The standards we are primarily investigating are about connecting one computer system with another, at the network level and at the information content level. The (semantic) interoperability of systems conforming to a standard must be tested to ensure that no conceivable ambiguities within the standard itself leave room for multiple incompatible interpretations of the standard.

All of this takes time, more time than the current pace of technical and scientific progress allows. As a result, vendors and users are forced to rush into implementations of systems where large uncertainties still exist, and must patch in solutions as problems arise.

In order to solve these time related issues, a fundamental change in how systems come to communicate with each other is already envisioned today. This new approach is called "self-integrating systems."

It is beyond the scope of this document to describe this approach in detail. We can, however, state that the following components for a realisation of such a system need to be developed:

• A negotiation protocol that will specify how two systems will greet one another.
• A semantic equivalence metric, i.e. a quantitative measure of how alike two concepts are.
• A reasoning or inferencing capability within the communicating systems, to enable the systems to make judgements and draw conclusions about the meanings of terms.
• A common ontology as a unified model of concepts and definitions

It should be noted there are persistent problems regarding semantic equivalence metrics which are considered so serious by some researcher that they question that a sound basis for such an endeavour can be build. Most of their concerns are related to the rather opaque nature of concepts themselves and the lack of good, noncircular and unambiguous definition of them.