On 16 December 2014 at 18:42, Simon Spero <[log in to unmask]> wrote:
> I am would like to suggest some background reading that should help people
> understand some of the different concepts involved, and how they have been
> handled in other ontologies.
> 1. Events as conceptually related things that happened around a certain
> time / place.
> This is the sense of the term that refers to historical events like the
> assassination of JFK, which cover more than just a man in Dallas pulling
> the trigger on a rifle.
> See Ryan Shaw's work on LODE:
> Shaw, R., Troncy, R., & Hardman, L. (2009). Lode: Linking open descriptions
> of events. In The Semantic Web (pp. 153-167). Springer Berlin Heidelberg.
> As well as the ontology itself:
> For a broader review see Ryan's dissertation :
> "Events and Periods as Concepts for Organizing Historical Knowledge"
> 2. Event objects as representations of actions.
> (This is the default sense of the term to most ontologists and linguists).
> This usage is frequently associated with Donald Davidson, and his seminal
> paper "The Logical Form of Action Sentences" (collected in "Actions and
> The problem Davidson was addressing was how to model sentences involving
> actions, especially when there were adverbs involved . His conclusion was
> that events and actions were best treated as entities (things),rather than
> as predicates (properties). One reason was that the alternative approach
> required a new predicate for every new detail about the action or event).
> For example, "John painted 'the made up title' on Dec 15 2014, in his
> garage, with a brush", whilst be watched by his cat.
> could be translated as
> There is a painting event. Its doer was John. Its output was 'the made up
> title'. It was done on Dec 15th 2014. Its location was John's garage. The
> instrument used to do it was a brush. An observer was John's cat.
> See the Stanford Encyclopedia of Philosophy article :
> There are copies of Davidson's paper online in scanned form (Google the
> title), as well as from OUP -
> The CRM uses E7_Activity to describe this sense of Event (they define
> E5_Event as a superclass that includes sense 1)
> 3. Events as gatherings of people, organized by someone for some person.
> This is the sense of the term used in "Event calendar", and is so used in
> schema.org . It is the sense in which a conference is an event.
FWIW schema.org's Action hierarchy was very nearly a sub-type below
Event. We didn't do that, but not for reasons of deep ontological
purity. More that there were proposals bubbling along for describing
'repeating' (social) Events, while the Actions work was also being put
together. They are clearly closely related. There is also "event" in
the sense of a historical event, e.g. Lilliput's invasion of Blefuscu;
or Glastonbury Festival 2005. In both cases there's an intuitive
notion of sub-event that is hard to characterize very formally. These
can have aspects of (2.), e.g. the verb invade in the Lilliput
example. Identity criteria for sense-2 events are notoriously
slippery, whereas sense-3 things tend to be grounded in practical
details like tickets, invitations, Web pages and so on.
> 4. Events as things that create works.
> This usage is something of a pun or metonym for something to which
> authorship of a work produced as part of the activities in an event of type
> This use seems to be idiosyncratic.