Re: Null and Void() - Or, Abandon All Hope Ye Who allow

From: Wayne Conrad <wconrad(at)yagni(dot)com>
To: Scott Marlowe <smarlowe(at)g2switchworks(dot)com>
Cc: Scott Ribe <scott_ribe(at)killerbytes(dot)com>, pgsql general <pgsql-general(at)postgresql(dot)org>
Subject: Re: Null and Void() - Or, Abandon All Hope Ye Who allow
Date: 2006-07-05 21:08:14
Message-ID: 20060705210814.GB14258@mail.yagni.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

All good points.

The ability to store NULL, and the fact that there is just one kind of
NULL, seem to parallel what has become common and useful in
programming languages. Most support NULL at least for pointers, and
many support it for all data types. It doesn't have to have a defined
meaning to be useful, any more than the number "2" has to have a
defined meaning to be useful.

(Don't think NULL is useful for all types? Than look for all of the
int functions in a C program that are returning -1 to indicate "not
found," "error," etc.)

In the marketplace of special values, program language designers-- at
least the ones who designed the languages I use--agree that allowing
NULL (and just one kind of NULL) for all data types is the winner.
That alone makes NULL necessary in the database: Because storing and
retrieving data that might include NULL on the program end would be
torture if the database itself did not support NULL.

There's plenty of nits to pick here, I think. But it's clear to me
that practice has proven NULL to be too useful to ditch.

Wayne Conrad

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Karen Hill 2006-07-05 21:27:19 How do I revoke CREATE TABLE and other privileges?
Previous Message Webb Sprague 2006-07-05 20:42:49 Re: Form builder?