Re: Using Facebook Data Objects to illuminate Linked Data add-on re. structured data

On 6/15/11 11:35 AM, Richard Cyganiak wrote:
> Hi Kingsley,
>
> I have a few questions. See below.
>
> On 14 Jun 2011, at 22:24, Kingsley Idehen wrote:
>> On 6/13/11 9:29 AM, Kingsley Idehen wrote:
>>> Then you go on to say that it would be much better if it said:
>>>
>>>     "id": "https://www.facebook.com/kidehen#this"
>> For the record, after scratching my head a few times following responses from both Richard and Glenn re. item above, I now see that really meant to have posted:
>>
>> https://graph.facebook.com/kidehen#this
> Would you agree that Facebook are the owners of this URI?

I would say they own the URI: https://graph.facebook.com/kidehen

I use that URI as the basis for a disambiguated URI in my data space, 
for instance, as per my comments re. owl:sameAs relations in my data space.

> Would you agree that the owner of a URI gets to decide what it identifies?

Of course.

> Would you agree that a good URI is one where the URI owner has explicitly communicated (ideally through its representations) what it identifies?

Of course.
> If I wanted to state, using Facebook URIs, that I know you, would you expect me to do this:
>
>      <http://graph.facebook.com/richard.cyganiak#this>
>          foaf:knows<http://graph.facebook.com/kidehen#this>.

In a graph, in my data space, it would be meaningful and useful since I 
would have triples in place that provide additional information about 
the referents of the identifiers in the relation above. Outside my data 
space all bets are off.

> or this:
>
>      <http://graph.facebook.com/richard.cyganiak>
>          foaf:knows<http://graph.facebook.com/kidehen>.

Not in my data space (bar you having CRUD privileges there), of course 
you could make that assertion in yours :-)

My data space is a place where I control data CRUD operations. Its also 
a place where inference rules are optionally associated with SPARQL 
queries, for instance.


Kingsley
> ?
>
> Thanks,
> Richard
>
>
>
>> And yes, that's owl:sameAs:
>>
>> https://graph.facebook.com/605980750#this .
>>
>> Re. this whole matter of Names and Address ambiguity relative to usefulness, Facebook doesn't really grok the above, neither does it grok fact that:
>>
>> <https://graph.facebook.com/130118787030077#this>  owl:sameAs<https://www.facebook.com/kidehen#this>,<https://graph.facebook.com/605980750#this>  .
>>
>> Instead of burning cycles getting them to think along the lines above, best I get something made that takes advantage of this insight for the time being :-)
>>
>> -- 
>>
>> Regards,
>>
>> Kingsley Idehen	
>> President&   CEO
>> OpenLink Software
>> Web: http://www.openlinksw.com
>> Weblog: http://www.openlinksw.com/blog/~kidehen
>> Twitter/Identi.ca: kidehen
>>
>>
>>
>>
>>
>>
>
>


-- 

Regards,

Kingsley Idehen	
President&  CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen

Received on Wednesday, 15 June 2011 11:01:56 UTC