I
guess the OBO Foundry (?) may well serve as a lesson how to build
an Open Standard Ontology library, as a virtual depository or registry
of upper ontologies (for common use across different knowledge fields and
business domains and knowledge applications).
Among things to be done in the first place there
are:
1. Establishing a common unifying structure for
ontologies, for their domain entities and relations, for their special axioms,
definitions and meanings;
Otherwise, there will be a large number of field models
without any clear integrative structure, like the specific biomedical
ontologies needing explicit foundational structure.
2. Establishing the basal ontological
relations.
The formal relations of the
logic of classes as <class, class>; <instance, class>; <instance,
instance> are NOT "genuinely ontological relations...between entities in
reality", http://genomebiology.com/2005/6/5/R46.
The "genuine" ontological relations have always being the
things like causality, spatial relations, temporal relations, etc. if
only I missed some "ontology revolution" in the nature and order of
relationships.
The subject of biology is nothing but special
ontological entities: living beings and animate things or life, their properties
(structures and compositions), states, processes and the relationships
among organisms, and their constituent parts.
So, dealing with living things, the physico-chemical
properties of life, biology is tending to integrate all its knowledge and
research, including its interfaces as biophysics, biochemistry, sociobiology, or
paleobiology. Such a converging tendency of biological sciences is explained by
the fundamental fact that all these many divisions and subdivisions are
interrelated by few basic onto-biological principles of living things: Sameness
or unity (of structure and function and origin); diversity, change or evolution
(natural selection mechanisms); interactions or interrelationships (of animate
beings), and continuity (reproductive, successive
generations).
That means that bio-ontology allows a single
classification of living beings, which are now
done by:
the levels of biological organization,
bio-molecules, cells,..., individuals,
populations...;
the kinds of living things, plants, animals;
the structure and function or origin of
organisms;
the processes of growth and development of living
systems
It hardly makes a real
scientific progress just duplicating the extant structure of special fields
of biology, which subject names now followed with "ontology"
designation: like as plant ontology and animal ontology or
zoo-ontology, genomics ontology or gene ontology, cell ontology or
cyto-ontology, tissue ontology and organ ontology, individual ontology,
population ontology, eco-ontology or environment ontology; or physiology
ontology and morpho-ontology and anatomy ontology, starting from animal anatomy
ontology down to fly, spider and mosquito anatomy ontologies ,and so on,
http://en.wikipedia.org/wiki/OBO_Foundry.
Before soon it will be practically impossible to count
the number of ontologies and so leave any hope for any semantic
interoperability, if only ?neon? will give us some light...
Hardly the conceptual progenitors of ontology could
foresee such a wild scenario.
Bottom line:
Having an Open Standard Ontology (library, depositary,
registry, facility) built is not a luxury but a necessity, even an absolute must
and urgency.
Azamat Abdoullaev