===== Context Treatments ===== There are two kinds of what we might call 'Context' in play: - Word usage context - Concept original context ==== Word Usage Context ==== {{:vcoi-premise.png?600|Context Calculus}} This is the context of the owning Task Force or SIG and is part of the original remit of this work: you don't simple say 'This word means this' you say 'In the context of this TF, this SIG, this document, this effort etc., this word means this'. ==== Origin Context ==== Where possible, words used in a given TF or SIG will have been sourced from an appropriate community of practice outside of the TF itself. This might be the OMG as a whole (for terms like 'metadata' or 'model') o it might be some source selected by the TF or SIG members, so for example GovDTF might choose a particular group for the origin of all terms relating to statistics. Terms sourced from these external sources need to have a treatment of context relating to that source. Depending on the nature of the resource maintained by that external community of practice, we might be looking at: * A Vocabulary * An ontology (this distinction has not yet been discussed) === Vocabulary Context === For an external vocabulary these are words, so the same calculus applies as would be the case for this TF or SIG. Indeed he external source might be another OMG SIG or Task Force - for example most TFs would look to the blockchain PSIG for definition of Blockchain and DLT related words / terms == Homonymy and Heteronymy === See examples in meeting notes for 26 Oct. The intended usage of a given word, e.g. 'Tank' will be determined by the kinds of user (user context) as would the selection of one word over another for a given concept. The same thing happens in a range of kinds of context, including human languages, regional, departmental and organizational communities, along with variants such as in finance, different 'markets' with their differences in market practice. Then: * Different words referring to the same concept = synonym * The same word referring to different concept = heteronym? - Actually homograph since in the context (!) of this work we are interested only in whether terms (strings of characters) are spelled the same rather than whether they do or do not sound the same e.g. Bass v Bass. Here are the Nyms as a whole: {{::thenyms.png?600|The Nyms Venn Diagram}} === Ontology Context === This would be the contextual scope of the ontology in which the concept is defined. ==== References ==== Representing Context in OWL as a concept that varies in time * “Context Slices: Representing Contexts in OWL” * http://ceur-ws.org/Vol-671/pat01.pdf * Underpinning paper on representing concepts that change in time: “A Reusable Ontology for Fluents in OWL” * https://www.researchgate.net/publication/221234960_A_Reusable_Ontology_for_Fluents_in_OWL Paper on applying a 5 Ws (not 6 Ws!) approach to context. * A citable reference of prior art: * https://www.researchgate.net/publication/319231978_CAMeOnto_Context_Awareness_Meta_Ontology_Modeling