W3C home > Mailing lists > Public > public-owl-dev@w3.org > July to September 2007

RE: Legal Persons

From: John McClure <jmcclure@hypergrove.com>
Date: Fri, 24 Aug 2007 13:56:43 -0700
To: "Mark Montgomery" <markm@kyield.com>, "Owl Dev" <public-owl-dev@w3.org>
Message-ID: <MGEEIEEKKOMOLNHJAHMKOEMMEFAA.jmcclure@hypergrove.com>

Mark,
I agree with your sentiment. My belief has been that a legal ontology pertinent
to a certain jurisdiction should be defined pretty much solely with respect to
definitions found in jurisdictional legal material (statutes, opinions,
charters, constitutions, regulations, etc) -- hence, my own focus is on RDFA
annotations of HTML versions of those documents. A generic legal ontology
applicable across jurisdictions, while difficult to devise, is nevertheless
critical for applications that aim to support mutliple jurisdictions -- they
need as much as possible a common foundation of classes and properties (or a
heckuva lot of sameAs relations).

The notion of 'context' is relevant here; perhaps the RDF community could
benefit from ebXML's designation of five axes defining context -- eg location,
i.e., jurisdiction -- which I hope eventually will become standardized
owl:Ontology & rdf:Statement properties.... In the meantime, we're having to
create a class like "USAThing" that can be mixed in with those pertinent to US
jurisdictions (assuming the constraint of a single namespace), which is a
subclass of "JurisdictionalThing" which has a property "Jurisdiction" and so on.
Or a similar class or property annotation, but there are problems with that
approach.

Anyway your point that Corporation is more a LegalEntity than a LegalPerson is
valid though it boils down to what properties and restrictions are defined for
LegalEntity, LegalPerson, Human & Corporation primary of which is the set of
Rights accorded to each.  Those rights are defined by the documents mentioned
above, so it appears plainly to me that a generic legal ontology should eschew
defining properties and restrictions practically altogether, leaving just a
nominal, flat vocabulary, and instead look to the annotations of legal writs for
jurisdictionally-relevant definitions of those properties and restrictions.

That is why RDFA is so important and hence, OWL 1.1's class & property axioms.

And thanks to all for the comments about the markup. Of course, unionOf !
John McClure

>-----Original Message-----
>From: public-owl-dev-request@w3.org
>[mailto:public-owl-dev-request@w3.org]On Behalf Of Mark Montgomery
>Sent: Friday, August 24, 2007 12:23 PM
>To: John McClure; Owl Dev
>Subject: Re: Legal Persons
>
>
>
>A good example of a significant challenge facing the global medium since
>becoming public anyway. While I am not an attorney, I am surrounded by them,
>and I can say with some confidence that for many purposes a corporation is
>not considered a person in the U.S., but rather a legal entity, so I believe
>it will require substantially more sub classification for most applications.
>However, as the question suggests, as soon as the data passes a border, it
>will likely have different meaning entirely. One of the fundamental
>challenges facing one discipline attempting to create standards that affect
>others, and why I consider it to be a mega-disciplinary field. .02- MM,
>Kyield
>
>----- Original Message -----
>From: "John McClure" <jmcclure@hypergrove.com>
>To: "Owl Dev" <public-owl-dev@w3.org>
>Sent: Friday, August 24, 2007 12:04 PM
>Subject: Legal Persons
>
>
>>
>> A Person in US legal contexts is either a Human or a Corporation; every
>> Human is
>> a Person, and every Corporation is a Person.
>>
>> Is the following construct valid? Will or should reasoners be troubled by
>> <rdf:Alt> within a <rdfs:range>, and can or should <rdf:Alt> be used
>> within an
>> <owl:Restriction>?
>>
>> <owl:Class rdf:about="#LegalPerson">
>>     <owl:ObjectProperty rdf:about='#Parent'>
>>        <rdfs:range>
>>   <rdf:Alt>
>>        <li><owl:Class rdf:about="#Human"/></li>
>>        <li><owl:Class rdf:about="#Corporation"/></li>
>>   </rdf:Alt>
>>        </rdfs:range>
>>     </owl:ObjectProperty>
>>     <owl:subClassOf>
>>          <owl:Restriction>
>>             <owl:onProperty rdf:resource="#Parent"/>
>>             <owl:maxCardinality rdf:value='1'/>
>>          </owl:Restriction>
>>     </owl:subClassOf>
>> </owl:Class>
>> <owl:Class rdf:about="#Human">
>>     <owl:subClassOf rdf:resource="#LegalPerson"/>
>>     <owl:subClassOf>
>>          <owl:Restriction>
>>             <owl:onProperty rdf:resource="#Parent"/>
>>             <owl:allValuesFrom rdf:about="#Human"/>
>>          </owl:Restriction>
>>     </owl:subClassOf>
>>     <owl:subClassOf>
>>          <owl:Restriction>
>>             <owl:onProperty rdf:resource="#Parent"/>
>>             <owl:maxCardinality rdf:value='2'/>
>>          </owl:Restriction>
>>     </owl:subClassOf>
>> </owl:Class>
>> <owl:Class rdf:about="#Corporation">
>>     <owl:subClassOf rdf:resource="#LegalPerson"/>
>>     <owl:subClassOf>
>>          <owl:Restriction>
>>             <owl:onProperty rdf:resource="#Parent"/>
>>             <owl:allValuesFrom rdf:about="#Corporation"/>
>>          </owl:Restriction>
>>     </owl:subClassOf>
>>     <owl:subClassOf>
>>          <owl:Restriction>
>>             <owl:onProperty rdf:resource="#Parent"/>
>>             <owl:maxCardinality rdf:value='1'/>
>>          </owl:Restriction>
>>     </owl:subClassOf>
>> </owl:Class>
>>
>> I haven't found examples of this sort in the docs. I understand that an
>> <rdfs:range> is nothing more than
>>     <!-- alternative specification for rdfs:range-->
>>     <owl:subClassOf>
>>          <owl:Restriction>
>>             <owl:onProperty rdf:resource="#Parent"/>
>>             <owl:allValuesFrom>
>>   <rdf:Alt>
>>        <li><owl:Class rdf:about="#Human"/></li>
>>        <li><owl:Class rdf:about="#Corporation"/></li>
>>   </rdf:Alt>
>>             </owl:allValuesFrom>
>>          </owl:Restriction>
>>       </owl:subClassOf>
>>
>> Maybe there is a better way to model this? Maybe I should be asking this
>> in
>> another forum? Thanks for any comments,
>> John McClure
>>
>>
>
>
Received on Friday, 24 August 2007 20:56:16 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 10 December 2014 20:07:18 UTC