RE: SUO: Organizations/Positions
Hi Martin,
Thanks for your helpful comments. See my replies below.
-Ian
> -----Original Message-----
> From: Martin King [mailto:martin_king@UK.IBM.COM]
> Sent: Monday, September 03, 2001 4:01 AM
> To: standard-upper-ontology@ieee.org
> Subject: Re: SUO: Organizations/Positions
>
>
>
> Ian, I support your proposal in the context of the current
> structure of
> SUMO with comment.
>
> 1. The informal discussion in your point 2 about assets and
> liabilities
> risk confusion as these words are important to use in
> relationship to Legal
> Persons (alias Legal Entities), and the corresponding
> relationships with
> OU's, though similar in some ways, are importantly different
> from the point
> of view of a jurisdiction.
>
> 2. As at least one other person has pointed out, the limitation of one
> person to a position is too strict. I would like at least to see
> occupiesPosition recognized as limited in time. Some people
> might call
> that having temporal extent, others as having having a from
> and too date (&
> time?). In line with what I suggested several months ago, I
> would actually
> most prefer two concepts (instances?), assignedtoPosition (with a
> date/time), and removedfromPosition (also with date/time).
> With these two,
> one can also write a rule to deduce occupiesPosition.
> Whether one then
> wants further concepts to handle job-sharing or hand-over periods is a
> level of detail we may not wish to explore.
I agree with your criticism that 'occupiesPosition' should not require that
every position be filled by at most one person, and I've stripped out the
offending axiom. As for the temporal qualification of 'occupiesPosition',
you can just use the general temporal predicate 'holdsDuring'. Thus, the
fact that John occupied the position of research director on May 27, 1998
would be expressed as follows:
(holdsDuring (DayFn 27 (MonthFn 5 1998)) (occupiesPosition John
ResearchDirector))
I like your suggested predicates "assignedToPosition" and
"removedFromPosition", and I've tried to define them formally below:
(instance assignedToPosition QuaternaryPredicate)
(domain assignedToPosition 1 Human)
(domain assignedToPosition 2 Position)
(domain assignedToPosition 3 Organization)
(domain assignedToPosition 4 TimePosition)
(documentation assignedToPosition "(assignedToPosition ?PERSON ?POSITION
?ORG ?TIME) means that ?PERSON first occupied the position ?POSITION at ?ORG
at ?TIME.")
(=>
(assignedToPosition ?PERSON ?POSITION ?ORG ?TIME)
(and
(not
(holdsDuring (ImmediatePastFn ?TIME) (occupiesPosition ?PERSON
?POSITION ?ORG)))
(holdsDuring (ImmediateFutureFn ?TIME) (occupiesPosition ?PERSON
?POSITION ?ORG))))
(instance removedFromPosition QuaternaryPredicate)
(domain removedFromPosition 1 Human)
(domain removedFromPosition 2 Position)
(domain removedFromPosition 3 Organization)
(domain removedFromPosition 4 TimePosition)
(documentation assignedToPosition "(removedFromPosition ?PERSON ?POSITION
?ORG ?TIME) means that ?PERSON was removed from the position ?POSITION at
?ORG at ?TIME.")
(=>
(removedFromPosition ?PERSON ?POSITION ?ORG ?TIME)
(and
(not
(holdsDuring (ImmediateFutureFn ?TIME) (occupiesPosition ?PERSON
?POSITION ?ORG)))
(holdsDuring (ImmediatePastFn ?TIME) (occupiesPosition ?PERSON
?POSITION ?ORG))))
>
> As a much more general point, I question the practical value
> of the formal
> structure of OrganizationUnit being a sub-class of
> CognitiveAgent, but I
> see this as just one example of a fundamental question to the
> structure of
> SUMO. I am prepared to go along with the current direction because I
> believe it may have useful results even if I later challenge
> higher levels
> of the structure.
I agree that this is a tough call. I think it's clear that
'OrganizationUnit' should be a subclass of 'Agent', but there may be
persuasive reasons not to make it a subclass of 'CognitiveAgent'. I'd be
interested if anyone can think of concrete counterexamples to this placement
of the concepts.
>
> Yours, Martin
>
> IBM Global Services PHONE: +44-1707-363090 (Int 7-453090)
> Rosanne House (RH2A) FAX: +44-1707-338732
> Welwyn Garden City Internet: martin_king@uk.ibm.com
> GB - AL8 6UB IBMMAIL: GBIBM3WS
>
> There are two ways of constructing a software design. One is
> to make it so
> simple that there are obviously no deficiencies; the other is
> to make it so
> complicated that there are no obvious deficiencies. The
> first method is
> far more difficult.
>
>
> Ian Niles <iniles@teknowledge.com>@majordomo.ieee.org on 2001-08-31
> 23:52:16
>
> Please respond to Ian Niles <iniles@teknowledge.com>
>
> Sent by: owner-standard-upper-ontology@majordomo.ieee.org
>
>
> To: "Standard-Upper-Ontology (E-mail)"
> <standard-upper-ontology@ieee.org>
> cc:
> Subject: SUO: Organizations/Positions
>
>
>
>
> Hi All,
>
> I've been trying to formalize part of the exchange that we've had
> recently on the concepts of organization, postion within an
> organization,
> etc. The formal definitions for the new concepts and revised
> SUMO concepts
> is presented at the end of this message, but these definitions give
> essentially the following structure, which I think both
> Martin King and
> Chris Partridge accept:
>
> CognitiveAgent
> |
> |
> OrganizationUnit
> / \
> / \
> / \
> Organization Position
>
> Together with this structure is a set of predicates (also
> formally defined
> at the end of this message): 'occupiesPosition' (relates a
> person to the
> position they occupy within an organization),
> 'subOrganizations' (relates
> one organization to another of which it is a part), and
> 'employs' (relates
> a
> person to the organization of which he/she is a member).
>
> As you might recall, I tried in a previous email to
> extract various
> criteria for organizations from some of the traffic on this
> subject. These
> criteria are as follows:
>
> 1. An organization has agency, e.g. it exhibits
> intentionality, and
> it has rights, responsibilities, and obligations.
>
> 2. An organization may have members, but it is not
> required to have
> members. There are many examples of organizations, e.g.
> corporations and
> churches, that have assets, are liable for certain claims
> etc, even though
> they have no members.
>
> 3. An organization has temporal extent. It comes into
> being at a
> certain point in time, and it goes out of existence at another
> point.
>
> 4. An organization can have various sorts of members.
> Owners are
> members of organizations, and employees, directors, and other
> stakeholders may also be members of organizations
> (although perhaps in
> different senses). An organization may also have other
> organizations as
> members.
>
> I think that the structure and predicates presented
> informally above and
> formally below make significant headway in satisfying all of these
> criteria.
> Let's consider them in turn. As for 1 and 3, these are
> satisfied for the
> same reason that they were before. In the SUMO, 'Organization' is a
> subclass of 'Agent' (more specifically, 'CognitiveAgent' in the new
> proposal) and 'Agent' is a subclass of 'Object', so organizations are
> agents
> and they have a position in space-time.
>
> As for criteria 2 and 4, these were not completely satisfied
> by my earlier
> proposed function 'GroupAgentFn', but I think the concepts
> outlined above
> do
> satisfy them. Consider criterion 2. The sticking point here is the
> stipulation that organizations are not required to have
> members. In the
> structure above, we distinguish two senses of organization, viz.
> 'OrganizationUnit' and 'Organization'. The latter is required to have
> members (in fact, more than one member), because it is a subclass of
> 'Collection' (the details about this are presented in the
> formal section
> below). However, 'OrganizationUnit' is not a subclass of
> 'Collection',
> because it is meant to cover both organizations and
> positions, so there is
> no requirement that an 'OrganizationUnit' have any members.
> Accordingly,
> we
> now have a notion of organization that covers "empty"
> organizations. As
> for
> the fourth criterion, the new predicate 'occupiesPosition' allows us
> distinguish, via 'Position', all of the various member types of an
> organization. Another advantage of this predicate is that we
> can account
> for the important fact that the person who occupies a
> position may have
> different rights, responsibilities, etc. from the position
> itself. This is
> because, on the current proposal, the agent who fills the
> first slot of
> 'occupiesPosition' is potentially a different agent from the agent who
> fills
> the second slot of this predicate. In some cases, the two
> agents would be
> the same, but establishing this would require axioms specific to the
> position in question.
>
> =======================
> Formal SUMO Definitions
> =======================
>
> (subclass OrganizationUnit CognitiveAgent)
> (documentation OrganizationUnit "An &%Organization or a
> functional unit
> within an &%Organization, e.g. positions, divisions, and
> departments. For
> example, the Shell Corporation, the accounting department at
> Shell, the
> positions of CEO and mail room supervisor at Shell, etc. would all be
> instances of &%OrganizationUnit.")
>
> (subclass Organization OrganizationUnit)
> (subclass Organization GroupOfPeople)
> (documentation Organization "An &%Organization is a corporate or
> similar institution, distinguished from other &%Agents. The &%members
> of an &%Organization typically have a common purpose or function.
> The continued existence of an &%Organization is not dependent
> on any of
> its members, its location, or its particular facility. Note
> that parts
> of &%Organizations should not be included here, unless they are
> &%subOrganizations of an &%Organization.")
>
> (subclass Position OrganizationUnit)
> (relatedInternalConcept Position occupies)
> (documentation Position "A formal position of reponsibility within an
> &%Organization. Examples of &%Positions include president, laboratory
> director, senior researcher, sales representative, etc.")
>
> (instance occupiesPosition TernaryPredicate)
> (domain occupiesPosition 1 Human)
> (domain occupiesPosition 2 Position)
> (domain occupiesPosition 3 Organization)
> (documentation occupiesPosition "(&%occupiesPosition ?PERSON ?POSITION
> ?ORG)
> means that ?PERSON holds the &%Position ?POSITION at
> &%Organization ?ORG.
> For example, (&%occupiesPosition &%TomSmith &%ResearchDirector
> &%AcmeLaboratory) means that &%TomSmith is a research director at Acme
> Labs.")
>
> (=>
> (and
> (occupiesPosition ?PERSON1 ?POSITION ?ORG)
> (occupiesPosition ?PERSON2 ?POSITION ?ORG))
> (equal ?PERSON1 ?PERSON2))
>
> ;; The axiom above stipulates that no more than one person
> can occupy a
> given
> ;; position. Note that it follows from axioms already in the
> SUMO that an
> ;; Organization consists of more than one person, because
> 'Organization' is
> a
> ;; subclass of 'GroupOfPeople', which is a subclass of
> 'Group', which is a
> ;; subclass of 'Collection', and we have the following axiom:
>
> (=>
> (instance ?COLL Collection)
> (exists (?OBJ1 ?OBJ2)
> (and
> (member ?OBJ1 ?COLL)
> (member ?OBJ2 ?COLL)
> (not
> (equal ?OBJ1 ?OBJ2)))))
>
> (=>
> (occupiesPosition ?PERSON ?POSITION ?ORG)
> (employs ?ORG ?PERSON))
>
> (subrelation employs member)
> (domain employs 1 Organization)
> (domain employs 2 Human)
> (documentation employs "(&%employs ?ORG ?PERSON) means that ?ORG has
> hired ?PERSON and currently retains ?PERSON, on a salaried or
> contractual basis, to provide services in exchange for monetary
> compensation.")
>
> (subrelation subOrganizations subCollection)
> (domain subOrganizations 1 Organization)
> (domain subOrganizations 2 Organization)
> (documentation subOrganizations "(&%subOrganizations ?ORG1
> ?ORG2) means
> that ?ORG1 is an &%Organization which is a proper part of the
> &%Organization ?ORG2.")
>
>