Re: Does Codd's view of a relational database differ from that of Date & Darwin? [M.Gittens]
Date: 8 Jun 2005 05:50:12 -0700
Message-ID: <1118235012.575337.202080_at_f14g2000cwb.googlegroups.com>
"A simple workflow routine, channelling the appearances of any critical nulls (not taken care of by the constraints!) to the people in the organisation that are directlt responsible for the entry of that element of data, also fixes the problem. "
How do you deal with Unsaleried people, such as those that are retired.
If you enter a 0, then any averages might the thrown out.
If you enter their pension as a salary, they might receive a demand for outstanding tax.
Since the NULL has failed to properly model the reality, the problem is pushed out into application logic. The application is littered with code that has to filter out pensioners and housewives every time it wants to aggregate salary.
So far NULLs are probably the simplest solution that is workable, but does this mean that no better solution exists? Shouldn't theorists like Date and Darwen be searching for such possibilities? Received on Wed Jun 08 2005 - 14:50:12 CEST