Re: [1.2T-LC] 16.2.9 values attribute 'extended syntax' (ISSUE-2084)

Hi, Dr. Olaf-

Thanks for your comment.  I've recorded this as ISSUE-2084 in our
Tracker. The SVG WG will discuss this and get back to you soon.

Regards-
-Doug

Dr. Olaf Hoffmann wrote (on 10/1/08 10:00 AM):
> Hello SVG WG,
> 
> it is surprising, that there is already a lot of SVGT1.2 content with invalid
> values for the 'values' attribute, because such 'extended syntax' is nonsense
> both for SMIL and for SVG1.1 too.
> For me this indicates more simple errors from authors or editors, especially
> because the adobe plugin, for several years the most used viewer for animated
> SVG content has a quite different '(error) management' for such wrong syntax
> (if it is wrong and does not specify an allowed empty value, it indicates an
> error and assumes one more value than the number of semicola. If an empty
> value is possible, an empty value is correctly interpreted as empty value).
> Every author testing content with the adobe plugin should have noted
> the error and should have already fixed it before publication.
> Typically if I find errors in my documents or scripts, I simply fix them and
> do not expect, that the specification is modified to fix my own errors. 
> And even more, no one can expect, that the behaviour of already published
> versions of viewers can be modified, therefore this 'extended syntax' should
> never be used to ensure better backwards compatibility with older viewers,
> therefore such a superfluous trailing semicolon needs to be fixed anyway to
> ensure a predictable behaviour.
> Especially there is no benefit for authors or users from this 'extended
> syntax'. 
> The opposite is the case with something like "/a.txt; ; /b.txt; ;" to get the
> desired effect of an  empty list item only for SVGT1.2.
> For implementors of SVG1.1, SVGT1.2 and SMIL it gets even worse, because 
> they have to implement it differently for SVGT1.2 without any advantage for
> anyone.
> 
> My suggestion is to skip this SMIL and backwards incompatibility completely
> and to help authors to fix their documents, if it is known, which authors
> produce so much erratic content without testing it.
> If such content is already mentioned in the specification, some editor should
> know at least some of these authors and how much invalid content it is. Maybe
> references would be useful too for others interested in helping those authors.
> This is more friendly for those authors as to brand or to stigmatise them to
> be guilty to corrupt the well thought out SMIL syntax.
> 
> 
> Olaf
>

Received on Tuesday, 7 October 2008 13:31:57 UTC