OWL in RDFa (was Re: Understanding 'chaining')

A side note: the OWL example is a difficult one. Mark referred a blog of
a guy who began to do it. However, a major hurdle comes by the fact that
OWL makes a relatively heavy use of lists (collections). Eg, for the
enumeration of class content, union, etc. And we do not have shorthands
for that; moreover, it is relatively heavy to do that in RDFa (I would
have to look up my old notes to see whether a scheme I had in mind would
work with the new/old model, probably yes, but should be checked...)

I am _not_ saying that we should reopen this issue! Just noting a
problem we should be aware of...

Ivan

Mark Birbeck wrote:
> Hi Ben,
> 
>> ...as I was writing some RDFa for Creative Commons, and found myself
>> with exactly the issue Mark describes as:
>>
>>> the following does not do anything useful:
>>>
>>>   <div about="#me" rel="foaf:knows">
>>>     <span about="#mark" property="foaf:lastname>Birbeck</span>
>>>   </div>
>> which really sucks. Clearly the above needs to mean what Mark's proposal
>> makes it mean: that the @rel is hooked onto the @about, and that all is
>> well as a result.
> 
> I think it's reassuring that real use-cases are helping us resolve
> these issues. In your case it's CC, and I saw one the other day that
> uses OWL:
> 
>  <http://ontologyonline.blogspot.com/2007/11/embedding-owl-rdfs-syntax-in-xhtml-with.html>
> 
> The author uses RDFa to write OWL, and whilst I wouldn't recommend the
> post to new readers trying to learn RDFa ;) it's a very impressive
> article, and shows that RDFa 'works' at all sorts of levels of
> complexity.
> 
> One 'pattern' that gets used a lot by the author--due to the nature of
> OWL itself--is this:
> 
>  <div rel='rdfs:SubclassOf'>
>    <div rel='owl:Restriction'>
>      <span property='owl:onProperty'>develops_from</span>
>      <a rel='owl:someValuesFrom' href='c/CellTypeOntology/immature_B_cell'>
>        immature B cell
>      </a>
>    </div>
>  </div>
> 
> As we know, this already works in the rules we have, but as you can
> see--following on from your comment--if an @about were to be placed in
> the hierarchy, with the intention of 'naming' one of the previously
> unnamed OWL classes, the current rules mean that the whole structure
> would collapse:
> 
>  <div rel='rdfs:SubclassOf'>
>    <div about="somename" rel='owl:Restriction'>
>      <span property='owl:onProperty'>develops_from</span>
>      <a rel='owl:someValuesFrom' href='c/CellTypeOntology/immature_B_cell'>
>        immature B cell
>      </a>
>    </div>
>  </div>
> 
> Regards,
> 
> Mark
> 

-- 

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

Received on Monday, 3 December 2007 05:49:57 UTC