GFO

The discussion in this section is mainly based on the publicaton [HeHe2006a]. Page 1: "GFO exhibits a three layered meto-ontological architecture consisting of an abstract top level, an abstract core level, and a basic level".

Page 5: GFO takes a "realism" approach to the philosophical aspect of ontologies: "We use the term entity for everything that exists". We feel that this use of entity is too broad and prefer to use the term Object in the same way as it is used in BFO.

The following list describes such concepts that do not usually occur in normal language usage in ontology development and for which new names have been created and those that are composed of terms.

• topoid, configuroid, cronoid, situoid
• abstract role, continuant, culminating event, endurant, factural universal, functional item, generalized process, intended-has-function, individual perdurant, material structure, natural universal, presential, persistant, persisting persistant, process aggregate, processual role, property universal, property individual, quale, quality space, social object, structural layer, substrate

Page 22:

• "Every material structure consists of an amount of substrate".
• "There are several kinds of substrates, they may be classified as solid, fluid, and gaseous substrates". The associated physical property is called Physical State. The changes in physical state take place either through temperature or pressure changes.
• "Consider, for example, an amount G of gold".
• "There may be rings, teeth, broochs, lumps etc., whose substrates contain the same G as parts."
• "There are several properties that can be attributed to G (solidity, fluidity, gaseity)".

How do signs and tokens come in? On Page 6 we find the definitions: "Concepts are categories that are expressed by linguistic signs and are present in someone's mind. Symbolic structures are signs or texts that may be instantiated by tokens."  The definition of concept in GFO differs significantly from its use in the Formal Concept Analysis and as we introduce it later for formal concepts. Unfortunately, no concrete examples are given for symbolic structures either. We suspect that a bridge to linguistics should be made, as we discuss in Semiotic Triangles.

On page 17 a time region is defined as the mereological sum of chronoids. We think that this definition is redundant because it can be derived as follows, as modeled in section Space and Time: A chronoid is a region and is composed of any number of adjacent time intervals (^TimeInterval, ◊AdjacentOf, ^Chronoid). The same applies to the definition of space regions as merological sums of topoids. A topoid is a region and is composed of any number of adjacent geo Objects (^GeoObjects, ◊AdjacentOf, ^Topoid). Because the object property ◊AdjacentOf was used for the aggregation, it automatically follows that the time intervals involved or lines, surfaces and bodies must be adjacent to one another and are delimited at the points of contact or have a common boundary.

The definition on page 18 is important to be able to compare spatial objects: "... two topoids are congruent if they have the same shape and size".

Wittgenstein says in LW 6.3611 that processes cannot be compared with the passage of time and in LW 6.36111 he argues that congruent figures can only be superimposed if they are moved out of space.

The gist of comparing GFO to other Foundational Ontologies on page 56 is that very few categories correspond exactly to each other, and that some categories have at least an eventual correspondence in the other ontology. This statement also corresponds to our experience and is of course surprising because one would have to assume, especially with top ontologies, that there would have to be greater correspondences there.

The arguments are given below in excerpts.