Conceptual Data Model

It is necessary to outline a conceptual data model in which to frame the spaces and domains under discussion. The following is a simplified version of the Open GIS Consortium (OGC) conceptual data model [2]:

Real World:

the entire world in objective reality

Conceptual World:

the observed subset of the real world

Geospatial World:

a categorisation and classification of that subset

Dimensional World:

the classified entities with metric representations and spatial reference systems, but not yet represented in any software system

Project World:

the entities in a logical schema defined by a particular information community

Software Worlds:

a set of representations of the entities in an overlapping set of increasingly capable software systems with defined schemas

In this paper a ``feature'' will be taken to be a software representation of a real-world object [2], e.g. a lake, road or city, which can have associated with it a number of attributes, some of which are geometric representations (``geometries''), i.e. shapes with locations. (Note that this definition differs from a commonly understood meaning where a feature is the geometric representation in a spatial reference system.) Thus a school is represented as a feature with one associated complex geometry which is the set of polygons representing the floor plans of the buildings and another that is the boundary of the site.

If we examine the conceptual data model sketched out above, we will see that unique labelling can only be done for discrete objects which are already a categorisation of a subset of the real world.