Re: more closed-world chatter

On May 10, 8:31 pm, Marshall <marshall.spi..._at_gmail.com> wrote:
> On May 10, 10:10 am, Vadim Tropashko <vadimtro_inva..._at_yahoo.com>
> wrote:
>
>
>
> > On May 10, 9:55 am, Vadim Tropashko <vadimtro_inva..._at_yahoo.com>
> > wrote:
>
> > > The generalized Date
> > > > identity is:
>
> > > > (A /\ B) \/ (A /\ 00) = A \ (A \ B)
>
> > > > where /\ is join, \/ is generalized union, and \ is antijoin.
>
> > > The reason why this formula is fascinating is that setting B=11
> > > reduces it to
>
> > > (A /\ 11) \/ (A /\ 00) = A
>
> > > the "fundamental" decomposition identity (Proposition 1 fromhttp://arxiv.org/abs/cs/0603044).
>
> > Writing in small increments... And setting B=00 gives
>
> > A /\ 00 = A / A
>
> > which is more obvious (and, therefore, less interesting:-) identity.
>
> Is that supposed to be
>
> A /\ 00 = A \ A

A typo, correct. The expression "A \ A" came from the left side of the identily

(A /\ B) \/ (A /\ 00) = A \ (A \ B) Received on Fri May 11 2007 - 17:56:44 CEST