Re: Fwd: Put <> around URIs. Re: draft CURIE draft feedback

Ben Adida wrote:
>
> Shane McCarron wrote:
>> This has nothing to do with href.
>
> Really? Did we indeed vote to disallow:
>
> <div href="[_:foo]">
>
> ?
We did not change the datatype of @href nor of @src, so the above would 
be illegal in XHTML+RDFa.  I assume that's the question you were asking.
>
> I think I would be okay if we had, but even so, having different 
> syntax for URIs in @resource and @href would be quite confusing:
>
> <a href="http://example.org" resource="<http://example.org/alternate>">
>
> Changing the way URIs are written in HTML... that's some pretty risky 
> stuff. I'm *much* more in favor of thinking about better ways to 
> represent CURIEs, even if they require a few more characters.
Hmm.... I would characterize it as "Defining the way non-CURIEs are 
referenced in @about and @resource", not "Changing the way URIs are 
written in HTML."  Tantamount to the same thing though.  But those are 
new attributes, and we can populate them however we want.  I think what 
this really comes down to is:  Do we prefer that people use CURIEs or 
that people use URIs in these attributes?  I suspect the answer is 
URIs.  In which case, changing the syntax of URIs makes little sense.  
The only use case I know of for using a CURIE in @about is when 
referencing a bnode.  If that's true, then...  Maybe it is simpler to 
just say that in this edge case you need to use this weird syntax for 
your CURIEs and be done with it.
>
> (To make things worse, I'm pretty sure you have to XML-escape the < 
> and > characters inside attribute values if you want things to 
> validate, and that gets quite ugly.)
Excellent point.   Brackets would work tho....  That was the other 
syntax Tim proposed. 

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com

Received on Monday, 14 April 2008 21:11:08 UTC