On Wed, October 24, 2012 21:12, Hans Polzer wrote: (01)
> At the risk of extending the discussion, I believe you also need to
> specify the frame of reference used for your metric(s) (02)
This is necessary for any term defined in an ontology. (03)
> and the scope of the frames of reference used. (04)
On definition, one should define a context. One cannot know the
full scope of where the terms will be used. But in order to use a
theory (set of axioms) about some entities in the ontology, the user
should accept the context of that theory. New contexts and theories
are fine to use, so long as they inherit from the context in which the
theory is expressed. (05)
The same entities can have different theories expressed in different
contexts (US Tax Code of 2000, US Tax Code of 2012) such that
reasoning with the same knowledge base would come to different
conclusions. (06)
> Often these are institutional in nature (like what
> constitutes a planet or who is an employee), (07)
In the first case, new definitions are made for English words. For
ontological engineering purposes, a new class/type should be defined
with the new meaning. Rules for mapping between the NL words or
phrases should be defined and contextualized for the prior mappings. (08)
In the second case, few institutions are interested in whether a person
is an employee or not (IRS, unemployment bureaus, ... excepted). In
most cases, an institution's knowledge base is interested in the property
of whether a person is an employee of that institution -- best encoded
ontologically either as a state of affairs which may have many properties
associated with it, or as a temporal relation between the person and the
institution. (09)
> and we need to be up front about that. (010)
Agreed. (011)
> I know a lot of people want to see some absolute,
> context-independent, ontologies. (012)
Contexts are certainly necessary. (013)
> But there aren't any because language
> itself is a human social/institutional creation. (014)
Language issues are the provenance of NL. Certainly NL research
informs ontological research. But ontological design need not be
constrained by the vagaries of any individual natural language.
[The vagaries of the ontological language chosen for encoding are
a different matter. 8)# ] (015)
-- doug foxvog (016)
> Hans (017)
> -----Original Message-----
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Pat Hayes
> Sent: Wednesday, October 24, 2012 1:31 PM
> To: [ontolog-forum]
> Subject: Re: [ontolog-forum] Ontologies vs. Web Ontologies
>
> On Oct 24, 2012, at 11:39 AM, doug foxvog wrote:
>
>> On Tue, October 23, 2012 20:22, Liam D. Gray wrote:
>>> ...
>>> And event might be a thing described by date/time. But it kinda makes
>>> sense for date/time to be metrics rather than things.
>>
>> It depends upon what "thing" means. Aren't metrics something?
>
> Yes, they are.
>
> I would suggest that arguing about what a "thing" is, and any reference to
> philosophical views or debates on the nature or scope of existence, is
> counterproductive in this forum, and that anyone who has strong views on
> this matter should submit them to an appropriate academic journal. As far
> as
> ontologies are concerned, a "thing" is anything that you use a name or
> expression to refer to. If you want to use a expression which denotes a
> metric, then you are talking about metrics, and metrics are things in your
> ontology. End of discussion.
>
> Pat
>
>
>
> _________________________________________________________________
> 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
>
>
>
> _________________________________________________________________
> 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
>
> (018)
_________________________________________________________________
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 (019)
|