Re: Facebook Linked Data

On 9/26/11 11:37 AM, Alvaro Graves wrote:
>
> We don't have to bring httpRange-14 and its timeless imbroglio
> into every conversation re. Linked Data :-) 
>
>
> True :) the only important thing is to make clear that facebook URIs 
> for documents are different from URIs for people.
I prefer to speak about (with computer science continnum benefits in 
mind re. terminology):
1. Data Object Identifier (Name)
2. Data Object Representation via directed graphs
3. Data Object Access Address.
Indirection (via de-reference and address-of operators) re. data access 
by reference pattern is old, thus there's a large receptive audience at 
one's disposal when using said terminology.
It costs about 12+ years to fix the broken terminology pot :-)
Kingsley
> ----
> Alvaro Graves
>
>
> On Mon, Sep 26, 2011 at 8:34 AM, Kingsley Idehen 
> <kidehen@openlinksw.com <mailto:kidehen@openlinksw.com>> wrote:
>
> On 9/26/11 11:16 AM, Sebastian Schaffert wrote:
>
> But then I would say the server should at least reply with a
> 30x redirect ;-)
>
>
> Not necessarily, they choosen to implement indirection internally,
> rather than via HTTP response headers. Naturally, doing via HTTP
> is more flexible and thereby desirable, but we have to accept that
> "half bread is better than none" re. this matter i.e., any kind of
> indirection is better than no indirection re. disambiguation of
> Data Object ID and Address for accessing its Representation.
>
> Kingsley
>
> Greetings,
>
> Sebastian
>
> Am 26.09.2011 um 17:05 schrieb Alvaro Graves:
>
> Hi Sebastian,
>
> AFAIK it's not a bug, but a feature :). This is done to
> comply with the httpRange-14 issue (i.e., you can't
> retrieve a person through HTTP but you can retrieve a
> document _about_ a person through HTTP). Since a person
> and a document about a person are different entities, they
> should have different URIs.
>
> http://www.w3.org/2001/tag/issues.html#httpRange-14
>
> ----
> Alvaro Graves
>
> On Mon, Sep 26, 2011 at 7:46 AM, Sebastian
> Schaffert<sebastian.schaffert@salzburgresearch.at
> <mailto:sebastian.schaffert@salzburgresearch.at>> wrote:
> Dear Jesse,
>
> Thanks for the effort! I am just experimenting with this.
> If I request my own Vanity URL
>
> http://graph.facebook.com/sebastian.schaffert
>
> The data I get back is:
>
> @prefix rdf:<http://www.w3.org/1999/02/22-df-syntax-ns#> .
> @prefix rdfs:<http://www.w3.org/2000/01/rdf-schema#> .
> @prefix owl:<http://www.w3.org/2002/07/owl#> .
> @prefix xsd:<http://www.w3.org/2001/XMLSchema#> .
> @prefix api:<tag:graph.facebook.com
> <http://graph.facebook.com>,2011:/> .
> @prefix og:<http://ogp.me/ns#> .
> @prefix fb:<http://ogp.me/ns/fb#> .
> @prefix :<http://graph.facebook.com/schema/~/
> <http://graph.facebook.com/schema/%7E/>> .
> @prefix user:<http://graph.facebook.com/schema/user#> .
> </561666514#>
> user:id "561666514" ;
> user:name "Sebastian Schaffert" ;
> user:first_name "Sebastian" ;
> user:last_name "Schaffert" ;
> user:link<http://www.facebook.com/sebastian.schaffert>
>
>
>
> Now the problem I see here is that the URI I requested is
> not the same URI as used in the subject of the RDF
> triples. Same holds btw if I request the data using the ID
> including "#". Which is bad in our case because we filter
> out triples that do not fulfill this condition to avoid
> importing "invalid" data.
>
> Also, the data should IMHO contain a @base statement
> defining the base for the</561666514#>, because when
> importing the data the original URI is sometimes no longer
> available.
>
> Lastly, the returned data does not contain the trailing
> "." required by turtle (see
> http://www.w3.org/TeamSubmission/turtle/#sec-grammar-grammar).
>
> Are there plans to fix this? For me, the more readable
> data would look like this:
>
> @prefix rdf:<http://www.w3.org/1999/02/22-df-syntax-ns#> .
> @prefix rdfs:<http://www.w3.org/2000/01/rdf-schema#> .
> @prefix owl:<http://www.w3.org/2002/07/owl#> .
> @prefix xsd:<http://www.w3.org/2001/XMLSchema#> .
> @prefix api:<tag:graph.facebook.com
> <http://graph.facebook.com>,2011:/> .
> @prefix og:<http://ogp.me/ns#> .
> @prefix fb:<http://ogp.me/ns/fb#> .
> @prefix user:<http://graph.facebook.com/schema/user#> .
> <http://graph.facebook.com/sebastian.schaffert>
> user:id "561666514" ;
> user:name "Sebastian Schaffert" ;
> user:first_name "Sebastian" ;
> user:last_name "Schaffert" ;
> user:link<http://www.facebook.com/sebastian.schaffert>
>
>
>
> Am 23.09.2011 um 14:09 schrieb Jesse Weaver:
>
> APOLOGIES FOR CROSS-POSTING
>
> I would like to bring to subscribers' attention that
> Facebook now
> supports RDF with Linked Data URIs from its Graph API.
> The RDF is in
> Turtle syntax, and all of the HTTP(S) URIs in the RDF
> are dereferenceable
> in accordance with httpRange-14. Please take some
> time to check it out.
>
> If you have a vanity URL (mine is jesserweaver), you
> can get RDF about you:
>
> curl -H 'Accept: text/turtle'
> http://graph.facebook.com/<vanity-url>
> curl -H 'Accept: text/turtle'
> http://graph.facebook.com/jesserweaver
> If you don't have a vanity URL but know your Facebook
> ID, you can use
> that instead (which is actually the fundamental method).
>
> curl -H 'Accept: text/turtle'
> http://graph.facebook.com/<facebook-id>
> curl -H 'Accept: text/turtle'
> http://graph.facebook.com/1340421292
>
> From there, try dereferencing URIs in the Turtle.
> Have fun!
>
> Jesse Weaver
> Ph.D. Student, Patroon Fellow
> Tetherless World Constellation
> Rensselaer Polytechnic Institute
> http://www.cs.rpi.edu/~weavej3/
> <http://www.cs.rpi.edu/%7Eweavej3/>
>
>
>
>
>
> Sebastian
> --
> | Dr. Sebastian Schaffert
> sebastian.schaffert@salzburgresearch.at
> <mailto:sebastian.schaffert@salzburgresearch.at>
> | Salzburg Research Forschungsgesellschaft
> http://www.salzburgresearch.at
> | Head of Knowledge and Media Technologies Group +43 662
> 2288 423 <tel:%2B43%20662%202288%20423>
> | Jakob-Haringer Strasse 5/II
> | A-5020 Salzburg
>
>
>
> Sebastian
>
>
>
> -- 
>
> Regards,
>
> Kingsley Idehen
> President& CEO
> OpenLink Software
> Web: http://www.openlinksw.com
> Weblog: http://www.openlinksw.com/blog/~kidehen
> <http://www.openlinksw.com/blog/%7Ekidehen>
> 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