Hi, Michelle.

I use INSERT INTO, and ran into this issue early on. I have a stored procedure for generating my key ids. And I set it as the default value at the table level, just for "in-case", but otherwise I just call the stored procedure in my INSERT INTO call if I need the id. Otherwise, I just ignore it. Since I use optimistic row buffering (as a general rule), I don't run into any problems.

>I don't want to store a random value to the key. I want to store the sequential key that goes in automatically when I do an APPEND BLANK. I can't do INSERT INTO because I'm not inserting anything into it. If I try it, it will put a 0 in for the key field and then I'll get a duplicate key error. That's why I use APPEND BLANK.
>
...snip