>I've just started a new job and will be be working alot with an old version
>of MS SQL (6.5). The system is very slow and has a number of capacity issues. Any changes I make will have to be approved by the corporate IT department, so I expect some battles. My SQL experience is limited to MS SQL 2000 so I definately feel that I'm stepping back in time having to start working with verison 6.5.
>
>I know that each connection to the server takes resources. The main software that we use disconnects the user's from the SQL back-end after 10 minutes of non-use. This is done to conserve resources. You have to click on a "Reconnect" button before you can use the application again.
>
>My questions are these:
1) What is the default "connection" behavior of a remote view created in VFP? I.E., does it keep a persistant connection to the server whenever the view is open (i.e., USEd) ?
Remote views consume a bunch of resources. If your project can live with it, use a "pass-through" VFP cursor to show data. Once your "desktop" receives the data you can disconnect.

If your app adds data using a pass-through - do a onetime insert or update to the backebd and the append (or change) the data in the pass-through.

If you want the system fast - and need to conserve resources - and not have to deal with connection issues - use pass through and script the onetime retreivals or updates to the backend to make your pass-through cursor "appear" as a "view".

SQL 6.5 was a good version of SQL

>3) Or, is the connection only active when data is being pulled into a cursor?
>4) Does a parameterized view work the same or is there any difference between a non-parameterized view?
>5) If the connection is constant using a view, is there a way to cause it to disconnect after 10 minutes of use and offer a reconnect button if the connection has died, like the application we are using?
>
>Thanks,
>Robert Wright