Never mind the absurdity of it:
But how *secure* would it be to use and develop a split Access 2007 database by having 100% of the database components on a local computer and installing a virtual machine using the Windows Vista OS that is essentially crippled with security features (no context menus, no execution of applications, no registry access, Access runtime-only installed, the list goes on. . .), with a remote connection between the client virtual machine to a file share managed by advanced Windows file permissions on the local machine negotiated by a virtual private networking utility called "Hamachi" wherein the users may be granted access to the network, blocked, or banned, and must always be approved before being added to the VPN, and having a further split back-end database that runs perhaps once daily based off of append and update queries so as to manage having multiple clients' tables with permissions set on a per-client basis (CLIENT A + CLIENT B = MASTER)?
Each virtual machine would be associated with negotiating a connection to a separate front-end Access database application on the VPN Local Machine and this front-end database would be password encrypted, compiled in the .accde format, and have internal script-based security for authenticating users periodically ("Are you still there? Please enter your pin.").
One obvious advantage of this is the ability to update the client application as needed and monitor client access to the application in real-time. Just like the internets!!
So. . .
How secure is this implementation? Any other considerations?