[Data-modeling] Events

[Ryan Shaw]

On 2/12/08, Kirrily Robert <kirrily at metaweb.com> wrote:

>  A while ago I modeled some "event" types in my user domain and these got promoted into the "Time" domain as /time/event and related types.  Now we're starting to realise that what's there isn't really working for us, so I wanted to open up some discussion about how better to model events and time periods.

What specifically isn't working with the current schema? (Not
disagreeing, just interested in specific issues you've run into.)

>  As I see it (now), the fundamental building block is just a time period which is simply two date/times, a start and an end.  This can be used for just about anything you might want, eg. the duration of a war, the run of a Broadway play, or the time when a sports event occurs.

One problem is that sometimes (especially in ancient history and the
geological sciences) time periods are not identified using specific
temporal ranges, but using some other criteria. For example, an
ancient historical period might be defined as the period during which
a specific kind of pottery was prevalent. It could be resolvable to an
approximate time range given the estimated ages of the oldest and
youngest discovered pieces of that pottery, but that doesn't mean that
the period is defined as extending over that temporal span. So if,
say, an older piece of that pottery were found, or carbon dating
procedures were made more accurate, the temporal span could change
even though the identity criteria did not.

In other cases, a period might be defined as beginning after a certain
event, even though the date of that event is not known.

These examples suggest that some sort of marker type, without
start/end dates, might be appropriate. Topics with that marker type
could participate in various sorts of temporal relations (e.g.
happened after, happened before) and could in theory be associated
with or resolved to date ranges but wouldn't necessarily have to be.

>  Currently the /time/event includes additional properties such as "Location" and "People involved" which I don't think should necessarily be on the event type.  I think that subtypes should do the necessary things there.  For instance, for a war, it might be sensible to have location, but instead of people involved you'd probably just rely on the combatant properties that "military conflict" has.  So I'm thinking that we should probably leave the basic building block (whether we call it time period, event, or whatever) as bare as possible and rely on other types to get specific where necessary.

That is probably a good idea. The philosopher Peter Hacker has an
excellent article entitled "Events and Objects in Space and Time"[1]
in which he discusses the problems associated with trying to establish
the location of events like bankruptcies or solar eclipses. Where
something happened really depends on what kind of something happened.