W3C home > Mailing lists > Public > www-svg@w3.org > May 2012

Re: using CSS syntax for presentation attributes

From: Cameron McCormack <cam@mcc.id.au>
Date: Thu, 24 May 2012 09:30:38 +1000
Message-ID: <4FBD731E.4050103@mcc.id.au>
To: www-svg@w3.org
Hi David,

My question was more about whether we should use CSS syntax exactly 
(apart from the required SVG differences -- scientific notation, 
unitless lengths, spaces in transform I think?), which would allow white 
space, CSS comments /* like this */, CSS escapes like th\000069s, and so 
on.  These things are current disallowed by the spec, but some 
implementations allow them (because they, rightfully, want to reuse 
their CSS parser).

ddailey@zoominternet.net:
> See, for example, http://cs.sru.edu/~ddailey/Guangzhou/video.htm (where
> no two browsers among the five (O,F,I,C,S) respond remotely similarly in
> attempting to apply SVG effects to HTML5 (as, for example, in video)

For filters, this will hopefully be taken care of by the separated out 
Filter Effects spec.  For clip-path, we probably need to write a spec 
for that.  I believe there is some interest in the FXTF to define masks 
so that they apply to HTML content too.

> At any rate, I think that instead of thinking of "Presentation" vs
> "semantics" -- a distinction that makes sense when one is styling
> hypertext, one needs to, maybe instead, think of Things versus Modifiers
> when semantics is graphical, as in SVG.
>
> In SVG things that modify include clippaths, filters, masks, gradients,
> animations, replicates, transforms etc. Not all are strictly
> "presentational" since in the case of SVG, semantics IS appearance.

Yes, that's how I like to approach the fact that we have two ways of 
specifying the styling of SVG content, the presentation attributes and 
style sheets.  I sometimes think of the former as required for the 
semantics, and the latter as ancillary presentation.  (Funny, since the 
word "presentation" is in the former.)

> Why not allow full-fledged SVG semantics (albeit geometric) to be
> bundled into predicate nominative phrases that modify Things? It is a
> syntactic device that hundreds of human languages have used over
> millenia, so there is no reason to think it should not work now. Trying
> to reinvent wheels (ala <canvas>) might be a bit redundant, superfluous,
> nugatory and silly ┬┐que no?

I unfortunately did not follow this paragraph. :)  Do you want the same 
kind of re-use that CSS affords but to consider it as an important part 
of the semantics of the document?
Received on Wednesday, 23 May 2012 23:31:14 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:51 GMT