--- Knowledge organization for its own sake: relationships between theory and applications abstract of video keynote at 2nd Brazilian ISKO Conference Rio de Janeiro, 28 May 2013 by Claudio Gnoli University of Pavia. Science and Technology Library, Pavia, Italy gnoli@aib.it | @scritur For the presentation please see http://mate.unipv.it/gnoli/sake.pdf The domain of knowledge organization (KO) consists of several layers: KO theory looks for the best principles to arrange the parts of human knowledge into general, consistent views; KO systems provide structures and symbols to do it; KO representation records these as formalized, nowadays digital, data; KO applications make use of all this for the particular needs of various information services. It is important that each of these layers be in sound relationships with the other ones. I have elsewhere discussed relationships between theory and systems [1] and between systems and representation [2]. In this occasion I intend to focus on relationships between theory (together with systems and representations) and applications. It is often said that KO should be driven by the needs of its users, so that user surveys can be a major source for KO theory. Supporters of domain analysis believe that each community needs a different organization of knowledge, so their approach also starts from the final layer to determine the shape of the previous ones. I would like to propose an approach that is in some way opposite: KO may identify theoretical principles, and produce certain systems, that are not in compliance with any particular need, beside the intellectual satisfaction of understanding the world and our knowledge of it. Such desire has existed among men in all times, although it is now often forgotten due to the dominance of technology and pragmatics in modern life. One could argue that KO for its own sake be a useless luxury that we cannot afford. However, experience with research teaches that many major innovations originated from research that was purely theoretical. None can know which will ever be the applications of discovering the Higgs boson, despite scientists feel that understanding it is a central aim of contemporary research. Research on fullerenes, spheres and tubes of carbon atoms lattice, originally had not any application purpose, but was later found extremely productive and their authors were awarded the Nobel Prize for it in 1996. In the same way, KO theory, and systems reflecting it, should be allowed to develop freely without necessarily having any particular application in sight. As witnessed by Ernst Mayr, KO systems that best reflect the properties of reality as it is often turn out to be also the most efficient ones for the purposes of information retrieval. -- References 1: C Gnoli, Ontological foundations in KO: the theory of integrative levels applied in citation order, Ibersid 2011, in Scire, 17, n. 1, p. 29-34 2: C Gnoli et al., Representing the structural elements of a freely faceted classification, in Slavic & Civallero eds., Classification and ontology, Ergon, Würzburg 2011, p. 193-206 This page: http://mate.unipv.it/gnoli/sake.txt : 2013 - 2013.05.29 -