W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > August 2011

Re: Bug in RDF Distiller

From: Ivan Herman <ivan@w3.org>
Date: Tue, 23 Aug 2011 10:41:50 +0200
Cc: Jeni Tennison <jeni@jenitennison.com>, W3C RDFWA WG <public-rdfa-wg@w3.org>
Message-Id: <B49AFAF2-B5A1-43DC-A880-1F86ECFD8BB8@w3.org>
To: Gregg Kellogg <gregg@kellogg-assoc.com>

On Aug 22, 2011, at 21:29 , Gregg Kellogg wrote:

> I'm CCing the RDFa WG, as this suggests a possible issue with precedence of terms from the default profile and @vocab, which I believe has been discussed before.
> 

Actually... yes, that is an issue but, in this particular case, there is another one. I think we have an action on Nathan that he would review the default HTML 'terms' (like 'first') to see whether these really make sense in a semantic sense. That action should determine which of these would be part of the default set of terms in HTML.

Ivan



>> Hi Gregg,
>> 
>> I'm trying out mapping some microdata examples into RDFa, and came up with this RDFa to represent a list:
>> 
>>   <div vocab="#" typeof>
>>    <p>Flavors in my favorite ice cream:</p>
>>    <div rel="flavor">
>>     <ul vocab="http://www.w3.org/1999/02/22-rdf-syntax-ns#" typeof="List">
>>      <li property="first">Lemon sorbet</li>
>>      <li rel="rest">
>>       <span typeof="List">
>>        <span property="first">Apricot sorbet</span>
>>        <span rel="rest" resource="nil"></span>
>>       </span>
>>      </li>
>>     </ul>
>>    </div>
>>   </div>
>> 
>> aiming to get:
>> 
>>   @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
>>   [] 
>>     <#flavor> [
>>       a rdf:List ;
>>       rdf:first "Lemon sorbet" ;
>>       rdf:rest [
>>         rdf:first "Apricot sorbet" ;
>>         rdf:rest rdf:nil ;
>>       ] ;
>>     ] ;
>>     .
>> 
>> What I actually get with your RDF Distiller is:
>> 
>> @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
>> 
>> <> <#flavor> [ a rdf:List],
>>  [ <#rest> [ <#first> "Apricot sorbet";
>>      <#rest> <nil>;
>>      a <#List>]] .
> 
> As previously mentioned, there is a special rule on the root element (and <head> and <body> with HTML+RDFa) in which @about="" is assumed, so this will not generate a BNode as the top subject.
> 
> It's not parsing @resource="nil" as a vocab-expanded URI, but as a document-relative URI. This is because the rules for @resource don't include terms, and @vocab uses the same rules as term (SafeCURIEorCURIEorURI not TERMorCURIEorAbsURI). To make this work, you'll need to use a full URI or a CURIE. Using @resource="rdf:nil" should work, as the 'rdf' prefix is defined in the default profile.
> 
> One more gotcha I discovered: If you process the document as HTML+RDFa, @property="first" doesn't generate rdf:first, as the @vocab might indicate; rather, it generate xhv:first; this is because the (X)HTML+RDFa spec defines "first" as a term in the default profile, which trumps @vocab. So, really, your example can only work reliably if you use rdf: prefixes throughout. It worked in the distiller, as it wasn't identified as being an HTML document.
> 
>   <div vocab="#" typeof>
>    <p>Flavors in my favorite ice cream:</p>
>    <div rel="flavor">
>     <u typeof="rdf:List">
>      <li property="rdf:first">Lemon sorbet</li>
>      <li rel="rdf:rest">
>       <span typeof="rdf:List">
>        <span property="rdf:first">Apricot sorbet</span>
>        <span rel="rdf:rest" resource="rdf:nil"></span>
>       </span>
>      </li>
>     </ul>
>    </div>
>   </div>
> 
> Would you assume that @vocab should trump terms in the default profile (now renamed to something else, due to dropping @profile)? If so, please file a bug.
> 
> Gregg
> 
> On Aug 14, 2011, at 3:16 PM, Jeni Tennison wrote:
> 
>> ...
>> Cheers,
>> 
>> Jeni
>> -- 
>> Jeni Tennison
>> http://www.jenitennison.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
Received on Tuesday, 23 August 2011 08:39:27 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:52 UTC