Re: A Logical Model for Lists as Relations

David Cressey wrote:
> x wrote:

>>David Cressey wrote:
>>>Jay Dee wrote:
>>>
>>>>>If one has a numeric index that differs for each tuple, 
>>>>>one never has duplication. 
>>>>>If one has duplication, one wonders how to refer to the
>>>>>duplicates. As Codd observed long ago, once one has said
>>>>>a thing is true, what does saying it again achieve?
>>>>
>>>>Yes, I agree.  If one were, for example, compiling a list of what
>>>>folks were going to bring to the picnic, it might be nice to know
>>>>that potato salad had been recorded many more times than fried
>>>>chicken...
>>
>>>That's counting, not asserting. Asserting something twice is no more
>>>consequential than asserting it once.
>>
>>The assertion depend on when, where, who, ... made it. :-)

> As data in databases is generally understood, the database itself is making
> an assertion, once a transaction has been committed and accepted that makes
> that assertion. Of course, an application could attach to every assertion
> an indication of the author of the assertion. In which case, that same
> author, asserting something more than once is no more consequential than
> that author asserting it once.

I said this before:
Implementation strategies we are comfortable with tend to bias the way we state problems.

At 10:10 John asserts "Two minutes ago Bob insulted me."

At 11:30 John asserts "Two minutes ago Bob insulted me again."

At 15:10 John asserts "Two minutes ago Bob insulted me again."

Hardly consequential, though. Admitted.

:-) Received on Thu May 11 2006 - 19:45:43 CEST