Re: more on delete from join

On Sun, 30 Aug 2009 15:31:03 +1000, Marshall <marshall.spight_at_gmail.com> wrote:

> On Aug 29, 7:44 am, Bob Badour <bbad..._at_pei.sympatico.ca> wrote:
>> paul c wrote:
>> > Bob Badour wrote:
>>
>> >> But computers do solve systems of equations:
>>
>> >> x+y = 3
>> >> x-y = 1
>>
>> > That seems a very useful interpretation to me, maybe much better than
>> my
>> > argument, interpretation is a big and tricky area, coalescing an
>> > accurate interpretation with a useful one is a talent Codd had.
>>
>> I cannot take credit for it. From what I understand, the whole point of
>> the work Vadim and Marshall have done is to find a way to solve systems
>> of relational equations and what Marshall once told me leads me to
>> suspect the idea originates with Vadim.
>
> I thrashed around various ideas for view updates, none very good,
> before
> hearing Vadim assert that the right way to think of it was solving a
> system
> of equations. I didn't particularly cotton to the idea right away, but
> over
> time I've become quite convinced of the rightness of it.
>
>

The field of non-monotonic logic seems particularly close to this approach. Those interested might enjoy:

http://plato.stanford.edu/entries/logic-belief-revision/

and

http://en.wikipedia.org/wiki/Default_logic

(That latter one is Ray Reiter's work, who I believe was active in database research before he went into "pure" logic).

Cheers,
Joe Received on Mon Aug 31 2009 - 00:58:43 CEST