SVG 1.2 Comment: Animation

13.1:

The attlist refs should be linked to where the relevant lists of attributes are 
defined.  In fact, this applies to all RelaxNG definitions in the SVG 1.2 
specification, so I'll stop saying it with the understanding that this last call 
comment applies to them all.

The SMIL specification needs to be clearly referenced normatively, again.  "The 
usual features" is too vague.

What the "preserveAspectRatio" attribute does is not defined anywhere that I can 
see.

I can't find where the SVGMediaElement interface (which SVGAnimationElement is 
defined to inherit from) is defined.

13.1.1:

It's not clear to me whether the text means that there is a transition element 
in SMIL already.  If there is, why is it not simply being reused here?  If there 
is not, the text should be reworded to not imply that the transition element is 
a SMIL transition effect.

13.2:

The "SMIL media elements" link should link to the relevant section, not to the 
table of contents of the SMIL specification.  It seems that it tries to do that, 
but gets the URI wrong and gets redirected.

There's a missing space before "define".

This section should probably precede the definitions of video and audio if it's 
going to add normative information about them...  If it's simply informative, 
it's not clear to me why it's here at all; surely the definitions of video and 
audio already say that they define their own timelines, etc?  If they don't, 
they should.

13.3:

Again, this text should be in the sections defining the relevant elements.

Again, the link to SMIL is using the wrong URI for the SMIL specification.

13.6:

The behavior of the ElementTimeControl is totally unspecified.  What does 
beginElement() do?  What should it do if the element is already "running"?  What 
if it's paused?  What does endElement() do?  What does endElementAt() do?  What 
are the units for the offsets?  What corresponds to an offset of 0?  What do 
negative offsets mean?  Can these methods throw exceptions?  If so, which ones 
and under what circumstances?

I would like to request that this section of the specification actually be 
written and submitted for public review.  I don't consider it possible to make 
Last Call comments on this section at this time, since the content that should 
be here simply doesn't exist....

13.7:

I don't understand what this section is trying to say.  In particular, the 
"which behaves as if stopPropagation() and preventDefault() have both been 
invoked" language makes no sense... A listener that behaves as if both of those 
have happened would most likely simply not fire.

-Boris

Received on Thursday, 25 November 2004 05:38:24 UTC