Re: Is relational theory irrelevant?
Date: Wed, 19 Nov 2003 18:53:46 -0500
Message-ID: <bpgvnd$i8s$1_at_hanover.torolab.ibm.com>
Bob Badour wrote:
> D nests relational expressions within variable operations and nests variable
> operations within variable operations. It does not, and should not, nest
> variable operations within value operations. Value operations with
> side-effects is a really, really, really bad idea.
I would love to agree, but it its value is really, really high.
Let's consider auditing. Whenever someone reads your health record you
want this fact to be tracked. Eseentially you want a select-trigger in
this case.
I think the bottom line is:
Should we prohibit powerful function because it makes the (virtual)
world a less clean place?
Must we denounce Einstein because he made such a mess of Newton?
I can't say I have the answer to this. The only thing I must say is that
I have gone a long way from insisting in uncompromising cleanliness to
learning to trust our customers.
Maybe I got corrupted, may be I just realized that the world isn't clean
and we have to model it.
Cheers
Serge
-- Serge Rielau DB2 SQL Compiler Development IBM Toronto LabReceived on Thu Nov 20 2003 - 00:53:46 CET