Re: SVG as Root (was: SVG Feedback on HTML5 SVG Proposal)

On Tue, 10 Mar 2009, Doug Schepers wrote:
> 
> [...] I don't have any grand suggestion for fixing the case where SVG is 
> served with the wrong MIME Type, or where authors got lazy or made some 
> other mistake.  Does your current proposal deal with this, and if so, 
> how?

It's not clear to me that there's anything to deal with. That is to say, 
it is dealt with in the same manner as people publishing SVG files as 
text/plain (which is likely more common), or publishing CSS as text/html, 
or people writing bad HTML markup in text/html. The rules are 
well-defined, but the result is unlikely to be what the author intended. 
When the author finds the page broken in some way (e.g. rendering at the 
wrong size), then they go and fix the problem.


> If by "this is an error and should not be supported" you mean that the 
> file simply wouldn't render anything... hmmm, ugly, but I might agree 
> with that.

I mean that it should simply do whatever the rules for conforming content 
end up doing, and we shouldn't worry about optimising the rules to 
gracefully handle this error in a "nice" way.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Wednesday, 11 March 2009 01:20:18 UTC