W3C home > Mailing lists > Public > www-rdf-interest@w3.org > December 2002

Re: KR and RDF Redux

From: Charles McCathieNevile <charles@w3.org>
Date: Thu, 12 Dec 2002 03:18:39 -0500 (EST)
To: "Richard H. McCullough" <rhm@cdepot.net>
cc: Seth Ladd <seth@brivo.net>, <www-rdf-interest@w3.org>
Message-ID: <Pine.LNX.4.30.0212120316130.32507-100000@tux.w3.org>

On Wed, 11 Dec 2002, Richard H. McCullough wrote:

>KR does have canonical forms.

Yes, but my point is that there is often little need for them, since the
ideas we are representing don't themselves have canonical forms. Anything
that has the information, and can interoperate, works...

(mental exercise - can we represent this discussion?)

Chaals

>Where RDF has one form (the triple) for all properties,
>KR has a different form/syntax for each different type of property:
>attribute, part, action, relation.
>============
>Dick McCullough
>knowledge := man do identify od existent done
>knowledge haspart proposition list
>
>  ----- Original Message -----
>  From: Charles McCathieNevile
>  To: Seth Ladd
>  Cc: www-rdf-interest@w3.org
>  Sent: Wednesday, December 11, 2002 9:34 AM
>  Subject: Re: KR and RDF Redux
>
>
>
>  I think the real point is that we are trying to represent information that
>  does not have a single canonical form in natural language, so the particular
>  syntax we use to represent it isn't that important given that it is selected
>  from among the many that meet the requirements. (There, of course lie the
>  details wherein the devil hides).
>
>  'John takes a walk' uses 'walk' as a noun
>  'John walks' uses it as a verb.
>
>  In some cases there are shades of different meaning, in others they are
>  equivalent statements. If you have a case where they are meaningfully
>  different then this discussion isn't getting fine enough and this example
>  isn't relevant anyway, but it does seem to illustrate the general point.
>
>  Cheers
>
>  Chaals
>
>  On Wed, 11 Dec 2002, Seth Ladd wrote:
>
>  >
>  >Charles McCathieNevile wrote:
>  >> journey
>  >>   by John
>  >>   from hisHouse
>  >>   to theStore
>  >>   purpose
>  >>     purchase milk
>  >>
>  >> (all except the purpose bit is readily available in the work that Dan
>  >> Connolly, Jim Ley, I and others have done in encoding flight information.
>  >> Adding a purpose for the journey is a good idea - especially since our use
>  >> case often involves going to meetings (already expresed as RDF) and the rest
>  >> seems to be trying to meet friends (fair amount of data available in FOAF)
>  >
>  >I think the point here is that the solution to "John walked to the store"
>  >is turned inside out from a verb to a noun.  Create a noun (object) that is
>  >the result of the action performed.  Then you can add many properties to
>  >this object and talk about it.
>  >
>  >This also helps when trying to use CRUD functions on everything (in an
>  >attempt to follow REST principles).  Inside of "door.enter()" you can
>  >"Entrance.create()".  Once I've done that, I can then talk about (apply
>  >properties to) an Entrance, because it is now a Thing.
>  >
>  >Seth
>  >
>
>  --
>  Charles McCathieNevile  http://www.w3.org/People/Charles  tel: +61 409 134 136
>  SWAD-E http://www.w3.org/2001/sw/Europe ------------ WAI http://www.w3.org/WAI
>   21 Mitchell street, FOOTSCRAY Vic 3011, Australia  fax(fr): +33 4 92 38 78 22
>   W3C, 2004 Route des Lucioles, 06902 Sophia Antipolis Cedex, France
>
>

-- 
Charles McCathieNevile  http://www.w3.org/People/Charles  tel: +61 409 134 136
SWAD-E http://www.w3.org/2001/sw/Europe ------------ WAI http://www.w3.org/WAI
 21 Mitchell street, FOOTSCRAY Vic 3011, Australia  fax(fr): +33 4 92 38 78 22
 W3C, 2004 Route des Lucioles, 06902 Sophia Antipolis Cedex, France
Received on Thursday, 12 December 2002 03:18:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:57 GMT