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

RE: Reification

From: Miles Sabin <MSabin@interx.com>
Date: Thu, 5 Apr 2001 10:03:24 +0100
Message-ID: <69B15B675E99D411A4110008C786DA23DED8B5@exwest_01.interx.com>
To: www-rdf-logic@w3.org
Pierre-Antoine CHAMPIN wrote,
> What if I write
>
>  (implies (is Reification Great) (likes Drew Reification) )
>
> I do not use quoting here, still the outer parenthesis seem to
> have a quite different meaning than the inner parenthesis : I 
> do *not* assert anything about Reification being great nor you 
> liking it, though I *do* assert the implication.

It might be helpful to use an example with a less contested[1]
operator,

  (or (is Reification Great) (likes Drew Reification) )

Much the same can be said here: I do not assert that Reification
is great; nor do I assert that Drew likes reification; but I do
assert the disjunction.

> My understanding is that Triples do have the same "power" as 
> the outer parenthesis, while reification only has the "power" 
> of inner parenthesis.

I think this is confusing two quite distinct things. In,

  (or (is Reification Great) (likes Drew Reification) )

we have *occurrences* of three triples,

  (or (is Reification Great) (likes Drew Reification) )
  (is Reification Great)
  (likes Drew Reification)

Note, tho', that only the first of these is asserted.

I suspect that what's going on is that some folks are trying to
get double duty out of reification,

1. As a quotation mechanism.

2. As a hack to work around the problem that every occurrence of 
   a triple is currently treated as an assertion of that triple.

The problem, as Drew explained extremely well, is that these two 
roles have completely different requirements. In the quotation 
case we want opacity, so that, eg., from,

  (believes LoisLane '(capable Superman fly) )

and,

  (identical Superman ClarkKent)

we *can't* infer,

  (believes LoisLane '(capable ClarkKent fly) )

'opacity' meaning: no substitution of identicals into quoted
contexts.

In the occurrence-but-not-assertion case we want the exact
opposite, transparency, so that, eg., from,

  (or (capable Superman fly) (is Reification Great) )

and,

  (identical Superman ClarkKent)

we *can* infer,

  (or (capable ClarkKent fly) (is Reification Great) )

I don't believe there's any sensible way of tweaking reification
so that it can support both roles adequately.

[1] Exactly what 'implication' is is a long standing and
    extremely contentious topic in philosophical logic. Whilst
    its perfectly possible (and in many contexts useful) to
    read 'P implies Q' as '(!P)vQ' it's far from clear that this
    captures the typical senses in which 'implies' is used in
    informal language. For example, there's often a presumption
    that P is relevant to, or necessitating of, the truth of Q.
    Either of relevance or necessitation would make implication
    a kind of modal operator.

Cheers,


Miles

-- 
Miles Sabin                               InterX
Internet Systems Architect                5/6 Glenthorne Mews
+44 (0)20 8817 4030                       London, W6 0LJ, England
msabin@interx.com                         http://www.interx.com/
Received on Thursday, 5 April 2001 05:04:10 GMT

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