Re: Publishing a new draft (HTML5+RDFa)

On Jul 31, 2009, at 3:34 PM, Ian Hickson wrote:

> On Fri, 31 Jul 2009, Julian Reschke wrote:
>>
>> That being said, MANY people deal just fine with prefix-based
>> indirection, and at least one indirection mechanisms we already  
>> have in
>> HTML (class names -> CSS) is *far* more complicated.
>
> Class names and CSS are also a source of great author confusion.

Class names as CSS selectors are somewhat confusing to authors, but in  
my experience less so than namespace prefixes. I think there are a few  
reasons: (1) indirectly binding style is less surprising to people's  
expectations than indirectly binding meaning/identity, since the  
indirection links two separate things rather than modifying meaning;  
(2) it's harder to make the mistake where you think the class name for  
actually *is* the style, than to make the mistake where you think the  
prefix *is* the namespace, in fact I have not heard of anyone making  
this particular mistake; (3) the effects of incorrect style binding  
are likely to be more immediately visible in the context where they  
are authored; (4) the binding rules of CSS are simpler and so less  
likely to lead to unnoticed copy/paste errors in markup.

Regards,
Maciej