Re: Fwd: SVG animateMotion specification clarification request

Ken Stacey
> Yet another part of SMIL I dislike.
>

SMIL is much simpler. It has no animateTransform and
animateMotion has no additional rotation. 
These complications are introduced in SVG, not in SMIL.
SMIL has a straight forward sandwich model, if different
animations apply to the same attribute. This is not
the case for animateTransform and animateMotion,
because animateTransform animates an attribute and
animateMotion provides the functionality to move objects.

Of course, these additional features are pretty useful - and 
in combination will be always complex to define.
Because the complexity here is not within SMIL, but has
its applications in SVG, I think, it is a good idea to separate 
the SMIL model from specific SVG complications.
This has the advantage, that implementors can
implement SMIL animation without many specific
rules almost independent from the animation element.
This reduces the risk of bugs.
This is simple for authors as well, because there are
always the same rules, the animation follows and
not a huge amount of exceptions, one has to keep in
mind - else one really has to use always g-elements
to get the intended effect, either because some specific
rules are implemented wrong or not at all in some
viewers or one does not have in mind all specific
rules introduced only in SVG, which have nothing
to do with the generic animation concept.

Well, in practical use cases, for example a planetary
system with either the sun or the earth or the moon
as the center, I anyway have to use additional
g-elements to get the intended effect, because 
which transformations apply to which elements depends
on what to put back into the center with an inverse transformation.

Olaf

Received on Thursday, 31 March 2011 10:18:28 UTC