Re: Nearest Common Ancestor Report (XDb1's $1000 Challenge)

neo55592_at_hotmail.com (Neo) wrote in message news:<4b45d3ad.0406090842.4cb3ee78_at_posting.google.com>...
> > The word "brown" is not a fact. The statements "my name is Brown",
> > "my dog is brown" and "John's car is brown" are 3 distinct facts.
>
> I do not know what your definition of a fact is, so according to your
> definition, you may be correct. I do know that all of them are things
> and in the context of storing things (ie data, schema, facts, values,
> etc) in a db, "my dog is brown" is a distinct thing from "brown" which
> is yet distinct from the following things: "b", "r", "o", "w" and "n".
>
> Per C. J. Date's, "An Intro to Db Systems", 6th Ed, Chapter 10 -
> Further Normalization, pg 312, "the purpose of such reduction is to
> avoid redundancy, and hence to avoid certain update anomalies". If the
> example data is entered in RM Sol#1 or 2, "brown" is stored three
> times (once in T_thing, and twice in T_attribute_value). Changing any
> one of them and not the others, creates an update anomaly.

You have read it, but you haven't understood it. According to your warped normalisation, if we have stored a person with surname "Brown" and he tells us that actually it is spelt "Browne" then you will avoid an update anomoly by changing the word in one place, and now you will have "my car is Browne" too!

> > Do you seriously believe that Date or Codd would support your warped
> > definition?
>
> Yes, because it isn't warp.

Dream on! Received on Thu Jun 10 2004 - 11:40:09 CEST