Re: Warning about null and open question to Oracle

From: Thomas W. Myers <t-myers_at_nwu.edu>
Date: 1995/06/21
Message-ID: <3sakec$i45_at_news.acns.nwu.edu>#1/1


In article <1995Jun20.192230.5512_at_sequent.com>, krader_at_sequent.com says...
>
>John Jones <john_at_iluvatar.tip.duke.edu> writes:
>>I
>>just think that NOTHING should be equal to NOTHING. I have heard that
>>other databases do this and was really just sounding off hoping Oracle
>>would do the same. Oh well, can't have everything I guess. Thanks for
>>listening.
>
>The problem, however, is that NULL does not mean "nothing", NULL
>means "unknown". And if two values are unknown you can't know if
>they are equal to each other or not. The problem with NULL is that
>it requires three-valued logic (T, F and ?) and people tend to
>think in terms of two-valued logic (T and F).
>

Oracle's logic is sound, for tri-state logic. It just depends on what kind of logic-religion to which you hold fast. Many people avoid this problem in their own databases by requiring most fields, and using codes to indicate missing and unknown data.

Does make you think, though when you want 'Not = X' or 'x.a = y.a (+) and y.b = z'.

-- 
Tom Myers						Email: t-myers_at_nwu.edu
Data Administrator, Attorneys' Liability Assurance Society, Inc.
Computer Studies Lecturer, University College, Northwestern University
Received on Wed Jun 21 1995 - 00:00:00 CEST

Original text of this message