Re: Updated RDFa Core Editor's Draft available

Hey Shane,

On Apr 9, 2010, at 18:52 , Shane McCarron wrote:

[snip]

> 
>>   - the issue on default namespace^h^h^h^h^h^h^h^hvocabulary URI being the XHTML one is indeed explicitly flagged as a question (as you did).
>> 
>> maybe other participants of yesterday's discussion can help out here...
>> 
>> - I see you use URIorCURIE as a type for @about and you say 'term or URIorCURIE' for, say, @property. I also see that you have separated term processing from the general CURIE processing. I must admit I preferred the previous version where term processing was part of the CURIE processing; I think it is better for implementers to have everything in on place, so to say. (But I can live with the current version, if the majority of the group is fine with this.)   
> 
> The group asked me to make this change whilst you were away.  It makes the datatypes more explicit.

Ah. Actually, thinking a bit more about this (term|URIorCURIE) approach I actually like it. It forces me to redo my implementation here and there but, well, that goes with the territory:-)

>  
>> But...
>> 
>>  - Is it stated somewhere that the term processing must precede the CURIE processing (I have not found it)
>>  
> 
> I believe that the processing order is defined in the datatype eBNF ( ( term | URIorCURIE )+ indicates term is evaluated first, as eBNF is evaluated left to right), but I can put it in the prose as well.  Also, I note that this eBNF didn't really make it into section 5 like I meant it to, so that section (which is prose) needs to be made more explicit.  I will take care of it.

Thanks.

> 
>>  - What happens with relative URI-s? At the moment, if I follow section 8 for, say, @about="./something" then this will be rejected because the the text says that we prohibit a CURIE without a colon. Where is that taken care of?   
> 
> Section 8 describes CURIE processing.  For a datatype of URIorCURIE, since that is not a CURIE it would be processed as a URI.  This is defined in section 6.4 - and in particular in section 6.4.2.

Yep, I see it now.


> 
>>  - And... if we do take care of that, would that mean that I can use @property="./something", ie, a relative URI for a @property? I thought we disallowed that...
>>  
> 
> Well...  since @property has a datatype of (term | URIorCURIE)+, term gets evaluated first.  Your relative URI in this example does not match the production for term, (see section 6.4.3) so it is evaluated as a URIorCURIE.  It does not match the production rules for a CURIE in RDFa, so it is treated as a (relative) URI.  This is something that got changed whilst you were away.

Aha! That is a departure from what we had before, where @property/@rel could not have relative URIs, if I remember well. But if it works spec-wise (and it seems to work indeed), then there is no reason to disallow it (though, I must admit, I do not see to much use for it in practice, but that is another matter).

Thanks!

Ivan


> 
>> Ivan
>> 
>> On Apr 8, 2010, at 19:41 , Shane McCarron wrote:
>> 
>>  
>>> As per my action items today, I have updated the RDFa Core spec and pushed a new Editor's Draft.  This is available via our drafts page at http://www.w3.org/2010/02/rdfa/drafts
>>> 
>>> I have also done some work on our publishing toolchain, so we are getting diff marks, postscript, and PDF versions of specs automatically now.
>>> 
>>> Please take a close look at this document, as I expect it is in the nearly final form for FPWD.
>>> 
>>> -- 
>>> Shane P. McCarron                          Phone: +1 763 786-8160 x120
>>> Managing Director                            Fax: +1 763 786-8180
>>> ApTest Minnesota                            Inet: shane@aptest.com
>>> 
>>> 
>>> 
>>>    
>> 
>> 
>> ----
>> Ivan Herman, W3C Semantic Web Activity Lead
>> Home: http://www.w3.org/People/Ivan/
>> mobile: +31-641044153
>> PGP Key: http://www.ivan-herman.net/pgpkey.html
>> FOAF: http://www.ivan-herman.net/foaf.rdf
>> 
>> 
>> 
>> 
>> 
>>  
> 
> -- 
> Shane P. McCarron                          Phone: +1 763 786-8160 x120
> Managing Director                            Fax: +1 763 786-8180
> ApTest Minnesota                            Inet: shane@aptest.com
> 
> 


----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf