Re: c.d.theory glossary - NULL
Date: Sat, 24 Apr 2004 14:26:30 -0300
Message-ID: <pan.2004.04.24.17.26.30.638187_at_dutra.fastmail.fm>
Em Sat, 24 Apr 2004 11:09:44 -0400, Laconic2 escreveu:
> Second, the NULL marker (bit, if you like) is far from the insanity bit.
It is.
> In most cases where it's used, it's more like the humility bit. The
> database doesn't know everything, and admitting that fact, where
> relevant, is better than making a random assertion of something for
> which there is no evidence.
You can do that much better with special values.
> Some cells should have default values, but some should not.
That's orthogonal.
> Next, NULL is not a value. I don't care what Date or Kimball say on
> the subject. They don't know nothing.
By saying that you prove they (or Date, at least) do and you don't. Date agree NULL is not a value.
-- Leandro Guimarães Faria Corsetti Dutra +55 (11) 5685 2219 Av Sgto Geraldo Santana, 1100 6/71 +55 (11) 5686 9607 04.674-000 São Paulo, SP BRASIL http://br.geocities.com./lgcdutra/Received on Sat Apr 24 2004 - 19:26:30 CEST