> the synchronization scheme pushes all of the record keeping and conflict resolution work onto the client so that the service can perform synchronization in a scalable “stateless” manner. This means that the client needs to keep track of the state of the server during each sync, and then use this information to send and receive updates on the next sync.
Does this means that, if an error occurs on the client side and this erroneous state is propagated to the server and other clients, the data would be lost for ever?
> the synchronization scheme pushes all of the record keeping and conflict resolution work onto the client so that the service can perform synchronization in a scalable “stateless” manner. This means that the client needs to keep track of the state of the server during each sync, and then use this information to send and receive updates on the next sync.
Does this means that, if an error occurs on the client side and this erroneous state is propagated to the server and other clients, the data would be lost for ever?