Temporal Entities: patterns¶
Introduction
The document outlines modelling choices in relation to temporal entities such as activities and statements which specify agency, causality, and other properties that, within CRM, can be considered to belong to this domain.
The modelling choices are recorded as generalisable patterns. Each pattern is made explicit using an example which it is both encoded in RDF using Turtle and presented as a codebox and in a graphical form. Each codebox is hosted as Gist on GitHub and embedded on the page. This solution has been taken to give the possibility to the readers to comment, download and embed the codebox. Each graphical representation of the codebox is created using the ontologyviz tool. It is possible to click on the pictures to enlarge it.
All the prefix used within the document can be found on http://prefix.cc/.
Creation of an artefact¶
We model the physical creation of artefacts, such as paintings or photos, using the class E12 Production, a subclass of Event which described a specific activity which results in the creation of one or more new physical object. This type of creation is modelled in CRM using the E12 Production Entity, using the pattern:
E22 Man-Made Object → P108 was produced by → E12 Production
Technique used¶
A production itself can be contextualised using diverse properties that augment the information about this activity. A quite important one for us was the technique used. The CRM modelling follows:
E12 Production → P33 used specific technique → E29 Design or Procedure
The encoding of the modelling in Turtle is pretty straightforward:
Codebox 1. Technique. Click here to comment the modelling.
Agency¶
Another important aspect to declare is the agency within an activity, which can be modelled simply as:
E12 Production → P14 carried out by → E21 Person
While the encoding of the modelling in Turtle can be seen below
Codebox 1 . Agency. Click here to comment the modelling.
Agent role in the activity¶
The role of the actor within the activity can be further modelled with a role. The modelling of this pattern is a bit more complicated matter. It would be:
E12 Production → P01 is domain of → PC14 carried out by
PC14 carried out by → P012 has range → E21 Person
PC14 carried out by → P14.1 in the role of → E55Type
We document in great details this modelling pattern in the corresponding General section. An example related to the production event would be the following:
Codebox 2. Production: agent role. Click here to comment the modelling.
Exhibition¶
We model temporary exhibitions as activities, because of their existence in time. Moving exhibitions, which do keep their collection but change location/museum are also modelled as a bigger activity (the whole exhibit), which are composed by sub-activities which do they take place in diverse locations.
The modelling in CIDOC-CRM is quite simple:
E7 Activity → P9 consist of → E7 Activity
For a better understanding, in the codebox below it is possible to see how an exhibition has been modelled as composed by two different ones, each of them happening at a different time and place.
Codebox 3. Exhibition. Click here to comment the modelling.
Provenance of an object¶
We model the provenance of an object as a series of acquisition events, which transfer an object from person to person. This allows us to follow the object throughout the years. This solution is not perfect, because in an open world it does not imply that the object stays with the same owner between acquisition events (despite the fact that we do know about it). A better modelling would be using a State which, however, are not fully considered within CRM at the moment of writing this document. While the solution is not better, it is still quite functional and can be seen in the codebox below where three diverse acquisition events are modelled
Codebox 4. Provenance. Click here to comment the modelling.