Re: Extending my question. Was: The relational model and relational

Bob Badour wrote:

>"Bernard Peek" <bap_at_shrdlu.com> wrote in message
>news:zPLtuiImBjU+Ew23_at_shrdlu.co.uk...
>
>
>>In message <0ai4a.10$m52.1404041_at_mantis.golden.net>, Bob Badour
>><bbadour_at_golden.net> writes
>>
>>
>>
>
>
>>>
>>>
>> From a purely pragmatic viewpoint I invariably choose databases that at
>>least claim to be relational. My book collection database has an
>>acquisition number field.
>>
>>
>
>Then your database prohibits duplicates obviating your own arguments in
>favour of them.
>
>
Bravo! This is exactly the point. And this is the same Bernard who said "Acquisition numbers are simply identity fields. They should not be considered part of the logical data structure."

Let's see: they should not be considered part of the logical data structure, but they are absolutely essential to your database. I prefer my physical models to have a basis in logic, so I would put the acquisition number into the logical model. You can choose not to, but if you do, don't blame the logical model for not being able to represent what the physical model can.

Bernard is arguing that physical models should not correspond to the logical models they purport to implement. Ok. By the way, I have this Oscar Meyer wiener whistle that's a physical model of the Cleveland Symphony Orchestra, and you know what? They're a really lousy orchestra. Wanna hear?

SK

>
>
>
>>--
>>Bernard Peek
>>bap_at_shrdlu.com
>>www.diversebooks.com: SF & Computing book reviews and more.....
>>
>>In search of cognoscenti
>>
>>
>>
>
>
>
>
Received on Tue Feb 18 2003 - 23:20:25 CET