Re: Cleaning House

On May 4, 2007, at 2:19 AM, Julian Reschke wrote:

>
> Lachlan Hunt wrote:
>> Julian Reschke wrote:
>>> Speaking of which, I'm not really sure why the UA parsing  
>>> requirements with respect to processing invalid content need to  
>>> be normative at all. As far as I can tell, there's really no need  
>>> to make them normative.
>> If you don't make the processing requirements normative, then UAs  
>> can just implement whatever they like and claim conformance.  That  
>> doesn't help anyone at all, it just leaves us with the same  
>> situation we're in now.  We're trying to fix the problem, not just  
>> ignore it.
>
> If the processing requirements are written and agreed upon, and  
> implemented by Apple/MS/Mozilla/Opera, it really makes no  
> difference in practice what normative status they have.

If you think it makes no difference, then perhaps you could let those  
who think it does make a difference get their preference.

I think it makes a difference because it only makes sense to write a  
conformance test suite for normative conformance requirements; you  
can't test an informative note. To actually achieve interoperability  
will require an extensive test suite which requires making processing  
requirements normative for some conformance class.

Regards,
Maciej

Received on Friday, 4 May 2007 09:33:20 UTC