W3C home > Mailing lists > Public > semantic-web@w3.org > February 2015

Semantics of events

From: Xanthor <xanthor@riseup.net>
Date: Tue, 24 Feb 2015 02:40:48 +0100
Message-ID: <54EBD6A0.3060204@riseup.net>
To: semantic-web@w3.org
Hello

I have an issue with event semantics.

Here is a simple testcase:

Alice wants to go to theatre with Bob, Charly and Debbie.

They all have needs, so Alice will describe everything in RDF.

Bob wants to add the event to his calendar, in order to block his evening.

Charly wants to browse the actors to choose the performance.

Debbie wants to share the performance with friends.


Suppose we have an ontology to describe theatre (prefix: th)

We can imagine something like this:

Classes:

th:Play
th:Performance

Properties:

th:hasPlay
	domain: th:Performance
	range: th:Play

th:hasActor
	domain: th:Performance
	range: foaf:Person

th:hasPlace
	domain: th:Performance
	range: geo:SpatialThing / gn:Feature

th:hasSpectator
	domain: th:Performance
	range: foaf:Person

th:hasTheatreDirector
	domain: th:Performance
	range: foaf:Person


But now, comes the issue of "event" (which will handle the time aspect).

Semantically : is a performance an event, or is a performance related to 
an event?

That is to say, without any existent vocabulary, do we have:

th:Performance rdfs:subClassOf event:Event.

or

th:Performance event:hasEvent :event;
:event a event:Event.

It seems to me that the first solution is the way to use RDFIcal, and 
the second is more http://motools.sourceforge.net/event/event.html

For me, but I can be wrong:
The first solution answer directly to Bob testcase, but present semantic 
issues (Is is ok for an event to have a Theatre director? Is it ok for a 
performance to have a ical:prodid?)

The second solution has better semantic, but we need tools to answer Bob 
testcase.

What to you think?
Received on Wednesday, 25 February 2015 20:46:20 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 08:45:41 UTC