It’s just data


Ian Hickson: I have removed SVG support from the HTML parser specification as per your request. (I have left MathML support in the specification.)  Hat tip: Shelley Powers.

I wonder how fast the SVG WG will come up with the proposal they mentioned in [link]. (Where is your text-to-link script? I’m impressed that it recognizes dots and question marks at the end.)

Posted by Anne van Kesteren at

Oh well. That was fun while it lasted.

I suspect that the Nth iteration of the thoroughly inane @alt debate has left Ian Hickson in a foul mood. Perhaps he should offer to remove <img> from the Spec, pending a non-nutball proposal from those arguing that @alt must remain mandatory.

Posted by Jacques Distler at

Oooh, removing <img> is not a bad idea...

And yeah, removing <svg> sucks. But it’s hard to argue against other working groups in the W3C. The text is still there in the spec, it’s just commented out. If they don’t come up with something to replace it, I’ll put it back, don’t worry. (And in the extreme case where the W3C process ends up meaning we can’t put it back... well, we still have the WHATWG as an independent entity to resolve that kind of issue. But I don’t think we’ll have to go there.)

Posted by Ian Hickson at

the Nth iteration of the thoroughly inane @alt debate

Damn you, that’s an hour of my life I’ll never get back.

Posted by Mark at

I apologize as I guess I am misconstruing Andrew’s comment that SVG in text/html “is incompatible with all implementations to date.” What is Amaya 10 doing that allows rendering of SVG in text/html as I found to be the case for a page I tested with Amaya 10? Maybe Amaya does not count as an implementation. Still, Amaya demonstrates that SVG can be rendered in text/html, with or without namespace declarations.

Posted by Dana Lee Ling at

I am too looking forward to seeing the SVG WG’s proposal on how to do this. I believe the competence on the HTML parsing algorithm and deployed base of browsers and HTML documents to be higher within the HTML WG than in the SVG WG and expect the SVG WG’s proposal to be completely incompatible with the HTML parsing algorithm. I hope to be proven wrong, though.

Posted by Asbjørn Ulsberg at

Sam Ruby: HTML5 -= SVG


Excerpt from at

Where is your text-to-link script? I’m impressed that it recognizes dots and question marks at the end.

Code -> mombo ->  Search for # naked urls become hypertext links.

Posted by Sam Ruby at

Of the options on the table right now, removing SVG from text/html seems the most sane. But then again the argument that text/html can never be truly SVG-compatible on the basis of media type and/or byte-compatibility seems a little defeatist. If the SVG can come up with HTML5-style parsing, great (but that looks very hard). There’s also at least one possible way forward that doesn’t seem to have been explored.

(X)HTML 4 has @profile as an extensibility point, which is put to good use in GRDDL by offering a path to a mapping between the HTML document and the RDF model. The mapping itself is typically expressed via an XSLT transformation (any easy win when the input’s XHTML).

What if HTML5 used @profile to indicate a means by which the HTML+SVG could be interpreted directly as SVG? The interpretation could happen server-side (say HTML+SVG - Tidy - XSLT - application/xml+svg) or client-side, if an analog to the “GRDDL-aware agent” was defined.

Posted by Danny at

Sam Ruby: HTML5 -= SVG

[link] [comments]...

Excerpt from programming: what's new online at

Just to be clear, the SVG WG is very interested in getting SVG into HTML.  I personally have been pushing for this for years.  Hixie told me back in November that it wasn’t a priority for him and that it would be more likely for HTML5.1, so I put it on the back burner.  Because of popular demand, it seems that this moved up his priority list.

All the SVG WG is asking for is a few weeks to evaluate options, become conversant in the HTML5 parser and tree construction algorithms, and prepare a proposal that meets SVG’s needs within the constraints of HTML5. 

I’m pleased to see that others have also been pushing for this, and we will be working on a solution that we hope will best serve the Web community and be mutually beneficial to the HTML and SVG WGs.  Given the current point in the browser releases cycles, and the anticipated schedule for the HTML5 specification itself, this short delay shouldn’t have any effect on users, authors, or implementors.  Suggestions or help with this can be sent to the SVG WG.

Posted by Doug Schepers at

SVG News Digest: 2008-04-16

I haven’t really given a good ‘SVG News Digest’ in well over a year, but there was enough recent news that I thought I should post a little bit about what’s going on in the Scalable Vector Graphics world. Fair warning: This blog post is long, I...

Excerpt from Something Witty Goes Here at

Add your comment