Re: Relation Schemata vs. Relation Variables
From: paul c <toledobythesea_at_oohay.ac>
Date: Thu, 24 Aug 2006 16:43:13 GMT
Message-ID: <BGkHg.452012$iF6.92696_at_pd7tw2no>
>> <brian_at_selzer-software.com> wrote in message
>> news:1156344427.551602.122270_at_h48g2000cwc.googlegroups.com...
>>
>>
>>> A candidate key value can only identify a tuple within a single
>>> database state.
>> Here's where you're going wrong, Brian. A candidate key (from the logical
>> model) doesn't identify
>> a tuple. Not even within a single database state.
>>
> ...
Date: Thu, 24 Aug 2006 16:43:13 GMT
Message-ID: <BGkHg.452012$iF6.92696_at_pd7tw2no>
brian_at_selzer-software.com wrote:
> David Cressey wrote:
>> <brian_at_selzer-software.com> wrote in message
>> news:1156344427.551602.122270_at_h48g2000cwc.googlegroups.com...
>>
>>
>>> A candidate key value can only identify a tuple within a single
>>> database state.
>> Here's where you're going wrong, Brian. A candidate key (from the logical
>> model) doesn't identify
>> a tuple. Not even within a single database state.
>>
> ...
PMFJI, I would say that the VALUE of a candidate key identifies one and only one tuple FOREVER! Nothing to do with `states`. The value can`t ever change, rather a different value would identify a different tuple.
Surely since we can distinguish two tuples based only on their values, then we can identify them.
p Received on Thu Aug 24 2006 - 18:43:13 CEST