>The reason for needing this, is that I'm not using a view at all. I want my database to be as lose from the form as possible. I'm running my forms on SQLEXEC's now, If I want to move to SQL Server the only thing I need to do is change the connection field in a local database. If I want to move to any other database type (ORACLE,...). I just change the connection again. I only need the ODBC driver.
>I do a SQLEXEC("SELECT * FROM ?tablename",lcCursorname) in the beginning of my form. And I dynamically create an UPDATE/INSERT command at the end of my form. My form needs only three properties for all this (tablename, cursorname, primarykey). So instead of creating a view I only need to set 3 parameters and I'm finished. Why take this approach ? It is mainly out of frustration with working with remote views.
>
>1) It's a hassle to do error checking.
>2) Empty dates get changed into 1800 something, bad support for .Null. Values (It will show .null. in your fields)
>3) String too long errors.
>I'm all prepared for these now. And the actual design is faster too.
>
>Guess there is no other solution then to set the controlsource of my look-ups in the init phase of my form.
>
>Thanks for all the help, any comments or questions are appreciated.

Jeroen,

I understand your points. I'm using a framework that allows me to shift easily from local to remote views with a small convention: lv_ views are local and rv_ are remote. But since I don't have any serious experience with remote views the only thing that I can say is "I don't know".