Re: A pk is *both* a physical and a logical object.

"David Cressey" <cressey73_at_verizon.net> wrote in message news:Cmhni.6545$Gx5.2883_at_trndny02...
>
> "Brian Selzer" <brian_at_selzer-software.com> wrote in message
> news:oXdni.23174$Rw1.4623_at_newssvr25.news.prodigy.net...
>>
>> "David Cressey" <cressey73_at_verizon.net> wrote in message
>> news:hbcni.9923$yx4.9381_at_trndny08...
>> >
>> > "Brian Selzer" <brian_at_selzer-software.com> wrote in message
>> > news:_jani.39744$Um6.23567_at_newssvr12.news.prodigy.net...
>> >>
>> >> "Roy Hann" <specially_at_processed.almost.meat> wrote in message
>> >> news:7e2dncCYhYD45wHbRVnytAA_at_pipex.net...
>> >> > "Brian Selzer" <brian_at_selzer-software.com> wrote in message
>> >> > news:1Pwmi.39574$Um6.32783_at_newssvr12.news.prodigy.net...
>> >> > [snip]
>> >> >> In addition, the definition of a candidate key does not demand that
>> >> >> its
>> >> >> values rigidly designate individuals in the universe of discourse.
>> >> >
>> >> > I've been waiting for someone else to pick up on this comment, but
>> >> > since
>> >> > they haven't I'm going to bite. When you talk about "individuals"
>> >> > here,
>> >> > do you intend that to be understood as "individual propositions"?
>> >> >
>> >> No.
>> >> > My layman's understanding is that the value of a candidate key most
>> >> > definitely *is* a unique identifier (and is irreducible).
>> >> >
>> >>
>> >> Indeed. But does the same key value identify the same individual in
> all
>> >> possible relation values? Clearly this is not the case. Consider a
>> >> relation schema in which the entire heading is the key. Now suppose
> that
>> > a
>> >> user issues a update. The tuple that identified an individual in the
>> >> universe prior to the update is now different from the tuple that
>> > identifies
>> >> the same individual after the update. Because the entire heading is
> the
>> >> key, it is clear that although the key uniquely identifies an
> individual
>> > and
>> >> is irreducible, it does not rigidly designate that individual. A
>> >> rigid
>> >> designator identifies the same individual in all possible relation
>> >> values,
>> >> not just the one that happens to be actual at any given point in time.
>> >>
>> >
>> > You can call what the user issued an update, if you care to, but what
>> > really happened is that an individual, identified by the before value
> of
>> > the key, was deleted, and a new individual, identified by the after
>> > value
>> > of the key was inserted. The fact that the old value and the new value
>> > used
>> > the same storage ("row", if you like) is irrelevant. They are
> different
>> > values, and they designate different individuals.
>> >
>>
>> Not so. Consider the following statement:
>>
>> I moved the widgit with lot number 203 at location 22 to location 44.
>>
>> Now assuming that there can only be one widgit from the same lot at a
>> particular location, then the definite description before the update,
>>
>
> widgets do not have identity in this scheme.

Ah, but they do. {lot_number, location} is a key. You could also have a relation widgit_part_numbers {lot_number, location, customer, customer_part_number} that references widgits. What widgits don't have is rigid designation. That's the problem: it cannot be determined if a referent at one possible database value is the same as a referent at another.

> If you only have a table
> describing
> widgets (lot_number, location), then you don't have a table designed to
> store the fact
>
> "I moved the widget with lot number 203 at location 22 to location 44".
>
> You can only express the following:
>
> "The widget with lot number 203 at location 22 no longer exists,
> and there is now a widget with lot number 203 at location 44."
>

I disagree. the statement,

UPDATE widgits SET location=44 WHERE lot_number=203 and location=22

clearly asserts that /the/ widgit with lot number 203 that was at location 22 is now at location 44. From that it logically follows that the widgit was moved.

> The table deosn't tell you whether a widget was moved, or whether a
> widget
> was removed, and another one inserted somewhere else.
>

You're proving my point: it can't be determined whether it is a different individual or the same individual because the key value is not rigid. Since it can't be proven that the individuals are different, a categorical denial that they are the same is patently false.

> And if you modify the table(s) so that you CAN store the above fact, you
> will no longer have a case where the key is the entire header.
>
>
>
Received on Wed Jul 18 2007 - 17:18:27 CEST