Dear all, This email relates to part of a homework related to the use and learn section of the CRM SIG site. Warning, relatively long email.
Upshot: here is presented a proposal for how to write useful descriptions around our data modelling patterns that might aid a learner to pick up how to use the standard. Interested members may read on to see a proposal of an initial proposed 'format' for writing these texts and two examples of what that might look like. What we're looking for: if you're interested in this topic please read on to see if you think the proposed format makes sense (will actually work, doesn't miss anything) and have a look at the two examples and see how it looks in practice. At the next SIG, we'd like to move this forward by adopting a 'format' like this and then getting volunteers to write descriptions according to the format for one or two of the diagrams. If senior members pitch in, this task can be completed without too much trouble (it is a lot of work for one person though). ---- For the longer background on this issue, please consult the issue history on the site here: https://cidoc-crm.org/Issue/ID-687-review-the-textual-descriptions-to-diagrams-in-the-functional-overview In this issue we are looking to continue the work of updating the section of the Use and Learn part of the site that introduces different modelling patterns and constructs enabled by the CIDOC CRM. This section is called "Functional Overview" and can be found here: https://cidoc-crm.org/functional-units In this section, there are diagrams which are meant to help learners appropriate (as well as remind ourselves) how to apply the CIDOC CRM in different situations. The diagrams have been updated to draw.io from a past jpeg version. The diagrams, however, lack good explanatory text around them to really help the learner appreciate what they are seeing and what it is meant to help them understand. For this reason, we really need to come up with some text that describes them. As part of this homework, I am proffering both a pattern for how I think these texts should be written and an example for two cases. This is the general pattern I think the text could follow: - *General Idea*: what area of documentation in the real world does this diagram attempt to address - *Ontological Enjeu*: how does this real world problem get addressed into a modelling pattern using CIDOC CRM - *Direct Classes and Properties*: what are the most salient classes and properties of this diagram that you should learn about. Here we provide a less formal description of the classes and properties that may be more easily digested in considering the overall pattern. Learners are always provided links to the full formal class and property definitions. We explain not only the classes and properties but important distinctions and decisions within the ontology that are raised by the diagram and the particular application of these classes and properties in this scenario. - *Examples*: we put in real world, simple examples that are easy to understand and of fairly broad general intellectual appeal (as far as this possible, depending on the classes and properties!) - *Cross Pattern Mentions*: one semantic modelling pattern may have important references to another. E.g.: acquisition might reference transfer of custody, title may inherit from general appellation and so on. - *Anti-Examples*: what doesn't apply here but you might think would, see instead x pattern Here are two diagrams and two texts in which I attempted this: *Subject Information* https://cidoc-crm.org/FunctionalUnits/subject-information Recording the subject information of a cultural heritage object or “what it is about”, is a key means of documenting and retrieving an object based on its intended meaning and content. Subject information gives a key facet by which to find cultural heritage objects with certain topics, be they very specific, e.g. about a particular person, place or event, or categorical, e.g. kinds of people, place or event. For example, a manuscript can be about Jean Paul Sartre, a photograph may have the subject New York, an article may describe World War II, a journal may be about philosophy. When modelling subject information using the CIDOC CRM we must recall that the ontology clearly divides between physical and conceptual objects. It is the class of conceptual objects, through the more specific class E89 Propositional Object that begins to enable the expression of reference to or being about the world, i.e.: having a subject. The key patterns then for documenting something having a subject then are: E89 Propositional Object P67 refers to E1 CRM Entity And E89 Propositional Object P129 is about E1 CRM Entity The distinction between these two properties is that in the former case there is a mention in the subject propositional object of the referred to target object, while in the second case, the object referenced is a central theme or concern of the subject. Consider the following examples: Jean-Paul Sartre: A Life (2005) by Hazel Rowley (E89 Propositional Object) P67 refers to Simone de Beauvoir (E21 Person) Jean-Paul Sartre: A Life (2005) by Hazel Rowley (E89 Propositional Object) P129 is about Jean Paul Sartre (E21 Person) It is important to note of both these properties (P67 and P129) that their ranges are defined as E1 CRM Entity. This is to say that a proposition can be about anything in the world. By implication, a proposition can be about all the major classes of entities of interest to cultural heritage objects including physical things, conceptual objects, events, places and actors. When documenting a physical cultural heritage object, the CIDOC CRM does allow a direct shortcut property between a physical object and something that it depicts in the world through the pattern: E24 Physical Human-Made Thing P62 depicts E1 CRM Entity Consider the example The Mona Lisa (E22 Human-Made Object) P62 depicts Lisa del Giocondo (E21 Person) It is, however, best practice in most cases to describe the physical and intellectual content of a cultural heritage object distinctly, for example The Mona Lisa (E22 Human-Made Object) P65 shows visual item the Image of the Mona Lisa (E36 Image) P138 represents Lisa del Giocondo (E21 Person) This latter form is preferable insofar as the actual entity doing the referring is the propositional object and not the physical object. Capturing its referential information at this point allows the accumulation of documentation to the most relevant instance in the knowledge graph. Related Patterns: N/A (tbd) and *Object Title Information* https://drive.google.com/file/d/1aCIT2nd3ksuYcig7O2QCgT0f4Xayc7dj/view?usp=drive_link The recording of title information for cultural heritage objects is a key aspect of documentation, allowing the representation of the names that have been ascribed over time to an object to be recorded. Titles are a primary way of identifying objects and retrieving information about them. To model this information in a uniform way, the CIDOC CRM provides the specialized class E35 Title in order to record the existence of the title itself. We note that the class E35 Title is a subclass of the E41 Appellation and E33 Linguistic Object classes. This has the implication that it is both a name and a linguistically encoded sign, meaning that in addition to the properties of E41 Appellation we can also use the properties pertaining to language to describe the title. Thus we can connect an instance of E35 Title to an instance of language (E56 Language) through the property P72 has language. Likewise we can connect an instance of E35 titler to another instance of title of which it is the translation through the property P73 has translation. Instances of E71 Human Made-Thing are linked to the their titles through the P102 has title property. Ontologically, we may note that the pattern E71 Human Made-Thing P102 has title E35 Title is a specialization of the general pattern E1 CRM Entity P1 is identified by E41 Appellation.The specialization occurs as titles are especially given to things that are human made and provide a subject of interest in itself. A classic example of this pattern is the painting the Mona Lisa. This painting has many names, two of which that are most well known are "La Gioconda" in Italian and the "Mona Lisa" in English. The relation between cultural heritage object and title can be expressed as The Mona Lisa (E22 Human Made Object) P102 has title "The Mona Lisa" (E35 Title) The Mona Lisa (E22 Human Made Object) P102 has title "la Giaconda" (E35 Title) The linguistic information regarding the titles can be expressed as: "The Mona Lisa" (E35 Title) P72 has language English (E56 Language) "la Giaconda" (E35 Title) P72 has language Italian (E56 Language) "The Mona Lisa" (E35 Title) P73 has translation "la Giaconda" (E35 Title) The information which can be modelled in this pattern thus helps answer questions such as what is the title of this/these object(s), what is/are the languages of the title(s), what translation(s) exist/s, as well, of course, as aiding in finding a particular cultural heritage object by a known title. Related Patterns: Naming ----- Look forward to your critical feedback! Best, George
_______________________________________________ Crm-sig mailing list [email protected] http://cidoc-crm.org/crm-sig-mailing-list
