Re: Proposal: Change end attribute behaviour

Brian Birtles:
>
> All of that aside. I'm genuinely interested to hear feedback about the
> specific proposal. Has anyone else been tripped up by this end
> instance condition? (I have been, too many times to count, and I
> implemented it!) Does anyone rely on this behaviour?
>

Well, I had this problem several times, with the preliminary result,
that I added always 'indefinite' to the end list, if I do not need
an ultimative end. Than I discovered, that this is for some contructions 
not necessary, because sometimes the end is removed again - 
of course combined with typical bugs of current viewers it is a lot of fun 
for authors to get it right both for current viewers and for future viewers
with a proper implementation as well, because oviously there is no 
interest, to check the behaviour for already existing documents again 
and again ;o)
The still open question is how to really end something for
current viewers with bugs, because often they ignore such a
rule, therefore either I did not publish such applications, that need
this functionality of the end attribute or I used a workaround 
that blows up the source code with several more animated
elements to get a similar effect, what really looks nasty.
Therefore I'm excited about the time, several viewers will
have implemented this properly to be able to publish some
applications, that need the functionality of an animation with
an ultimative end preventing restarts. Time containers are
currently not available too in SVG, therefore there are no
simple alternatives to work around the current bugs.

Obviously these are complex rules, but to manage begin- and 
end-time-lists in general is complex. Not sure, if there is a simpler
method to do it without restricting something. 
It may take a longer time and several SMIL experts (do they still
exist?) to be sure, that a change of this complex system does
not create unintended new problems, gaps and inconsistencies.
My experience in the past with SMIL and the SYMM group is,
that it was even difficult to fix existing inconsistencies because
there was not much motivation to manipulate a relatively good
thought trough system. To invent a new one is even more
difficult. To get a simpler system with less options for authors 
is not very interesting (at least for me).


Olaf

Received on Friday, 25 March 2011 14:26:10 UTC