The slide expects that an animation can be started and stopped
> twice by calling twice beginElement() and endElement(). Our understanding
> of the sequence of events is the following :
>
> - beginElement() is called - a "begin" time is added and the animation
> starts
> - endElement() is called - an "end" time is added and the animation stops
> - beginElement() is called - we are now in the same scenario as the one
> described in media-audio-207-t.svg : there are more begin times than end
> times and the latest begin is greater than the latest end. We believe the
> animation should not restart because of the algorithm mentioned above,
> though the slide's description expects it to restart.
>
> Thank you for your feedback on this issue.
>