Re: Interesting question and debate on ddl tables with date fieldsthatwill not always have a value

There are technical reasons why nulls are better.
1 - the CPYTOIMPF issue I had yesterday where, without null I got tab
space tab and with null I got tab tab. This was critical with the partner
(whom I still think is an a$$, but my opinion doesn't matter).
2 - summary functions are built to handle nulls For example
JOE 5.00
ROB 7.00
SAM null
The average of that is 6, not 4. If you just HAVE to include it in your
calculation for average you can coalesce, or ifnull, it to some other
value.
Nulls get your attention that special processing may be needed (at least
in SQL). An incompetent may blow by the terminated date is 0001-01-01.

Rob Berendt