Question about modeling Interpretation as a link between Activities and Records in RiC-O
Sylvain Loiseau
Dear all,
I am a linguist interested in using the Records in Contexts (RiC) ontology to model data produced during linguistic fieldwork.
Linguistic fieldwork generates a complex network of interrelated data. Several archiving bodies in our discipline (see the www.delaman.org network) support the preservation of these materials, which in many cases constitute the only surviving testimony of endangered languages. However, only a fragment of the data produced is typically archived, partly because such data are difficult to describe in the first place.
Several years ago, F. Clavaud kindly provided guidance that helped me better understand the RiC model. I am now conducting a small experiment with IT students to develop a prototype tool that would help linguists describe their data more effectively using RiC-O.
Linguistic fieldwork data consist, in particular, of relationships between Activities and Records. For instance, SpeechEvents (a kind of Activity) are documented through Records, while DataSessions (another type of Activity) are concerned with these Records and can, in turn, produce additional Records (such as field notes) that document the analytical process.
I am currently struggling to identify an appropriate relation for linking a DataSession Activity to the Record(s) it analyzes. Conceptually, the Activity represents an interpretation of the Record — the Record is the subject of the interpretative Activity.
Most existing RiC relations connecting Activity and Record express the opposite direction — that is, that a Record results from an Activity:
rico:ActivityDocumentationRelation: “Connects at least one Record Resource or Instantiation to at least one Activity, when the Record Resource or Instantiation results from the Activity.”
rico:documentedBy
rico:isOrWasAffectedBy (expresses how an Event affects a Thing)
rico:resultsOrResultedIn (again, Event → Thing)
rico:hasOrHadMainSubject, whose domain is RecordResource, not Event.
What I am looking for instead is a way to express the inverse situation: how an Activity takes as its subject (or produces an interpretation of) a Record. By contrast with the "documents" relation, it is a "analyzes" relation.
I have considered linking the two Activities (the SpeechEvent under study and the DataSession analyzing it) or the two Records (the recording of the SpeechEvent and the field notes produced during the DataSession). However, this approach would introduce redundancies: multiple Records can be produced during a single DataSession, and the event-based model aims precisely to avoid duplicating information about properties that belong to the underlying Activity. More importantly, conceptually, interpretation always concerns a material trace (a Record), not the Event that produced it, whereas the interpretative process itself is an Activity, which may secondarily generate its own Records.
I would be very grateful for any advice or suggestions regarding how to model this relationship within RiC-O.
With best regards,
Sylvain Loiseau
Florence Clavaud
I would say you could use rico:affectsOrAffected (a property of rico:Event, of which rico:Activity is a subclass) to connect the DataSession activity to the record that results from the SpeechEvent. Or a sub-property of this property, that you could define in an extension of RiC-O; for example, if your ontology has IRI something like 'https://delaman.org/resources/ontology', a 'https://delaman.org/resources/ontology#analyzes' object property, with domain DataSession (unless I am wrong you have already created such a class? otherwise with domain rico:Activity) and with range rico:Record. You could of course add to this DataSession one to many participants (using rico:hasOrHadParticipant, or, to be more precise, rico:isOrWasPerformedBy), a date, a location, a description, and one to many instances of rico:RecordResource resulting from the activity (using rico:resultsOrResultedIn).
Hope this helps,head of the Lab, Archives nationales de France
chair of EGAD