Re: telecon Thursday, 1600 UTC

My understanding is:

Manu Sporny wrote:
> Mark, I think you're implying that the following triples are generated:
> 
> Mark Birbeck wrote:
>>   <img src="A" instanceof="B" />
> 

yes

> <A> rdf:type <B>
> 
>>   <div about="A" instanceof="B" />
> 

yes

> <A> rdf:type <B>
> 
>>   <div resource="A" instanceof="B" />
> 
> _:x rdf:type B
> 
> I thought we agreed that @instanceof doesn't apply to @resource.
>

Correct, but that is the point of my continuing confusion on the
semantics of @instanceof, and I was begging for an explanation for a
while...

If I had

<div resource="A" rel="a:b"/>

then the current rules are that "a:b" applies to the 'inherited'
subject. That can be a real URI in case of, say

<div about="#q"><div resource="A" rel="a:b"/></div>

which yields

<#q> a:b <A>

or the Bnode from a hanging rel, like

<div about="#q" rel="q:r"><div resource="A" rel="a:b"/></div>

<#q> q:r [ a:b <A> ].

The behaviour of @instanceof seems to follow the same pattern for most
of the cases, but you seem to imply that, somehow, in some cases, an
extra BNode is generated for the subject of @instanceof. Can anybody
tell me what the result of, say:

<div about="#q" rel="q:r"><div instanceof="q:s"/></div>

is? My instinct would say

<#q> q:r [ rdf:type q:s ]. Manu, you say above

>>  <div resource="A" instanceof="B" />
>
> _:x rdf:type B

But WHICH _:x are you talking about? A new one just created for the
local instanceof? Is it an inherited subject from the parent?

I have been asking for a clear answer on that one for weeks, and it
seems that you all seem to know something that I don't... Help!:-)


>>   <a href="A" instanceof="B">label</a>
> 
> _:x rdf:type B
> 
> Again, I thought we agreed that @instanceof doesn't apply to @href.
> 

More exactly: we agreed that, for almost all aspect of the processing,
@resource and @href behave in an identical manner.

Ivan


> -- manu
> 

-- 

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