RE: Another one bites the dust

[rob@xxxxxxxxx]

This is a multipart message in MIME format.
--
[ Picked text/plain from multipart/alternative ]
One of the problems I have with the AS/SET data model concept is that it
still issues many of the data integrity as modifications to their code.
Yet access to data outside of their code (ie C/S) can still violate these
coded data integrity checks versus true RI.

I understand your concern for 'soft deletes'.  Often I've had to address
this with triggers instead of RI.  Do not allow any new records with soft
deleted master file records, however let's not dispose of history yet.

Another reason I've used triggers is that BPCS insists on multiple members
for all their files.  And the RI constraint commands insist on single
member files.

Rob Berendt
--
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
Benjamin Franklin