>Ian,
>
>Why not just use pessimistic buffering, it does what you want automatically wihtout any work.

The only problem with that suggestion is that it requires the first user to actually make a change to the record to lock it. If user one retrieves a record, gets up and goes potty before making a change, a 2nd user can retrieve the record and modify it. I started a thread regarding RLOCK several months back and also decided that buffering was of no use due to the above situation. The person who I was researching this for could not accept that the above might happen. He wrote the app without buffering and mapped class properties to the forms performing his own locking and without buffering.

Steve