Re: SVG in text/html

If browsers can provide an 'export DOM as XML' function as part of its
standard functionality, then my concern melts away with the proposal as it
stands (stood), since round-tripping then becomes feasible.  Has this been
considered by both SVG and HTML groups?

I seem to remember a discussion of such a mechanism as a requirement on
HTML5 User Agents that was rejected a couple weeks back...


On 4/15/08, Maciej Stachowiak <> wrote:
> On Apr 15, 2008, at 4:06 AM, Andrew Emmons wrote:
> Hello HTML WG,
> While it is encouraging that SVG in text/html is being discussed[1], we
> feel that the proposal tries to change the SVG syntax in a way that is
> incompatible with all implementations to date.  We feel that such change is
> outside of the HTML Working Group's charter and request that it be removed
> from the text/html specification pending further research into proposed
> solutions, in order to avoid premature implementation.
> As an SVG and HTML implementor, I would find any reasonable choice for SVG
> in text/html about equally burdensome. Any approach by necessity would be
> "incompatible with all implementations to date", since no implementation
> supports SVG in text/html.
> We are happy to make rapid progress on allowing SVG in text/html while
> maximizing compatibility with the wide range of deployed SVG authoring tools
> and renderers.
> Authoring tools would presumably also need changes to support any kind of
> content in text/html. (If the idea is that authors would manually extract
> SVG snippets, I believe the spec also supports the cut & paste use case, but
> will not prevent the author from making the SVG content unusable to XML
> tools if edited after insertion into text/html content. To support safety in
> the latter case, the author would have to use an HTML5 parser to extract
> inline SVG from HTML.)
> Regards,
> Maciej

Received on Tuesday, 15 April 2008 18:23:33 UTC