Dear Matthew, Rich, Pat C, David, and Steven, (01)
The fundamental distinction is between a normative ontology intended
as a standard and a descriptive ontology that specifies a broad range
of uses for the terms. (02)
MW
> We were defining standards for use throughout the supply chain... (03)
That's good. But the point is the same no matter how large the group
of people involved: normative standards specify the definitions of
terms for those who agree to abide by the norms. (04)
JFS
>> But we also need to design our systems to interoperate in a world of
>> people and systems over which we have no control. (05)
MW
> Of course we do. That is what we have been doing, and you saying
> otherwise does not make it so. (06)
I apologize if anything I said was offensive. I have a very high
regard for your work, which I believe is important and should be
continued. And I was not suggesting that your standards can't
interoperate with systems that use other standards. (07)
JFS
>> we must also relate the normative concepts to the descriptive concepts
>> about a wide range of systems whose design is outside our control. (08)
MW
> It is the other way round. We need to relate the concepts in a wide
> range of systems to the normative standards. That gives some interesting
> requirements on those standards:
> - The standard needs to be able to have any concept from those legacy
> system mapped to it, which in turn means,
> - The standard needs to be extensible, but stable (you don't want to have
> to change what you have when you add something new - it's probably already
> being used) (09)
I agree with those points. My only qualification is about the phrase
"the other way round". It reminds me of a joke in the old Soviet Union:
> Under Capitalism, man exploits man.
> But under Communism, it's the other way round. (010)
It's common practice in ISO to use the OED for the base definitions
of English words. Those definitions are descriptive, not normative. (011)
Normative standards by ISO or any other organization add further
constraints to the underspecified descriptive terms in order to
specify a more precise, but narrower sense for the normative use. (012)
JFS
>> For example, Cyc can give underspecified definitions for time and space
>> that avoid any commitment to a 4D or a 3+1 D ontology. Then different
>> microtheories could add axioms that complete the spec (013)
MW
> Yes, but can it translate between them? Unless you have the means to
> bring different ways of describing the same thing into a single
> representation, you cannot bring together data from different sources. (014)
Two points: (015)
1. For most purposes, no translation is necessary. Names, addresses,
date of birth, and geographical coordinates are the same for both. (016)
2. The CycL logic is sufficiently expressive to specify any translation
that can be carried out by a Turing machine or any digital computer. (017)
MW
> ... for an integrating ontology, you only want one way to say something,
> but you want it to be the most expressive way of saying it, so that
> any other way can be expressed that way. (018)
In any version of logic, there are many equivalent ways of saying
anything. Even in simple Boolean logic, "p & q" is equivalent to
"q & p", "~(~p or ~q)", "~(if p, then ~q), ... (019)
Instead, I would say that a normative ontology should be expressed
in a formal logic and with precise criteria for conformance. (020)
RC
> There is a great deal of defensive benefit to corporations in certain
> circumstances keeping their products incompatible with other companies’
> products. (021)
Yes, but I would also add that there are sometimes huge benefits
to society when somebody gets "a better idea". In such cases,
many of the old standards become obsolete and counterproductive. (022)
PC
> I still believe that identifying the most primitive concepts and representing
> those would provide us (at the least cost) with a top-level ontology that can
> be used for integration at that most general level. (023)
I believe that your suggestion of the 2000 defining terms of Longman's
dictionary would be a good goal for a general upper level ontology.
An upper level along those lines could serve as a *descriptive* ontology
that could relate an open-ended variety of *normative* ontologies. (024)
DE
> I would argue that for all intents & purposes, INTEGRATION is not possible.
>
> The realistic goal should be INTEROPERABILITY rather than integration. (025)
Complete integration is only possible for those groups that adopt
a particular standard. But computer systems have been interoperating
since the 1950s with punched cards carried from one computer to another. (026)
SEZ
> ... manuals must be automatically generated from the implementation
> language (not comments therein) and compared to the expectations set
> upon them. Only then can you be sure of conformance. (027)
I strongly agree. The normative specifications should be stated in
a highly expressive formal logic and automatically translated to
(a) programming languages and (b) humanly readable languages. (028)
As an example of the formal logic, I would cite Common Logic. For (a),
I would cite the past 40 years of R & D in logic programming and in
compiler theory and practice. For (b), I would cite Common Logic
Controlled English (CLCE). But no progress will be made unless and
until we recognize that this goal is possible and necessary. (029)
John (030)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (031)
|