As databases (DB) developed in the 1960s and 1970s as the primary technology for supporting business use of information in computers, the study of “semantic” techniques under a variety of labels became important for meaningfully modeling and using such databases in organizations. Yet it was only with the advent of the World-Wide Web and the implicit ability to broadly share its vast distributed supply of information that made the need for automated understanding of the formal, agreed meaning of data in autonomous computer-stored resources inescapable. When this need was formulated and recognized by the AI and DB research communities, the Semantic Web (SW) was born.

Today the SW community has somewhat retreated from its earlier over-ambitious goals, as they became stated by AI researchers often ignoring reservations made by researchers and practitioners from the DB community. Even the term “Semantic Web” appears to be fading, and the more sophisticated uses of Linked Data threaten to go the same way, unfortunately. As befits perhaps a seminar in a (Summer) School, it could be time to draw educational lessons from this process.

As the saying goes, one who does not remember history is doomed to repeat it. Long preceding the current emergent “social dimension” in computing where groups, networks, or whole communities analyze, design and deploy systems by meaningful agreement,  in the 1970s and 80s similar modeling principles were identified and exploited for the design and implementation of databases and information systems within companies and other organizations of people. Notably the structured use of natural language (NL) during the modeling process deserves new attention and scrutiny, and in this seminar a method based on NL will be presented that became (and is still) successfully taught and applied in many practical business contexts where agreement on meaning, i.e. semantics, of information arising from and for communication is of the essence.  

The method is called NIAM (Natural Language Information Analysis Methodology) or ORM (Object-Role Modeling) depending on the literature source, and its NL principles were later re-applied to form the basis for DOGMA (Developing Ontology-Grounded Method and Application). These will be used in this seminar to establish ad illustrate those principles, and to show how the structuring of NL in general allows a formal foundation for e.g. semantic mapping and invariance, the involvement of groups of users, and the design of systems that support the emergence of agreements in such groups or communities.