ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] predicates in RDF

To: paola.dimaio@xxxxxxxxx
Cc: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Pat Hayes <phayes@xxxxxxx>
Date: Wed, 2 Jan 2008 00:44:06 -0800
Message-id: <p0623090dc3a0fdea7e1e@[192.168.1.6]>
>Thanks John and Pat
>
>I admit, I had a broad range of reactions to that statement
>
>>  # the predicate is the word "creator"
>(shrieeekkk, what?, surpise, dispair, bewilderment, incredulity)    (01)

Look, the fact that it is called "predicate" is of no deep 
significance. In fact, through most of the RDF literature, the middle 
URI of a triple is called a "property". But you can call it a 
"foodle" if you like. What you call it is not important. What matters 
is what its role is in the RDF formalism.    (02)

>Maybe thats because the semantic web is being designed by
>mathematicians and computer scientists, who have little appreciation
>for natural language?
>
>>
>>  >why would RDF not reflect/embed/support a basic rule of 
>>linguistic semantics?
>>
>>  Why would it?
>
>1. Because RDF is a framework to provide a structure to natural
>language which is part of the linguistics domain.    (03)

No, that is not what RDF is. It has almost nothing at all to do with 
NL. Where did you get that idea??    (04)

>  If I need to get
>'reasoners' to work with RDF then I need to have the logic expressed
>by the language reflected by the semantic structure    (05)

Indeed. The RDF specification documents give a semantics for RDF 
based on the rather minimal syntax of 'triples'. That meaning is what 
is conveyed by RDF and accessible to an RDF reasoner. In fact, there 
is a complete set of inference rules for RDF reasoning in the RDF 
spec.    (06)

>, else I have 'loss
>of meaning' big time
>
>2. otherwise RDF will not work for many applications    (07)

What kind of applications, and how will it not work?    (08)

Pat    (09)

>
>
>
>>  >is the question ill posed, or is RDF ill posed, in this respect?
>>  >
>
>
>I
>>  >
>>  >On 1/2/08, Chris Menzel <cmenzel@xxxxxxxx> wrote:
>>  >>  On Tue, 1 Jan 2008, paola.dimaio@xxxxxxxxx wrote:
>>  >>  > Coming from a linguistic background, I have always thought that
>>  >>  > predicate is a verb
>>  >>
>>  >>  Actually, the predicate of a sentence is usually identified with a verb
>>  >>  *phrase*, the part of the sentence that "makes the assertion about the
>>  >>  subject", as in your definition:
>>  >>
>>  >>  > CF. The predicate of a sentence is the part of the sentence that makes
>>  >>  > the assertion about the subject. The main part of the predicate is a
>>  >>  > finite verb (which must be present). The predicate can be a verb
>>  >>  > alone, or a verb and other words related to it. ...
>>  >>  > www.ntgreek.org/learn_nt_greek/terms.htm
>>  >>
>>  >>  Thus, e.g., the predicate of "John runs" is "runs" and the predicate of
>>  >>  "John loves Mary" is "loves Mary".
>>  >>
>>  >>  > but on the W3C spec, there is no such requirement/constraint
>>  >>  >
>>  >>  > http://www.w3.org/TR/rdf-concepts/#ref-rdf-semantics
>>  >>  >
>>  >>  > Is there no such implication that predicate must be a verb in rdf?
>>  >>
>>  >>  The question is ill-formed, as there is no such concept as "verb" in
>>  >>  RDF.  Indeed, there are no predicates per se in RDF, there are only
>>  >>  names (more specifically, URIs and literals) which denote "resources".
>>  >>  "predicate" itself just indicates a *role* that a name can play in an
>>  >>  RDF triple -- it is the name that "connects" the other two names in the
>>  >>  triple.  Any name can be the predicate in a triple.  That said, if an
>>  >>  RDF triple is used to represent a simple natural language sentence like
>>  >>  "John loves Mary" -- "<ex:j> <ex:l> <ex:m>", say -- the predicate in the
>>  >>  triple "<ex:l>" does in fact correspond to the verb in the sentence in
>>  >>  question.
>>  >>
>>  >>  The W3C RDF semantics document (authored chiefly by Pat Hayes)
>>  >>  http://www.w3.org/TR/2004/REC-rdf-mt-20040210 provides a very clear
>>  >>  account of these matters, though, it assumes some familiarity with basic
>>  >>  mathematical logic and set theory.
>>  >>
>>  >>  -chris
>>  >>
>>  >>
>>  >>  _________________________________________________________________
>>  >>  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
>>  >>
>>  >>
>>  >
>>  >
>>  >--
>>  >Paola Di Maio
>>  >School of IT
>>  >www.mfu.ac.th
>>  >*********************************************
>>  >
>>  >_________________________________________________________________
>>  >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
>>  >
>>
>>
>>  --
>>  ---------------------------------------------------------------------
>>  IHMC            (850)434 8903 or (650)494 3973   home
>>  40 South Alcaniz St.    (850)202 4416   office
>>  Pensacola                       (850)202 4440   fax
>>  FL 32502                        (850)291 0667    cell
>>  phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
>>
>>
>
>
>
>--
>Paola Di Maio
>School of IT
>www.mfu.ac.th
>*********************************************    (010)


-- 
---------------------------------------------------------------------
IHMC            (850)434 8903 or (650)494 3973   home
40 South Alcaniz St.    (850)202 4416   office
Pensacola                       (850)202 4440   fax
FL 32502                        (850)291 0667    cell
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes    (011)


_________________________________________________________________
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    (012)

<Prev in Thread] Current Thread [Next in Thread>