>Maybe the question I am about to ask you is not so hard, but for me, because I could not find any examples or documentations about it, it is.
>The basic question is how should I build the server and the client. The connection should be in the client ptoject or in the server one? How can I make the connection imposible to use without a password? I know the SQL sintax that is used for retrieving data, the problem would be in implementing and using a connection.
>How would be the implementation for a LAN different from one on modem ?

I'm not sure what you're asking here. If you want to create a client-server application, the most common approach is to use a "back-end" database server like SQL Server, Oracle, PostgreSQL or MySQL running on a server computer. The latter two can be obtained free of charge and run on Linux. You then create a VFP application and configure its data handling to interact with the back-end database via either SQL Pass-through (SPT), remote views, or perhaps ADO.

Another approach is to use a Web-based solution like West Wind Web Connect - check out the technical papers at www.west-wind.com.

It is possible to create a VFP COM server that can be accessed remotely (via Distributed COM, or DCOM). If this is what you want to do, make sure you weigh the costs of doing so against the alternatives listed above.

As far as LAN vs. dialup, the important thing is to get some sort of client-server architecture in place to avoid moving large VFP tables or indices over the wire. This is acceptable in a LAN but not in a WAN at dialup speeds. If you have a client-server system in place it will work the same way, whether LAN or dialup. The only difference is in the underlying "pipes" - the LAN will be faster.