On Thursday, April 28, 2011 08:31:09 PM Scott Ribe wrote: > Well, natural keys are quite obviously the way to go, when they exist. The > problem is, they usually don't really exist. What's usually proposed as a > natural key, will upon further investigation, either not be guaranteed > unique, or not guaranteed to be unchanging, or both. There is no fundamental problem with changing primary keys. Sure, there are challenges and situations where thats annoying and problematic, but it's not as bad as often assumed. Andres -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
