Re: NPR, Godel, Semantic Web

["Thomas B. Passin" <tpassin@home.com>]

[Simon St.Laurent]

> >It may well trip on authority, but this claim suggests that it also trips
> >on inherent limitations of logical processing.
>
> I should clarify this claim as it was made on the radio.
>
> It's not that such systems will return wrong answers - it's that there are
> right answers they cannot find.  How that would echo through a system or
> whether users would even notice the missing information isn't clear.
>

Godel statements are found by the result of meta-arithmetic reasoning - that
is, by operations not covered by the axioms.  They can be shown to be true,
because they can be constructed by correct rules, but they can't be proven
to derive from the axioms.

In Semantic Web inferencing - notice that that is the term usually used -
we're
1) not talking usually about deduction but inference, perhaps with degrees
of trust or certainty,
2) not normally talking about stepping outside the area of normally-used
logic to apply some meta-logic (third order logic??).  And why should some
web service provider create some, let us say, service description using
godel-like meta-techniques when that would prevent possible customers from
decoding the service description?  Perhaps some hacker would have some fun
that way, but it would be much easier to simply be misleading instead of
unprovable.

What would be more likely would be that someone might come up with a theory
about how densly interconnected data networks function, perhaps some chaotic
qeueing theory, that appears to be true but might be formally unprovable in
a Godel sense.

Actually, I'm not clear whether, given any instance of a Godel number
without knowing how it was arrived at, you can determine that it represents
an unprovable statement.

Cheers,

Tom P