Re: The Theoretical Foundations of the Relational Model

In message <3D168294.C823240E_at_managesoft.com>, Clifford Heath <cjh_nospam_at_managesoft.com> writes
>JRStern wrote:
>> And that's the big, black hole in relational theory: identity. OOP
>> posits identity as important.
>
>So does relational theory. It just requires that you define the attributes
>by which a tuple will be distinguished from all similar tuples. It further
>requires that these attributes are not synthetic, because a synthetic
>identifier doesn't unveil the true distinction.
>
>The question devolves into a spitting contest about the time that OO and
>relational devotees attempt to define their arts as an attempt to structure
>reality. It's nothing of the sort of course; it's merely an attempt to
>structure understanding. As such, alternate and variant structures are
>valid, though each will present its important values in different areas.

Good point. OO and RM are maps, neither should be mistaken for the territory.

>
>I brought up the Laws of Form because GSB begins with the concept of
>distinction (as opposed to identity), and rebuilds the whole of Boolean
>logic around a single operator, in a system requiring only two axioms,
>not the five(?) of first-order logic. I'm not a mathematician, but his
>algebra of distinction is remarkably simple and enlightening.
>
>Memories of when my children were young are illustrative here. At 2 years,
>one referred to all four-legged domestic quadrupeds as "cat", even the
>canines. He hadn't learnt to distinguish them. Later there were cats and
>dogs, but no Burmese or Dalmatians. The awareness of the distinction came
>before the words of course, and that's informative too. The structure of
>knowledge isn't based in facts, but in distinctions.

That shows my main objection to the OO model in its current form. Having defined a class called Cat it becomes necessary to redefine some members as Feline Cats and Canine Cats or possibly not cats at all.
>
>To my mind, the OO practitioner's insistence on disembodied identity is
>an attempt to record that some things are believed to be distinct, even
>though we might not yet be able to codify the distinction. In any case
>there will be a need to distinguish things, but we can delay defining the
>distinction. In other words, we can delay codifying our understanding.

Another good point. The RM assumes that there is a set of attributes which always map 1:1 with the objects they are describing. OO looks as if it should be able to deal with an error in identifying that set. In practise it seems that substituting class membership for the primary-key negates that advantage. Excising classes from the OO model looks to me to be the way to go.

-- 
Bernard Peek
bap_at_shrdlu.com

In search of cognoscenti