Re: In an RDBMS, what does "Data" mean?

From: Anthony W. Youngman <>
Date: Thu, 10 Jun 2004 00:13:22 +0100
Message-ID: <>

In message <QL7xc.49$>, Mikito Harakiri <> writes
>"Anthony W. Youngman" <> wrote in message
>> In message <a73wc.31$>, Mikito Harakiri
>> <> writes
>> >P.S. Anthony, admittedly your reasoning is not without intelligence. What
>> >are you doing in the Pick camp?
>> >
>> If you mean me (I think I'm the only Anthony, though there's several
>> Tonys), what other camp do you expect me to belong to?
>> As far as I'm concerned, relational theory is MATHEMATICS. I'm a
>> SCIENTIST. And it seems to me the mathematicians have an unhealthy
>> disrespect for reality. I don't. You can only get so far with logic, and
>> that does not include proving that your maths actually works in the real
>> world. My experience tells me it doesn't.
>> Or are you saying that you believe that "there is one true database, and
>> Codd is its prophet" :-)
>Relational world is filled with many many interesting things. Scientist
>should enjoy (re)discovering them. If you follow database literature
>closely, you'll discover that not many works are citing Codd (or Date, for
>that matter) today. In fact, outside c.d.t and few other notorious places,
>not that much religious zealotry involved.
>But may I ask you, how is that narrow Pick world (with those ugly field
>delimiters) satisfies your scientific urge of exploration?
Well, I should ask you, how is it that the narrow relational world, with its fixed-width columns, satisfies your urges?

Yep, I know, I've just turned your question around. But relational, with its ugly fixed widths, has exactly the same problems you accuse Pick of having.

And the reason Pick satisfies my "scientific urge of exploration" is extremely easy to explain. It makes modelling the real world (which after all is the aim of science) EASY. Whereas relational results in a spaghetti rats-nest of related (or unrelated) tables and views and $DEITY knows what confusing the picture somewhat chronic! ...

In Pick, the mapping between real-world objects, and the database objects that model them, is one-to-one. In relational, it's one-to-randomly-many. One to one is a hell of a lot easier to grasp (and for the relational purists, the database can do the conversion to "tables" and "views" for me :-)


Anthony W. Youngman - wol at thewolery dot demon dot co dot uk
HEX wondered how much he should tell the Wizards. He felt it would not be a
good idea to burden them with too much input. Hex always thought of his reports
as Lies-to-People.
The Science of Discworld : (c) Terry Pratchett 1999
Received on Thu Jun 10 2004 - 01:13:22 CEST

Original text of this message