Hi Elliot,

>
>Unfortunately, you must requery() the datasource.

That's what I've found. My question is WHEN? Do you ReQuery after each Save? Or do you send them back to the query screen after changes were made? The ReQuery after Save wouldn't be too bad for a Single record, but in a Parent/Child situation this could return hundreds of records again. Not good if no repeat edit is required!

>In developing client/server apps I have always tried to keep a minimalist approach by giving the user a query screen that pulls up a few key fields worth of data.
>
>The user selects the actual record they want to modify, delete, etc and a separate single record query is run based on the primary key of the record the user selected. This returns a single record cursor that is current at that time.
>
>I have found that using a parameterized view works well for the editable record, and use spt (sequel pass through) for the read only navigation cursor.

This is my approach as well.