Knowledge Organisation

04.12.2004

Kim H. Veltman

“Challenges for the Semantic Web and Information Systems from Culture”:

SIGSEMIS Newsletter: Special Theme: Semantic Web Challenges for Information Systems, March 2004, pp. 1-5

Abstract

This paper notes that 20th century advances in computer science and information systems narrowed the earlier scope of semantics and outlines six challenges for information systems to:

1) return to the earlier scope of semantics at the turn of the 20th century
2) address all six basic questions
3) create cognitive models that link with the physical world
4) create entities to allow integration of changing paradigm shifts and world views
5) include multi-lingual and multi-cultural dimensions
6) develop models that reflect differences and limitations in media types.

At the turn of the 20th century, Semantics was a “science that studies the content (meaning) side of linguistic signs.” It was linked with five other fields: 1) Semiotics a “science of general properties of sign systems”; 2) Lexicology, a “science that studies vocabulary of language also called Lexical Semantics”; 3) Lexicography, a “science of dictionaries and their creation”; 4) Semasiology, a “branch of semantics that seeks meaning departing from expression side of language that studies the development and changing meaning of words” and 5) Onomasiology, a “branch of semantics which departs from a meaning side of language and asks what expressions exist in other languages.”

During the 20th century there were many breakthroughs in computers and information systems. Paradoxically this quest to achieve automated forms of communication also led to a great simplification of semantic problems and a serious dumbing-down with respect to solutions. As a first challenge, this paper argues that we need to integrate what we left behind if we are to go ahead. In so doing, information systems can meet a second challenge of going beyond Who? and What? questions to deal also Where?, When? How? and Why? questions.


Read full article