Re: Database design

"Marshall Spight" <marshall.spight_at_gmail.com> wrote:

>Mark Johnson wrote:

>> A relation is a set of unsorted, unordered unique tuples,

>There's some signficant redundancy in there:

>set => unsorted
>unsorted => unordered
>unordered => unsorted
>set => unique

Wrong. Unsorted, unordered is the same thing. I know you just pointed that out. But get over stuff like that. It was just there for emphasis. And no, the tuples, these specific entities, these instances, MUST be unique within a relation.

>"an n-ary relation is a subset of the cross product of n sets."

Referring to a single tuple in one relation, here. And in that context, the instance is said to be unsorted, and we both agree (I think) that it must be unique.

But:

>> which may only contain data - not text.

>Text is a kind of data.

But that's just a definitional sleight. Suddenly a 'scalar' become a 'sentence'. What is the character string? Just as example:

"Text is more subtle, because it can be viewed in many different ways simultaneously: e.g., as a sequence of characters (for encryption), as a sequence of words (for indexing and word analysis), as a sequence of sentences, etc., as various hierarchies of such sequences, or as arbitrary further structures that might be derived by syntactic analysis."

And Codd's #1:

All information in the database should be represented in one and only one way -- as values in a table.

> > But which leads back to the question. If a sort column/attribute is
>> included, is it definitionally destroyed by saying it doesn't sort, or
>> that whatever is sorted is not a, relation?

>An ordered set is something else: it is pair.

In this context, it's not something else. It's what is at issue. It is something else because the term can additionally refer to:

>The pair is (relation,ordering relation.)

But it can also refer to the relation - unless, one simply defines that as impossible, even for the suspicion and presence of order. True? And if such is the approach, doesn't that seem more just mere semantics, by the common use of that term? Received on Wed Feb 22 2006 - 10:34:44 CET