Re: Suggested revised text for HTML/XML report intro

On Tue, 16 Aug 2011 08:45:12 +0200, John Cowan <>  
> Noah Mendelsohn scripsit:
>> Hmm. First of all, I certainly would be glad to find out that XML5
>> could "work", and be widely accepted; it's just the sort of direction
>> I was hoping the task force would find to be practical.
> I continue to believe that XML5 is not practical because of the
> generality of XML.  Malformed HTML can be repaired on the principle of
> "change it to be well-formed HTML that browsers will treat identically
> to the original."  No equivalent principle exists for XML, because there
> are no privileged XML-processing applications; the canonical application
> for XML depends on the variety of XML, and may not even exist for many
> varieties.

Your principle is wrong. HTML is not repaired; processing just does not  
stop. Doing the same for XML is fairly trivial. (As is doing it for CSS,  
WebVTT, event streams, etc.)

Anne van Kesteren

Received on Tuesday, 16 August 2011 10:33:35 UTC