Some days ago I was checking how I could use the SIOC (Semantically Interlinked Online Communities) ontology to broadcast/share Talk Digger’s content in a meaningful way. (read more: Using SIOC ontology to connect Talk Digger with other online communities)
The implementation using RDF
Now it’s time to implement it by formatting that content using RDF/XML following the ontology’s guideline.
To create a good ontology you have to process by iteration: refining the ontology with testing and peer reviews.
Implementing an existing ontology in a system (such as Talk Digger) also has that process: generating a RDF file accordingly to the ontology and then trying to figure out how to link everything together (defining URI classes, defining resources, linking to resources, etc.) to optimize the graph (optimizing the relations betweens the resources to have as much meaning (easy to query) as possible).
Visualizing the RDF graph using IsaViz
My procedure was simple:
- Generating the RDF/XML files (accordingly to SIOC and FOAF) with Talk Digger’s content database.
- Importing the RDF/XML file in IsaViz.
- Visualizing and analyzing the resulting graph.
- I checked all the relations between the resources and tried to figure out if it was possible to cut/add some of them to simplify/optimize the resulting graph.
- I checked all the anonymous nodes (bNodes) of the graph and checked if it was possible to relate them to an existing resource.
- I performed these 5 steps until I was satisfied by the resulting graph.
Playing with URIs and xml:base
What is great is that I can distribute Talk Digger’s content anywhere on the Web (with different URLs) and a crawler can download all these snipped of content (FOAF profiles, conversations content, etc.), aggregate them and merge them in a unique RDF graph. That way they can have their hands on all the relations that exist in the Talk Digger system and then querying it(the RDF graph) in useful and meaningful ways.
All that magic is possible by the fact that we can define a different URI for a given RDF/XML document using the xml:base attribute. That way I can:
- Host a RDF/XML document at the URL http://talkdigger.com.com/a.php
- Define the xml:base with the URI “http://talkdigger.com.com/db/”
- Host a RDF/XML document at the URL http://talkdigger.com.com/b.php
- Also Defining the xml:base with the URI “http://talkdigger.com.com/db/”
Then if a crawler downloads both RDF documents “a.php” and “b.php”, it can merge them to recreate the single RDF document defined at “http://talkdigger.com.com/db/”. By example, this merged RDF document would be the graph of all relations defined between Talk Digger’s content.
Talk Digger’s URI classes
I refer to a “URI class” when I talk about a “part” of a URI that is general to many URI “instances”. I refer to an “URI instance” when I talk about a URI that refer to a resource.
By example, the “URI class” of Talk Digger subscribed users is:
But an “instance” of that “URI class” would be the URI that describe a particular Talk Digger user:
In that example, this “instance” refers to a resource that is the Talk Digger subscribed user called “fgiasson”.
There are the “URI classes” defined in Talk Digger:
- URI class referring to a conversation container (work as a container for the conversation components)
- URI class referring to a conversation
- URI class referring to a usergroup (a group of users tracking that conversation)
- URI class referring to a subscribed user
SIOC and FOAF instances generated with Talk Digger content
In the next sections I will show you how Talk Digger will use the SIOC and FOAF ontologies to broadcast/share its content.
I’ll use 3 RDF documents for my presentation:
- sioc-fgiasson.rdf. This file describe the conversation evolving around my personal website [fgiasson.com] (in fact it is only a snippet for the example’s sake).
- foaf-fgiasson.rdf. This file is the FOAF profile of my Talk Digger user [fgiasson].
- sioc-blogspot.rdf. This file describe the conversation evolving around a specific blog post on a Blogspot blog called “Guilty of being…”. This blog post refers to an article I wrote on my blog.
All the RDF graphs have been exported in SVG files from IsaViz. Many mainstream applications like Visio or CorelDraw can read these files. If you don’t have access to one of these software you can refer to that list of tools to read SVG files (I am unfortunately not able to find any tool to convert a SVG file into a huge JPEG or GIF file).
Note: the path of the interesting results I found in these graphs are highlighted in red.
The SIOC document describing the conversation evolving around fgiasson.com
You can download the RDF document here: sioc-fgiasson.rdf.
The resulting RDF graph of this document is available here: fgiasson-conversation.svg
What is interesting in that graph is the interaction between
-some sioc: Post
Merging a FOAF document with this SIOC document
Now I merged the FOAF document that describe more in deep my sioc:User fgiasson.
The resulting RDF graph of these documents is available here: fgiasson-conversation-foaf.svg
The link between the FOAF to the SIOC document is made using the foaf:holdsAccount property. The result is:
— foaf:holdsAccount —>
The link between the SIOC to the FOAF document is made using the rdfs:seeAlso property. The result is:
— rdfs:seeAlso —>
If you take a closer look at the resulting RDF graph, you will see that I highlighted a “loop” in red. This is one of the really interesting results when you merge different ontologies.
What is happening here? By merging the FOAF document of the “fgiasson” user with the SIOC document of the “fgiasson.com” conversation, I uncovered some interesting relations.
In fact, the sioc:User “fgiasson” is a friend of “dgiasson” another sioc:User. However, it was impossible to know that fact before merging the two documents because that information is held in the FOAF document and not the SIOC one! At the same time, we know that “fgiasson” and “dgiasson” also belong to the same sioc:Usergroup
That way we are able to extend the power of our queries because we are able to know who are the people that belongs to a sioc:Usergroup and that are also friends. We could possibly even extend that feature by using the “relationship” ontology and then developing some type of trust networks by analyzing the relationships between people belonging to a same group.
Merging a new SIOC document to that RDF graph
Finally I merged another SIOC document that describes a conversation evolving around a blog post that link-back to the conversation about “fgiasson.com”.
The resulting RDF graph of these documents is available here: fgiasson-conversation-foaf-other-conversation.svg
In this graph I highlighted in red the fact that a sioc:User “fgiasson” is a member of two different sioc:Usergroup and that a sioc:Forum is the parent of another sioc:Forum.
In this blog post I described how I would use the SIOC and the FOAF ontologies to share the content available on Talk Digger.
There are probably some glitches, errors, or best-practices (I don’t follow) in the way I handle some classes and/or properties of these ontologies (specially about how to relate a sioc:User and a foaf: Person). Considering this it would be really appreciated to report any glitch/errors/best-practices you find to help me improving this implementation in Talk Digger.