Virtual Maastricht McLuhan Institute (VMMI)

Knowledge Organization

Marshall McLuhan drew attention to the limitations of print media. While electronic media potentially offer ways beyond these limitations, many projects today simply continue using the limitations of the previous media just as McLuhan warned.

The early enthusiasts of computing (Engelbart, Quillian, Nelson, Wood) recognized that computers entailed a new chapter in the history of linking: indeed that links in terms of relations are central to definitions of knowledge and that the new media ultimately require a fundamental re-organization of knowledge; that hypertext and hypermedia were the tip of a new augmented knowledge.

The semantic web today uses triples (tryadic, 3), which permits sharing of information, but only information that is assumed to be true. This entails statements and propositions about entities and their relations in terms of taxonomy (is a), partonomy (has a) and, in general terms, properties. As a result the Semantic Web addresses questions of Who? and What? but does not deal with activities and dimensions. Or, to use the distinctions of Perreault, the semantic web today deals with subsumptive relations, but not with determinative and ordinal relations. Hence, it tends to define knowledge as entities (taxonomy, partonomy, properties) and to dismiss the rest as data. Information requires tryadic (3) systems to deal with Who? and What?. Knowledge requires hexadic (6) systems to deal with Who?, What?, Where?, When?, How? and Why?.

To create its system the semantic web uses logic in terms of semantic primitives and semantic relations. The good news is that this creates a solid system for machine to machine (M2M) communication. So semantic is used in a very specific sense, namely, the logical meanings of instructions that the computer needs to process actions and transactions. This approach seeks commercialisation in the area of Entreprise Application Integration (EAI), as i.e. evident in the Semantically Enabled Serice-Oriented Architecture (SEMA).The seemingly good news is that this leads to a static system, which is always true and can be sold as a web of trust. Unfortunately, this completely overlooks the insights of science and philosophy especially of the past century (e.g. Heisenberg, Polanyi, Lorenz, Cassirer, Whitehead, Popper) and indeed ignores major trends in computing (Turing, systems theory, cybernetics, complexity).

On the positive side, there are a number of new trends that address the technological limitations of the present fashions of the day: autonomic computing, evolutionary computing, natural computing and organic computing.

Our concerns are with the human side and the implications for human knowledge: not just digital libraries and e-science, which reproduce the limits of the past, but new systems and new access. Here the bad news is that the semantic web of today has nothing to do with human meanings, with human interpretations, and hence has nothing do with dictionaries, encyclopaedias, history or culture. Worse, the system of triples, on which the semantic web is based, deals with information and not knowledge. It has no way to test the truth of statements beyond their logic. Ultimately it is a threat to the very idea of knowledge and ignores different traditions of knowledge and ways of knowing.

Indeed the whole approach of the semantic web pretends that there is one system of knowledge. one logic and ultimately one kind of interface, that applies to all machines and persons. This model pretends that the way a Buddhist looks at a Zen garden, the way a Russian Orthodox believer looks at an Icon; the way a European looks at art or cartoons are identical. Stated in dramatic terms, the logic of the semantic web today is that culture is irrelevant to its system. Hence, it ignores the unique expressions of individuals that inspire local and regional diversity around the world; that are the basis of tourism in economic terms and are ultimately our highest symbols of human freedom. The semantic web is creating a system where semantic entails only the effects of instructions; where actual meanings are meaningless. Our concern is to create new systems, which reflect dynamic knowledge organization and different ways of knowing.

Cultural Computing

Our concern is to explore models, which address semantic in its original 19 th century sense as a science of meaning, when it was linked with a number of other disciplines including semiotics, semasiology, lexicology, lexicography and onomasiology. This entails much more than humanities computing as it is now defined, which deals mainly with text. [ i] How can we create systems that continue giving access to alternative views; to unique insights as opposed to official claims? How can we create systems that strive for truth, which also provide alternative attributions; acknowledge levels of truth, levels of claims, levels of certainty, levels of quality? How can we create systems that entail more than striking images of hits and give us ordered access to our long and changing history of changing meaning, interpretation and ultimately wisdom?

McLuhan explored historical effects of media on the trivium (grammar, logic and rhetoric). VMMI explores how new media affect both the trivium and quadrivium, and implications for knowledge and culture today.

[ i] Work on projects such as the Dictionary of Old English (DOE), and Records of Early English Drama (REED) in the early 1980s were among the factors that inspired scholars such as Ian Lancashire, Willard McCarty and individuals such as Yuri Rubinsky to lay the cornerstones for what has emerged as Humanities Computing. At a very practical level this explored how new methods such as Standardized General Markup Language (SGML) could be used to provide new analytical tools for the understanding of text.. This led in the short term to tools such as TACT, and in the longer term to awareness of a need for new metadata standards. This inspired the Dublin Core Metadata Initiative which took a pragmatic view towards solutions.