There is no one formal ontology for GIScience purposes. Agarwal notes Uschold and Gruninger (1996)’s four types of ontologies: ‘highly informal’, ‘semi-formal’, ‘formal’, and ‘rigorously formal’. Agarwal continues to outline other academics’ categories of ontologies, which can be loosely fit into the aforementioned four types. Most interesting to me are the ‘highly informal’ ontologies, which can comprise general or common ontologies and linguistic ontologies. How can these ontologies be incorporated into GIScience and into a GISystem? Do they need to be translated into a more formal or meta-ontology in order to be properly analysed, reproduced, and/or applied broadly across different applications? These are questions I don’t have answers for.
Agarwal acknowledges the lack of semantics in the ontological specifications. He notes that “explicit stating and consideration fo semantics allows better merging and sharing of ontologies” (p. 508)– perhaps it is from here, in the recognition of varying semantics across cultures and people, where we can move from informal to formal ontologies. Concepts can therefore be qualified with a criteria stemming from the merging and sharing of ontologies, and consequently increase our understanding and better our analyses.
-sidewalkballet
Tags: GEOG506