Re: [port-peer-review] review
Dear Janos, (01)
On Mon, 24 Jun 2002, janos sarbo wrote: (02)
> Review
> --------
> Paper: Making Dough's Dream Come True
> Author: Aldo de Moor
>
>
> Dear Aldo,
>
> There is no doubt about that collaboration activities
> should be supported. I think that simple tools like email,
> possibly extended with filtering and routing facilities
> should be preferred. I also agree with you that any
> collaboration may need some form of a common ontology
> like a lexicon or some implementation of a web of definitions
> and agreements. However, I do believe that, due to the dynamic
> nature of collaboration and communication, the problem of
> providing automatic tools for such definitions is more deeply
> related to the fundamental problem of knowledge representation
> than it is generally recognized. It seems me unlikely that existing
> methods like conceptual graphs and concept lattices could be
> the solution for this problem. (03)
I do not claim that the (higher-level) definitions can be automatically
generated. Of course not. To agree on those (continuously changing)
definitions that best suit the particular community to which they apply,
requires hard human negotiation. However, once agreed upon, these formal
definitions can be useful, for instance, in selecting the right people to
discuss a particular design problem. (04)
Simon Shum and Albert Selvin ("Structuring Discourse for Collective
Interpretation", Proc. of Distr.Collective Practices, Paris, 19-20
Sept.,2000) put it nicely: "[W]e are focusing on ontologies not
(primarily) for modelling the micro-worlds that are the normal focus of
attention, but for reifying the normally invisible (or at least,
unattended to) structure of *discourse-- ways of talking about* these
micro-worlds. The purpose of discourse structuring schemes is to support
multi-perspectival discussion, not to close it down with the imposition of
a master view." (05)
> But your paper is not focusing exclusively on automatic tools.
> What you call a `socio-technical-system' is a tool involving
> human communication processes, their organization and support. (06)
Be careful, a system is not a tool. I have a very restricted use of the
term tool here, in the sense of it purely being a software application.
The information system is the whole of tools, people, and procedures etc. (07)
> The realization of a pragmatic web will certainly need such
> tools and your paper seems to suggest the primary importance
> of human processes, and the secondary one of the automatic tools
> like ontologies. Such an approach to knowledge representation
> is certainly practical. Nevertheless, it may only provide a
> temporary solution. A reason for this, I think, is that the use
> of any kind of automatic tool requires _some_ formalization of
> knowledge. Therefore, on the long term, we will need to address
> the basic questions of knowledge representation like
> `how knowledge emerges', `how knowledge can be ordered' and
> `how formal and human (cognitive) concepts of knowledge are
> related'. I also believe that a stepwise approach for finding a
> solution for the problem of knowledge representation may not
> work. There is a need for a paradigmatical change. (08)
We have no disagreement here. This is where, I think, your work may
contribute. (09)
Interestingly, all four papers seem to be examples of the blind men
describing different parts of the same elephant. I hope that in Borovets
some enlightenment will happen, and we can all see the whole animal. (010)
Aldo (011)
==========================================================================
---/// e-mail: ademoor@kub.nl
IN|F/OLAB phone +31-13-4662914/3020, fax +31-13-4663069
|/// home page: http://infolab.kub.nl/people/ademoor (012)
Dr. Aldo de Moor
Infolab, Dept. of Information Management - Tilburg University
PO Box 90153, 5000 LE Tilburg, The Netherlands
========================================================================== (013)