From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of John Bottoms
Sent: Sunday, February 03, 2013
9:15 AM
To: ontolog-forum@xxxxxxxxxxxxxxxx
Subject: Re: [ontolog-forum]
Preferred Ontology Engineering Methodology?
All, Hear! Hear!
"Ontologists Classify Thyselves"
I agree and have said before that it is meaningless to air differences without
an understanding of how an ontology will be used. The segregation of ontology
types will provide focus to its use or misuse and is an exercise in
classification and entails predicates.
The purpose of the classification itself leads to segregation of typed
instances.
My suggestion:
Two different ontology types; weak AI and strong AI for lack of better labels.
They could be called by a number of different labels such as database, KR, AGI
or whatever, but it is the theory and use, NOT the tools that should steer our
course. How many of you remember the early SGML tools? Don't they are gone as
will be many of today's tools.
Two ontologies, two different discussion groups. They can borrow from each
other as appropriate but labeling of postings would probably not work without
some tool that enforces tagging by ontology type.
-John Bottoms, past Electronic Manuscript Committee member (developer of the
book tag set) and
inventor of the web browser.
FirstStar Systems
Concord, MA USA
On 2/3/2013 10:16 AM, Sjir Nijssen wrote:
To all,
I find it amazing
how at this stage in the development of “ontology”, one can
describe a protocol how to develop such an ill defined artefact as ontology.
Leo Obrst has
proposed to this forum to introduce the concept of a spectrum with several
different concepts of ontology, each with its specific distinguishing
name and characteristics.
I would recommend
to this forum to spend time on operationalizing Leo’s idea.
Thereafter (having the structure) and not before, the work on a protocol
(process, methodology) can start.
Other people would
say that as long as a concept is a homonym, it is nearly impossible to
communicate usefully about it.
In the world of
software and conceptual databases the development of methodologies has been
going on for two decades, the seventies and the eighties. It seems that for
most ontologists history began thereafter.
If one or more
points on the “ontology” spectrum have to become mainstream, I
would recommend not to let the ontology homonym era last too long.
Kind regards