Re: ID field as logical address

"paul c" <toledobythesea_at_oohay.ac> wrote in message news:9tUTl.29817$PH1.24288_at_edtnps82...
> Walter Mitty wrote:
>> ...
>> So, even if the relational model doesn't forbid updates on any attribute,
>> the neophyte learning on MS Access is apt to learn that autonumber fields
>> can't be updated.
>> ...
>
> The neophyte learning MS Access is learning MS Access, not relational
> theory. Probably the same can said of whoever designed MS Access. It's
> true of other fields too that the average practitioner knows very little
> theory but what is strange to me about this one is that compared to some
> fields, there is very little relational theory to learn! The question of
> how to implement 'updates', if asked with relational theory in mind,
> should be put as how to replace one relation with another. There are lots
> of ways to program that, how too pick one is a different question. In
> fact, such comments aren't about relational theory, they would be about
> 'table theory' if there were such a thing.

Not according to Codd: I suggest you read his papers and his book. Received on Sat May 30 2009 - 03:42:43 CEST