Re: So what's null then if it's not nothing?

From: John Doherty <jdoherty_at_nowhere.null.not>
Date: Thu, 08 Dec 2005 02:35:14 -0600
Message-ID: <dn8r7e01hjt_at_news2.newsguy.com>


In <1132882485.862127.147590_at_g14g2000cwa.googlegroups.com>, michael_at_preece.net wrote:

> In SQL, NULL is most widely used to represent missing data. I
> believe it should be possible to store only present data.

I know I'm jumping in late, but it seems to me that this belief of yours, and more particularly, your insistence that it be met by systems you approve of, is the crux of the problem you seem to have with the entire relational database model.

The entire model is based on the idea of tables. A table is zero or more rows of one or more columns and all rows have the same number of columns. You can like that or not, but that's the way it is.

In that sense, a table is very much like a spreadsheet: you can't have a spreadsheet in which, say, R13C42 doesn't exist. It must exist, by the very nature of a spreadsheet.

Now, you can argue all you like that something that was similar to a spreadsheet but that allowed for any arbitrary cell within it not to exist would be a terrific thing to have, and for many purposes, you would certainly be right. But whatever that thing was, it wouldn't be a spreadsheet.

Your belief in that, though, does not invalidate the perfectly obvious usefulness of spreadsheets.

In the same sense, your belief that something other than the relational model, in which tables are not "rectangular" -- that is, in which missing data is "left out" and rows can have different numbers of columns -- does not invalidate the perfectly obvious usefulness of the relational model.

And yet you seem to believe that it does.

> If a model insists that there be a representation for unknown or
> absent data then that is a fundamental flaw in the model.

You have made your belief in that clear, but you have not shown any reason to think that it's so. And FWIW, the theoretical underpinnings of that model and its widespread practical use seem to suggest that your belief is just wrong.

--
Received on Thu Dec 08 2005 - 09:35:14 CET

Original text of this message