>Although distance is meaningless using TCP/IP, network traffic costs us money. What we really do not want is all the records of a result set coming down the network, every time the view needs to get updated. We just want the updated/ inserted stuff to to travel, and I need confirmation of that!
>

Refresh()... will get the update values for a particular record or records. So, if you have a remote view, and you open it, and lets say it gets all customers in the state of CA...

Now, your user navigates to a customer and presses the EDIT button. You can Refresh() that record and the backend will only be queried for the single record. You can also refresh() for a certain number of records.

Requery()... will pass the view SQL to the back end which will return the whole cursor (recordset) to you again.

However, I don't know what you want your ap to do. You should not be concerned about keeping a whole view in sync with the back end. However, If that is what you are looking for, you may want to use a TimeStamp on your BackEnd... and every X minutes do a query that gets all records with TimeStamp > tLastGet... Then, with the returned values update a read-write cursor that you are keeping on your client. You are better of replicating the Oracle data locally if you want to do something like this. That is where the power of the database comes in... and replication is something you should let Oracle deal with.


>Thanks once again,

BOb