Re: SVG and MathML in text/html

On Mar 15, 2008, at 9:40 AM, Doug Schepers wrote:

>
> Hi, Geoffrey-
>
> Geoffrey Sneddon wrote (on 3/15/08 12:12 PM):
>> On 15 Mar 2008, at 15:16, Ben Boyle wrote:
>>> These things are authoring nightmares. Just don't do it.  
>>> Consistency please!
>> You can't plain have consistency: you need to be inconsistent  
>> somewhere.
>> There are two options as I see it:
>> 1) We remain consistent with SVG/MathML elsewhere, and require it  
>> to be well-formed XML within HTML, and refuse to render them if  
>> they are ill-formed.
>> 2) We remain consistent with HTML, and have full non-draconian  
>> error-handling.
>> The two are mutually exclusive. As we are dealing with HTML, I'd  
>> much rather see HTML remain consistent (i.e., option 2) and not go  
>> against the basic principles it relies on (e.g., not dying on an  
>> error).
>
> No, there's a third way.  Have non-draconian error handling that  
> does not cause the parser to halt, but which does ensure that SVG  
> that wouldn't work in existing SVG UAs doesn't render in HTML5 UAs.   
> It would still be parsed, put into the DOM, but attributes with  
> unquoted values (and the rest of that element) aren't rendered.   
> That way SVG isn't fractured, and it doesn't break the error  
> recovery of HTML for non-SVG elements.

HTML has the feature of two serializations: a classic serialization  
that is error-tolerant, and an XML-based serialization that has  
draconian error handling. These have different costs and benefits,  
ultimately it is a benefit to HTML authors that they have a choice. I  
think SVG deserves to have this feature as well, there's no reason it  
should fall short of HTML in this regard. Supporting SVG inline in  
text/html seems like a good opportunity to add this feature to SVG.

Regards,
Maciej

Received on Sunday, 16 March 2008 05:13:13 UTC