Re: A Normalization Question

> You'll have to explain why storing "brown" three times is redundant,
> while storing ->5 three times is okay.

With respect to dbs, normalization is the process of eliminating or replacing duplicate things with a reference to the original thing being represented. Within the context of a db, duplicate references are not considered redundant because they are unrelated to the thing being represented.

> Also, you confuse adding useless layers of indirection with normalization.

Normalization requires the equivalent of a references. It is you who is confused about the following tuple containing redundant data.

Thing Person Color Street
1 brown brown brown

The above tuple says (among other things) that there is a person named brown, a color named brown and a street named brown. The person, color and street in the tuple are all different things from each other and also different from their names which are the same thing, the string 'brown'. The above tuple stores that string 'brown' three times and this is redundant. Below is approximately how XDb1 normalizes the above tuple.

Thing Person Color Street
1 ->2 ->3 ->4

Person Name
2 ->5

Color Name
3 ->5

Street Name
4 ->5

Name Sym1 Sym2 Sym3 Sym4 Sym5 ....
5 ->6, ->7, ->8, ->9, ->10

Symbol

6      b
7      r
8      o
9      w
10     n

Note, there are no tables in XDb1, only things. However certain groups of things can be viewed a list, tree, table, matrix, etc. Received on Thu Jul 08 2004 - 04:56:29 CEST