Re: Southampton Pub data as linked open data

On Jul 29, 2008, at 8:44 PM, Richard Cyganiak wrote:

> On 29 Jul 2008, at 15:23, Bijan Parsia wrote:
>
>> On 29 Jul 2008, at 15:08, Tim Berners-Lee wrote:
>>
>>> Making something like pub:name a subProperty of rdfs:label is  
>>> essential
>>> for the Tabulator for example to know it can use names as labels  
>>> in the UI.  So please do it
>>
>> I would suggest that Tabulator find a better way. E.g., something  
>> Fresnelish.

For many reasons. I.e., I would recommend this even if subpropertying  
rdfs:label becomes legit OWL DL.

> Let me see.

Try *thinking* next time.

> 1. The RDF working group defines RDF and RDFS, including the useful  
> utility properties rdfs:label, rdfs:comment, rdfs:seeAlso and  
> rdfs:isDefinedBy.

Er...to say that they "defined" them is stretching things a bit. E.g.,

"""(rdfs:comment, rdfs:seeAlso, rdfs:isDefinedBy and rdfs:label are  
included here because some constraints which apply to their use can  
be stated using rdfs:domain, rdfs:range and rdfs:subPropertyOf. Other  
than this, the formal semantics does not assign them any particular  
meanings.)"""

> 2. RDF people start to use these properties in conformance with the  
> specs.

Also stretching things a bit.

> 3. While defining OWL-DL, the OWL WG introduces rather arbitrary  
> restriction on all these properties.

It's not actually arbitrary per se, but I do understand that  
technical considerations are a lost cause on you.

> 4. OWL people ask RDF people to stop subclassing these properties  
> in order to meet the restrictions imposed by OWL-DL.

Do whatever you want.

> This doesn't make any sense to me.

Many things don't, I see. I notice you elided the fact that you have  
a rather ridiculously constrained notion of the "open web". There's  
no *need* to be hostile, after all. (And yes, you started the tone in  
the gutter in the face of non-guttered tone.)

> The OWL WG blundered when they redeclared these properties as  
> annotation properties.

That part is quite sensible, actually. The design of RDF is  
problematic in several ways not least in mixing domain modeling and  
annotative stuff.

I wouldn't be surprised if these choices don't affect you. Good for  
you. I even understand that you don't care about people for whom it  
matters. Also good for you.

> [snip]
>>> Presumably OWL DL systems can be built to ignore the  
>>> rdfs:subClassOf fact when they do OWL-DL reasoning on the data.
>
> +1. You broke it, you fix it.


Thanks for playing.

Cheers,
Bijan.

Received on Tuesday, 29 July 2008 23:29:37 UTC