Re: NULLs: theoretical problems?

From: V.J. Kumar <vjkmail_at_gmail.com>
Date: Mon, 27 Aug 2007 11:24:14 -0700
Message-ID: <1188239054.049413.37180_at_k79g2000hse.googlegroups.com>


Marshall wrote:
> On Aug 27, 7:20 am, "V.J. Kumar" <vjkm..._at_gmail.com> wrote:
> >
> > Whenever we have def(x), any formula
> > will evaluate to 'false'.
>
> I'm not sure if you're just typing too fast and leaving stuff
> out or what.

Yeah, guilty as charged. Since yesterday, i've been and still am posting from my phone and let me tell you, it's not easy !

Obviously, i've lost "and x is undefined".

My real question has been what we gain by using the construct in comparison to the sql three value logic.

>
> In the proposed construct, it cannot be said what def(x):f(x)
> evaluates to without knowing at least whether x is defined
> or not, and if it is, further knowing what f is.
>
> You have to know those things before you can say what it
> evaluates to. The construct is not a new way to write "false"
> or anything like that.
>
>
> Marshall
Received on Mon Aug 27 2007 - 20:24:14 CEST

Original text of this message