W3C home > Mailing lists > Public > www-rdf-logic@w3.org > April 2001

RE: Reification

From: pat hayes <phayes@ai.uwf.edu>
Date: Mon, 9 Apr 2001 14:13:11 -0500
Message-Id: <v04210113b6f7b8c7a3e3@[130.107.66.237]>
To: "Danny Ayers" <danny@panlanka.net>
Cc: www-rdf-logic@w3.org
>Undeniably good argument in favour of having the exact meaning/standard
>somewhere, but why can't the standard be left in Paris (or Greenwich), and
>our machines have their own metre rules for day to day use?

THEY CAN!!. Nobody is suggesting that the machines USE the 
model-theoretic semantics. The machines use inference processes or 
rules (or various other kinds of wizardry). The connection between 
these rules and the semantics is that the rules should preserve 
meaning (according to the semantics), eg an inference rule is correct 
when it preserves truth, ie in any interpretation (see semantics) if 
the expressions input to the rule are true in that interpretation 
(see semantics for specs), then the expression the rule outputs is 
also true. The person who writes the rule (implements the software) 
can check that once, and then can use the rule (run the inference 
machine) in any situation knowing it isnt going to cause any problems 
screwing up the meanings.

>---
>Danny Ayers
>http://www.isacat.net
>
><- -----Original Message-----
><- From: pat hayes [mailto:phayes@ai.uwf.edu]
><- Sent: 09 April 2001 23:17
><- To: Danny Ayers
><- Cc: www-rdf-logic@w3.org
><- Subject: RE: Reification
><-
><-
><- >
><- >BTW, to my machine (and me)  "for y in AD, if <x,y> in IR(?P)
><- and <y,z> in
><- >IR(?P) then <x,z> in IR(?P)" doesn't mean anything more than e.g.  "not"
><- >does - where is this meaning exactly?
><-
><- The point is not to give a readable exposition of the meaning, but a
><- mathematically checkable standard. The trouble with just saying "not"
><- is that even logical words can be interpreted in all kinds of ways.
><- Some people take "not p" to mean that p isnt proven, others that p is
><- false, others that they are not asserting p one way or the other,
><- others yet to mean something like "I can give a constructive
><- refutation of any attempt to prove p", others even more yet to mean
><- something like "I can play a game of refute-versus-prove with p and
><- always win it".  So just saying "not" leaves the issue open; whereas
><- the model theory settles the issue very exactly. Like the Standard
><- Metre in Paris, it's not intended for daily use, but it does settle
><- any debates about exactly what is what.
><-
><- Pat Hayes
><-
><- ---------------------------------------------------------------------
><- IHMC					(850)434 8903   home
><- 40 South Alcaniz St.			(850)202 4416   office
><- Pensacola,  FL 32501			(850)202 4440   fax
><- phayes@ai.uwf.edu
><- http://www.coginst.uwf.edu/~phayes
><-

---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Monday, 9 April 2001 17:11:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:38 GMT