Business information has no “natural” constraints. This becomes evident when trying to reverse engineer a taxonomy or ontology within an existing enterprise. There are no natural boundaries, definitions or theories governing the meaning of terms. They are some generally accepted terms in the accounting arenas such as defined in GAPP but even those are subject to creative interpretation or manipulation. Another factor that diminishes the value of ontologies in a business environment is the need to have “adaptive” ontologies. This indicates there is already a concern than the ontology is not stable and must be adapted. What is the volatility of the ontology? How frequently will it need to be adapted and under what circumstances and by what process? What this seems to imply is that the ontologies are subjective, volatile and unstable.

Extracting the semantics from data when replacing legacy systems or attempting to integrate systems such as customer data integration has a purpose but that purpose and value is only limited to the duration of the project. Once the project is completed those artifacts, like requirements are historical and typically never referred to again. I suggest that building ontologies, taxonomies and pursing semantics have a limited half-life.

The lack of natural constraints and stability and the limited business value and half-life of semantics suggests that it although an interesting theoretical pursuit, will contribute little measurable value to the business. Even within the world of nature there is increasing anxiety with the ontology that defines the animals species given that with improved science some of the characteristics that defined a species are no longer considered unique and as a result the structure of the ontology no longer represents an accurate model. Imagine this volatility when attempting to define abstract, unstructured, ill-conceived and ambiguous business terms. The result may be interesting but of minimal value to improving the business.

Another factor to consider is that business training does not include skills in building ontologies, taxonomies or the understanding of semantics. This expertise is certainly no evident in IT nor is a hiring criteria for CIO’s. So where will the expertise to develop these elaborate ontologies come from and who will continually adapt them within businesses? Will there be a department of ontologies, taxonomies and semantics?

Ultimately unless we can show a strong correlation between the development and maintenance of ontologies, taxonomies and the use of semantics to the business, this will continue to be a theoretical pursuit. It is a pursuit that I thoroughly enjoy but unlikely one on which to build a career.