Re: Why not XHTML+RDF? was Re: Links are links

Tim Bray wrote:
> Paul Prescod wrote:
> 
>> Beyond XHTML, I think it is necessary to be able to say:
>>
>> <cv:doctor
>>    cv:doctorate="http://.../university"
>>    cv:specialty="http://.../"
>>    cv:homepage="http://.../"/>
>>
>> Basically I should get the benefits of standardized link recognition 
>> without contorting my vocabulary around XLink at all.
> 
> 
> OK, but let's think about user interface. 

"CSS3 module: Basic User Interface"

http://www.w3.org/TR/css3-ui/

> When I click on the cv:doctor 
> element I'd like to have a menu pop up saying 
> "Doctorate/Specialty/Homepage".   Where do I put that kind of per-link 
> labeling metadata? 

Admittedly the spec above is new and incomplete. But I am confident it 
points in the right direction. User interface is specified separately 
from the data for all of the usual reasons that either of us could 
repeat ad nauseum.

Also, it may be convenient to put some kinds of annotative metadata into 
The Schema or a new kind of Link Schema (one or the other is desperately 
needed IMO).

> ...  Then generalize this question for *any* kind of 
> information that you might want to provide per-link-end.  Also think 
> about internationalization.

Internationalization is another strong argument for moving menu titles 
out of the data and into the stylesheet. I would rather provide a Thai 
stylesheet than Thai-annotated data.

>   These were the considerations that drove 
> XLink to use subelements for link ends.  -Tim

In *my* markup language it is my perogative to weigh the costs and 
benefits of elements vs. attributes for myself. My choice should not be 
constrained by XLink or RDF M&S.

  Paul Prescod

Received on Wednesday, 2 October 2002 16:37:31 UTC