Thanks for the pointer, I've talked to them (SyncData) and it sounds pretty good.

Chris

>Take a look at Whats this? DBF Replication!!! Thread #700351.
>
>>Does anyone know of a toolkit/SDK/API that allows you to synchronize seperate copies of VFP tables?
>>
>>We have a mature VFP 6 application that uses local VFP tables. A customer wants to install copies of the program at multiple sites. In this case, the application would need to constantly synchronize the data files over a network connection as the application is used. If the network connection goes down, the application continues to run using it's local data files as they stand, and once the connection becomes available it starts synchronizing again. So, in normal use, each site usually has all the latest data from all the sites.
>>
>>I assume that there would probably be a master set of data files in a central location that all the sites synchronize with. They constantly post their own changes to the master set, and get any other changes from the master set on an ongoing basis.
>>
>>I think something like this could be built using insert, update, and delete triggers that logged changes to the tables, and that log would be used to make the changes to other copies of the tables, but I'd rather license/buy something if I can.
>>
>>The client needs this functionality becuase the remote sites all need access to each others data, but they cannot be dependent on the network. So terminal services (running the app on a server with dumb terminals) is no good, as is a WLAN. The term they use for this approach is "trickle" The changes/additions to the data trickle out from the sites to each other. If the network connections go down, the trickle stops, but they can still use the application, it just suspends the trickle process until the network connection is restored. With terminal services or a WLAN, if the network goes down they are out of business until it comes back on-line. They are aware of SLA's and backup network connections, it's just how they want it.
>>
>>A friend implemented such an application in the past using a 3rd party toolkit and the Pervasive SQL engine. They had to add a few fields to each table (timestamp, ID, etc.) and hook the toolkit in to the application. Other than a synchronization process that always runs it's transparent to the rest of the application.
>>
>>Has anyone heard or actually used anything like this?
>>
>>Thanks,
>>Chris