This document describes how to model locations using IBML.
The goal of this document is to describe how to use IBML to build a computer system that will understand and discuss location relations among things.
IBUKI uses a very general notion of location that includes both places and things. It follows the idea that the notion of location and its vocabulary allows us to refine this understanding to many areas, eg, geography, chronologies and more generally ideas. In our view, for example, the color of an object 'locar=tes' it in color space. This suggests that the use of [term]s as adjectives can reasonably and in general be thought of as a 'locator;. This idea is expanded on below.
The notion of [location] includes at least two different (essentially topological) ideas: points and areas. Things thought of as points will have attribures but not detailed internal structure. We can think of |New York City| as a dot on the map of |New York State| or we might think of it as made up of bouroughs (Brooklyn, Queens) or less well constituted as neighborhoods (Greenwich Village, the diamond district) or ... .
We call [location]s having areas [region]s. [region]s are, at their core, made up of points and are partially ordered by subset on the set of their points. That is, one [region] is a subregion of another just in case the set of points of the latter is a subset of the set of points of the former. Note that these definitions allow a [region] (and therefore a [location]) to consist only of a finitely enumerated set. A set of subregions can be disjoint from one another (like the |US States|) or overlap (like the |middle east| which covers parts of several countries), they can cover the entire [region] (like the |US states| which cover the |continental US|) or just part of it (like the |US states| do not cover all of the US (consider |Puerto Rico|)).
Our theory of [location] includes the possibility of 'telescoping into' a [point location] and expand it into a [region].
[region]s whose points are linearly ordered are called [interval]s. In the case of [date]s, for example, we have the [interval] called [year]. These can be relative [jan01-dec31] or absolute [1923] (of course some calander type is implicit here).
In geography relative [place]s we write 'London' by which we mean 'London, Englsnd' ...
The notion of [location] has it own vocabulary. In conversation we use this language pervasively.
For Geography
For Things
who - people
what - activities
how - tactics
where - geography
when - chronology
why - reason
In the case of |the blue block|