Re: Fitch's paradox and OWA
Date: Thu, 17 Dec 2009 01:22:09 -0800 (PST)
On Dec 17, 5:06 am, paul c <toledobythe..._at_oohay.ac> wrote:
> Nilone wrote:
> > On Dec 16, 11:09 pm, Jan Hidders <hidd..._at_gmail.com> wrote:
> >> On 16 dec, 12:07, Nilone <rea..._at_gmail.com> wrote:
> >>> Does Fitch's paradox prove an inherent contradiction in the open-world
> >>> assumption?
> >> Only if you assume that the database contains everything that is
> >> known. It rarely does. :-)
> >> Besides, do you really believe that everything that is true can be
> >> known? ;-)
> >> -- Jan Hidders
> > "All truths are known" is only problematic if you assume that truth
> > exists independent of cognition. From a phenomenalistic point of
> > view, both entities and predicates exist purely in the mind, which
> > means that all truths are known, but doesn't exclude the possibility
> > of creating new truths based on sense data. I'm currently leaning in
> > this direction (in the sense that Mach espoused), especially after
> > coming across relational quantum mechanics.
> > But I'm getting off-topic here. It seems to me now that Fitch's
> > paradox just illustrates the distinction between OWA and CWA formally.
> To be more clear/blunt, in the context of the RM, D&D have it that it is
> a value of the relation that has no attributes. This may seem obscure
> in general language, but in the db machine context it is much
> simpler/clearer than the above.
A relation with no attributes is a simple proposition, so you're saying the chosen assumption is a value in the db, right? I prefer this, although TTM states they subscribe to CWA.
Is it necessarily a db-wide choice, or could it be useful to distinguish the choice per relation? Received on Thu Dec 17 2009 - 10:22:09 CET