Re: What does this NULL mean?

From: David Cressey <dcressey_at_verizon.net>
Date: Fri, 16 Dec 2005 12:37:09 GMT
Message-ID: <Vxyof.18614$b57.1590_at_trndny06>


"mountain man" <hobbit_at_southern_seaweed.com.op> wrote in message news:jKqof.12951$V7.2671_at_news-server.bigpond.net.au...

> The RM is powerful but restricted.
> It does not answer all questions emergent
> in the modern database systems environment.

Agreed.

In connection with your earlier comment, and with the title of this thread, here's another case of "what does this NULL mean?"

A table is created (by the DBA).

The table is loaded (by an application written by a programmer).

The DBA comes along and does an

    ALTER TABLE ... ADD COLUMN. The DBMS Dutifully adds the column, and equally dutifully fills it with NULLS. The application program inserts a new row. Lo and behold, the application's actions are logically correct. All of the columns are filled in with correct values, except for the recently added column, which gets a NULL.

Someone performing a query sees all the NULLS, and goes back to the programmer, asking
"What does this NULL mean?"

The programmer responds, "I don't know. I didn't write it."

(We can interpret "I didn't write it." as "My program didn't write it.")

How do you make sense out of this dialogue? Who wrote the NULL?
The application programmer via his program? The DBA via the ADD COLUMN command?
The SW Engineer who built the DBMS, via the actions of the DBMS?
The SQL standards committee?
Ed Codd, when he wrote the paper that started all this?

I hope I'm not hijacking your subthread, mountain man. Received on Fri Dec 16 2005 - 13:37:09 CET

Original text of this message