SUO: RE: conformance
Folks,
I have not been following the debate incredibly closely but would not one of
the things the users of SUO would like - for inter-operability - is some
comfort that the SUO is being used with sufficiently similar intended
interpretations: that different groups developing the SUO will be using the
terms to talk about the same things. If this is so would we not want one
level of conformance to be a reasonable guarantee that the intended
interpretations are converging.
Or will it be argued that this is, in principle, impossible.
It is normal for conformance to standards such as SWIFT to do this in a way
that guarantees a degree of confidence that the conforming systems are
referring to the same things - important if you are exchanging instructions
for the payment of large sums of money etc.
If we include a notion such as Asset or Currency or Transaction in the SUO,
are we expecting the 'systems' that conform to the SUO to be talking about
the same things?
A number of philosophers (David Wiggins and Ruth Marcus spring to mind) have
suggested that it is impossible to understand what many 'concepts' mean
unless you are familiar with some examples/instances - either of them or
related concepts. This is one way in which SWIFT works - providing
lists/examples of the instances.
This closely relates to another point - the 'teachability' of the SUO terms.
It seems to me merely publishing KIF axioms - and some brief descriptions -
is insufficient to assure that the terms are used in similar ways by
different groups of people. If there is a requirement for them to be used in
sufficiently similar ways, surely there needs to be some though about how
people can be taught to use them in this way. Of course if there is no such
requirement, this should be made clear.
It seems to me that mere (logical) consistency is not necessarily the most
important aspect of conformance for many operational applications.
If it is not regarded as part of the job of the SUO to tie down intended
interpretations - then the users need to be clearly warned.
Chris
-----Original Message-----
From: owner-standard-upper-ontology@majordomo.ieee.org
[mailto:owner-standard-upper-ontology@majordomo.ieee.org]On Behalf Of Adam
Pease
Sent: 08 October 2001 23:47
To: standard-upper-ontology@ieee.org
Subject: SUO: conformance
Folks,
Here's a draft conformance clause (which I believe could apply to both
IFF and SUMO). Constructive feedback is welcome.
- Implementations of SUO are "ontologies" or "information models".
- A conforming implementation is an ontology or information model that:
[#1] Uses terms provided in the SUO, or
[#2] Uses terms that are defined (using SUO-KIF) by other terms in the
SUO
[#3] Is consistent: a contradiction cannot be derived by means of
first-order logic from the set of statements belonging to the
implementation and the SUO.
Adam
Adam Pease
Teknowledge
(650) 424-0500 x571