Re: More on SVG within HTML pages

Aryeh Gregor wrote:
> Why is it that well-formedness isn't required for SVGs in HTML pages
> to begin with?  If it's not well-formed, you could just say it doesn't
> display.  In HTML, well-formedness requirements would make it
> impossible to use legacy content, but that doesn't apply to SVG.  All
> other SVG applications require well-formedness (right?) -- if HTML
> doesn't, then it will be impossible to reliably copy SVGs from HTML
> pages to other applications, creating interoperability problems.
> While requiring well-formedness has its downsides, I can't see how
> they outweigh the interop problems this would create.
Aryeh, I agree with you -- I don't think the well formed rule should be 
relaxed, just because the SVG (or MathML) is dropped into an HTML page.

The issue is, I believe, and I can only speak as an observer on this 
one, that HTML elements can be embedded within the SVG, and those HTML 
elements may not be valid. The case could be in error, closing tags 
omitted, attributes unquoted.

Personally, I think this is a stretch. I'd rather well formed XML be the 
rule, and as long as the XML is well formed, it could all be included in 
the SVG element.


Received on Sunday, 6 September 2009 21:28:54 UTC