Re: Updated LOD Cloud Diagram - First draft and last feedback.

On 8/15/14 9:31 PM, Ruben Verborgh wrote:
> Hi Kingsley,
>
>> >I passed<http://data.mmlab.be/people/Anastasia+Dimou>  through our edition of Vapour [1]
> Thanks for checking this. Below is what happens on HTTP level.
> Looks fine to me. Do you spot an issue here?
>
> $ curl -H "Accept: text/turtle" -Lhttp://data.mmlab.be/people/Anastasia+Dimou  -i
> HTTP/1.1 303 See Other
> Server: nginx/1.1.19
> Date: Fri, 15 Aug 2014 20:25:52 GMT
> Transfer-Encoding: chunked
> Connection: keep-alive
> Cache-Control: public, max-age=3600
> Vary: Accept
> Access-Control-Allow-Origin: *
> X-Powered-By: Linked Data Fragments Server
> Location:http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou
>
> HTTP/1.1 200 OK
> Server: nginx/1.1.19
> Date: Fri, 15 Aug 2014 20:25:52 GMT
> Content-Type: text/turtle;charset=utf-8
> Transfer-Encoding: chunked
> Connection: keep-alive
> Vary: Accept-Encoding
> Cache-Control: public, max-age=3600
> Vary: Accept
> Access-Control-Allow-Origin: *
> X-Powered-By: Linked Data Fragments Server
>
>> >  and it produced an unexpected result [2]
> The result seems fine to me:
>
> Dereferencing Entity URI (requesting (X)HTML)	3/3
> Dereferencing Entity URI (requesting JSON-LD)	5/5
> Dereferencing Entity URI (requesting RDF/XML)	2/3
> Dereferencing Entity URI (requesting TURTLE)	5/5
> Dereferencing Entity URI (without content negotiation)	2/2
>
>
> All test that should pass, pass. (We don't offer XML.)
>
>> >i.e., unambiguous names resolve to description documents i.e., as exemplified by terms [3] in natural language, an identifier resolves to the description of its referent.
> That's what we do, right?
> -http://data.mmlab.be/people/Anastasia+Dimou  is the term
> -http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou  is the document about that term
>
>> >In the worst case, we'll fix any anomalies in our Vapour implementation.
> Looks fine to me. Did something change in the meantime?

The issues arise from the conclusions. Basically, the denotation (name) 
aspect of the term isn't associated with its connotation (description 
document), via a discernible RDF relation. Hence output such as:

<http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou> 
denotes a Web document bearing JSON-LD content.

You can fix this by adding a relation that associate 
<http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou> 
with <http://data.mmlab.be/people/Anastasia+Dimou> .

Suggestions:

## Assuming the document describes a single term

<http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou> 

foaf:primaryTopic <http://data.mmlab.be/people/Anastasia+Dimou> .


## Assuming the document describes a variety of termss

<http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou> 

foaf:topic <http://data.mmlab.be/people/Anastasia+Dimou> .

## More IANA friendly

<http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou> 

xhv:describes <http://data.mmlab.be/people/Anastasia+Dimou> .

## Powder exploitation

<http://data.mmlab.be/people/Anastasia+Dimou>
wdrs:described by
<http://data.mmlab.be/mmlab?subject=http%3A%2F%2Fdata.mmlab.be%2Fpeople%2FAnastasia%2BDimou>.

Via URIBurner [1][2] what you have works, due to the fact that we post 
process the description document and then inject missing relations to 
alleviate the issue highlighted in this thread.

Hope this clarifies matters.

Links:

[1] http://linkeddata.uriburner.com/c/9DQZ6O3L -- document description 
(note: Anastasia Dimou is denoted by a URI [confined to @href attribute 
of <a/>] that's used as the object of the foaf:topic relation)

[2] http://linkeddata.uriburner.com/c/9DQHQJ4L - actual description of 
Anastasia Dimou

[3] http://linkeddata.uriburner.com/c/9BQOSMMQ -- reified statement 
representing familyName relation .

-- 
Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog 1: http://kidehen.blogspot.com
Personal Weblog 2: http://www.openlinksw.com/blog/~kidehen
Twitter Profile: https://twitter.com/kidehen
Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen
Personal WebID: http://kingsley.idehen.net/dataspace/person/kidehen#this