Jan Hidders wrote:
> Kenneth Downs wrote:
>>
>> A choice quote is this by Date, in describing the original definition of
>> DK/NF:
>>
>> "Fagin defines a relvar R to be in DK/NF if and only if every constraint
>> on R is a logical consequence of what he (Fagin) calls the domain
>> constraints and key constraints on R."
>>
>> Perhaps we can rephrase this into the vernacular as saying all biz rules
>> are specified completely by primary and foreign keys?
>
> Careful, careful. Fagin is talking only about candidates keys, which is
> more than primary keys and does not include foreign keys.
>
> -- Jan Hidders
Your caution is appreciated. In my enthusiasm for the similarities, I
probably did not take enough care to enumerate the differences. The idea
was to take DK/NF as a starting point for conversation but to rapidly
depart into other areas.
--
Kenneth Downs
<?php echo "sig block. cool."; ?>
Received on Fri Nov 19 2004 - 18:11:40 CST
