Re: &foo= in attribute values (and why defining conformance matters)

On 6/17/09 10:09 AM, Michael(tm) Smith wrote:
> Rob Sayre<rsayre@mozilla.com>, 2009-06-12 18:49 -0400:
>
>    
>>   On 6/12/09 6:22 PM, Ian Hickson wrote:
>>      
>>> It defines what QA tools like conformance checkers
>>>        
>>   Oh, I forgot to mention: "conformance checker" is an incredibly vapid piece
>>   of newspeak. They're only checking "conformance" because you classified
>>   large parts of the working, interoperable grammar as such.
>>      
>
> Do you mean that you believe "conformance checker" is absolutely a
> misleading term -- and that you think there's some term other than
> "conformance checker" that should instead be used to describe
> validator.nu or other such tools? Or did you just mean that the
> term is misleading in the specific instance you cite above?
>    

The term is misleading because there are conformance requirements in the 
document that only manifest themselves in conformance checkers. It's 
circular.

>  >  I pose a serious question: what is the real benefit of making unescaped
>  >  ampersands non-conformant? (Of making anything "non-conformant"?)
>
>  It defines what QA tools like conformance checkers should highlight as
>  problems, as an aid to authors who wish to catch mistakes they did not
>  intend. That's it.


No one would complain about a lint tool that warns upon encountering a "<dvi>" element. The problem with these "conformance checkers" is that they are used to enforce a very prescriptive, centralized, and inconsistent view of the way markup "should be" written. Perhaps the term should be "ideology checker" or "loyalty checker" would be more appropriate.

- Rob