W3C home > Mailing lists > Public > www-svg@w3.org > January 2007

Re: SVG1.1 TestSuite 2006-12-13 animate-elem-65 min

From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
Date: Fri, 26 Jan 2007 12:17:52 +0100
To: Andreas Neumann <neumann@karto.baug.ethz.ch>
Cc: www-svg@w3.org
Message-Id: <200701261217.52501.Dr.O.Hoffmann@gmx.de>

Hello,

> Regarding the subtests 6 and 7: I understand that it will be useful for
> implementors to add your note that the sync-base values will be
> different from the visible part of the "active duration" or whatever it
> will be finally called. Regarding the clash in the meaning of "active
> duration" between the official SMIL definition and the meaning of
> "active duration" in the description of our subtests 6 and 7 I don't yet
> have a good solution. Do you have a suggested wording to resolve that
> problem?
>
> Would "visible part of the active duration" instead of just "active
> duration" work in your opinion?
>
> Andreas
>

Well, this situation is very specific (or academic) - therefore I think
there is no need for a specific expression, especially to be pedantic:
for fill freeze there is no visible change at the end of repeatCount ;o)

The wording in these examples with min and max 
is in general a little bit bumby and imprecise:
(0s <= t < 5s) is a point t in a time interval, not a duration.
(5s <= t < 10s) is related to another time interval with the same
duration. The duration is in both cases 5s - a number and
a time unit, no interval. There is no bijective relation between 
time intervals and durations.
Therefore maybe something like this is more elegant:

"
On the sixth row, the <set> animation has a dur value of 1s, an end
value of 2s, a repeatCount of 5 and an 8s min attribute value. 
The active duration resulting from dur, end and repeatCount 
would be 2s, because this is less than the min value (2s < 8s) 
the active duration has to be corrected to 8s. Anyway because the 
fill attribute is set to 'remove' on the <set> animation, this 
remove is applied at 5s, the end of the repeatCount. 
Note, that if the end of active duration would have been used as a
syncbase-value for another animation, the corrected end event at 
(begin +) min = 8s has to be used. 

On the seventh row, the <set> animation has a dur value of 1s, an end
value of 2s, a repeatCount of 5 and an 8s min attribute value. 
The active duration resulting from dur, end and repeatCount 
would be 2s, because this is less than the min value (2s < 8s) 
the active duration has to be corrected to 8s. Because the fill attribute 
is set to 'freeze' on the <set> animation, the animation is frozen at
5s, the end of the repeatCount, the <set> applies indefinitely.
Note, that if the end of active duration would have been used as a
syncbase-value for another animation, the corrected end event at 
(begin +) min = 8s has to be used. 
"

It is maybe useful to avoid the confusion about time intervals and
duration in the other examples too, but this is a matter of taste
or precision, because in the other examples there is only a low
chance for misunderstandings, if it is not improved.


Olaf
Received on Friday, 26 January 2007 11:21:15 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:36 GMT