Dear John, (01)
I think the data modelling community also thought it had been solved a long
time ago, but came up with a different answer :-) (02)
Firstly, I can see what you intend by your idea of employee. (03)
Consider this case. (04)
If I joined IBM ten years ago.
If I then joined HP six years ago (while still being employed by IBM),
Then five years ago I left IBM. (05)
Throughout the last 10 years I was an employee. (06)
Your notion of employee does not distinguish between employers - and it is
also temporalised - one is an employee at a time. (07)
JS>Subtype has a very clear and simple definition: X < Y means that every
instance of X is an instance of Y. (08)
It seems that implicitly mean that *at time t1*, X < Y means that every
instance of X is an instance of Y. Maybe better to say that X <t1 Y means
that at t1 every instance of X is an instance of Y. (09)
JS>That is covered by two subtypes of Role called the
PrehendingEntity and the PrehendedEntity. The employer
and the owner are prehending, and the employee and pet
are prehended. That's all in the KR book or on the web: (010)
I am not sure how this would deal with employees with multiple employers at
the same time. It looks as if it would have problems. Could you explain. (011)
However, this is not the notion I find when I look at systems.
Typically a system will say who the person is employed by, when he/she
started employment, and if appropriate when they left employment. (012)
This is the notion that Aristotle was capturing with his master/slave
relationship. (013)
In the example above, I would expect to find:
I was an IBM employee who joined 10 years ago and left five years ago.
I am an HP employee who joined six years ago. (014)
Of course, you could answer, that you can introduce into your scheme
sub-types of employee called IBM employee and HP employee. However, apart
from the name, how is this related to IBM and HP? (015)
However, the case of 'corporation sole' is much clearer in marking the finer
grained requirement. (016)
In your scheme of things, a bishop would be a sub-type of person. If James
were a bishop, you would say that there is only one James who is both a
person and a bishop (at time t1). (017)
However, in English Law there would be two legal persons, James and the
Bishop. If one left ones property to the Bishop, then when James dies in
office, the property would remain the property of the bishop and not pass to
James's heirs. (018)
I do not see how your scheme of things would deal with this requirement.
Again, I would welcome an explanation. (019)
Best regards,
Chris (020)
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of John F. Sowa
Sent: 19 March 2007 14:55
To: [ontolog-forum]
Subject: Re: [ontolog-forum] Ontology and methodology (021)
Dear Chris, Matthew, John, Adrian, and Florian, (022)
I thought that all those issues had been settled decades ago: (023)
CP> Would there be a single object (John), who is an instance
> of both employee and person, in your scheme of things?
> I think you need this to get your sub-type relation. (024)
Of course. (025)
CP> This enables me to allocate responsibility to the employee
> but not to the person. (026)
We don't have two individuals here. There is only one.
If I'm assigned a task as an employee, it's my task, not
the task of some virtual employee. (027)
MW> This is also what we now do with a 4-dimensionalist
> approach. However, it is this approach that also demonstrates,
> as I explained to Duane that employee is not a subtype of
> person, taking person as the person-for-the-whole-of-their-life. (028)
Subtype has a very clear and simple definition: X < Y
means that every instance of X is an instance of Y. (029)
This is true in a 3-D version and in a 4-D version. When a
person stops being an employee, there is no employee. But
as long as a person is an employee, that employee is one
and the same individual as that person. (030)
MW> Let us look at a couple of possibilities here: (031)
This definition covers every case. It is true in a 3-D
version and in a 4-D version. It implies that the entire
spatiotemporal extent of Employee is included in the
spatiotemporal extent of Person. (032)
MW> Note this does not stop roles being subtypes of role (we
> use the term participation). (033)
In the KR ontology (from my KR book), participation is a
subtype of role. (034)
JB> pet is the name of the second participant in
> a having-pet-relationship.
>
> In English, we have some minimal but productive
> morphology for such entities: e.g.,
>
> employer, employee
>
> This provides more background to what it means
> to place something under 'Role'. (035)
That is covered by two subtypes of Role called the
PrehendingEntity and the PrehendedEntity. The employer
and the owner are prehending, and the employee and pet
are prehended. That's all in the KR book or on the web: (036)
http://www.jfsowa.com/ontology/roles.htm
Roles and Relations (037)
JB> Another advantage is that it is easier to relate such
> treatments to a broader range of linguistic expressions
> and in a broader range of languages. (038)
That is one of my primary motivations, and the KR ontology
handles all that beautifully. (039)
AW> In other words, take the purpose of the modelling into account.
> As John Sowa and others have pointed out, natural language evolves
> through analogous usage. (040)
Yes, in natural language both natural kinds and role types
are represented as nouns. It is very important to have a
representation that supports both the similarities and the
distinctions in a systematic way. (041)
Children don't learn the distinctions between names and nouns
at the beginning, and the distinctions between natural kinds
and role types come much later. Furthermore, metaphor and
metonymy keep introducing expanding the relationships in a
systematic way. The formalism must support all the distinctions
and the transitions from one to another. (042)
AW> A third thought is that it may help to express prototypes
> of at least some of the applications in executable English**, (043)
I certainly agree. The version of executable English I have
been developing is called CLCE (Common Logic Controlled English). (044)
FP> I am wondering what a role is ontologically? What are the
> super-categories of ROLE? (045)
See the KR ontology (op. cit.). (046)
FP> Why should employee be a role while chair is not? (047)
For very good reasons: a chair can be recognized by the
properties it has in itself without looking at anything else
(but one can use things that are not chairs *as* a chair).
But you can't tell whether a person is an employee without
considering something else. See the KR book for a discussion
of Firstness, Secondness, and Thirdness. (048)
John (049)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: 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 Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (050)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: 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 Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (051)
|