Re: In an RDBMS, what does "Data" mean?
Date: 8 Jun 2004 02:51:14 -0700
Message-ID: <c0e3f26e.0406080151.2f2f3221_at_posting.google.com>
"Anthony W. Youngman" <wol_at_thewolery.demon.co.uk> wrote in message news:<zHiLa9IExOxAFw1d_at_thewolery.demon.co.uk>...
> This is for all those people who think "if I don't understand it, then
> it must be wrong" (is Tony listening :-)
Do you mean me? If so, I'm not sure what you are referring to. Do you mean that I think "If I (Tony) don't understand it, then it must be wrong"? If so, where did you get that idea?
> Now. It's not words of one syllable, I'm afraid, but I'm trying to
> explain something very heavy as simply as I can.
Thanks, Professor ;-)
> LOOK at the subject of this thread again. It is an AXIOM of relational
> theory that data comes in tuples. Show me that that's true! And because
> it's an axiom, mathematics itself tells you that logic CAN not give you
> an answer!
Where did you get that axiom from that "data comes in tuples"? Codd's rule #1 says that all data in the database is to be REPRESENTED in only one way: as values in attributes of tuples. It is a prescribed RULE for building relational databases, it is not a claim that anything in the real world "comes in tuples". We have a similar rule in English that all objects are represented by words made up from the 26 letters of the alphabet; it is not an "axiom" that says that objects "come in" combinations of the letters A-Z.
Your problem is that you consistently confuse data and reality.
Of course, this all doesn't mean that tuples are the BEST way to
represent data, or even that ALL data can be represented by tuples.
But you could easily disprove a theorem that said that "all data can
be represented by tuples" by finding a counter-example. Bet you can't
though!