Introduction:What Landmarks Are, and Why They Are Important - Landmarks: GIScience for Intelligent Services - page 4

Level 1 Position, place —(the particular portion of space occupied by something;
“he put the lamp back in its place”)
Level 2 Po in t —(the precise location of something; a spatially limited location;
“she walked to a point where she could survey the whole street”)
Level 3 Location —(a point or extent in space)
Level 4 Object, physical object —(a tangible and visible entity; an entity that can
cast a shadow; “it was full of rackets, balls and other objects”)
Level 5 Physical entity —(an entity that has physical existence)
Level 6 Entity —(that which is perceived or known or inferred to have its own
distinct existence (living or nonliving))

This hierarchy of hypernyms of landmark tells us something about the conceptual
space of the term landmark . Each level inherits from the more abstract concepts and
adds its own specific meanings. Especially the derivation from place and location
suggests a formal property of an image schema of contact to a place (e.g., [ 11 , 25 ] ),
a meaning that is reflected in below's approach to define landmarks as references to
locations.

1.1.1 An Extensional Approach

Theoretically, one approach to capturing the meaning of a concept is by forming
a taxonomy , and enlisting all entities that belong to a particular category in this
taxonomy. So let us assume our taxonomy—whether a cognitive or a database
taxonomy does not matter—contains a category landmarks . Is it then humanly
possible to enlist all landmarks, or, with other words, can the meaning of landmarks
be captured by providing a comprehensive list of all landmarks?
We could think of landmarks in our hometown and start a list. Asking friends
would add to the list. But asking friends of friends would add further, and may
bring up objects we were even not aware of. Are these all landmarks? Perhaps we
would also like to challenge some of the elements picked by others. So in order to
answer the question whether an extensional approach is the way forward the proper
response is: The approach is feasible, but unsatisfying. Our favorite search engine
may have such a list and use it on their webmaps, but there is no explanation in such
a list why these elements were considered to belong to the category, and others
are left out. Correspondingly, there may remain disagreement about individual
elements. The deeper issue with an extensional approach is about completeness. In a
closed world assumption only listed objects are landmarks. But there is no known
procedure guaranteeing completeness as long as the selection criteria are not made
transparent. In our dynamic world a related issue is maintaining the list by adding
and removing items.