David Brin pointed out this group to me awhile back, and their p2p architecture is nicely summed up on the system overview,
It's not scalable to think of objects that must be distributed across a planetary-scale network as though they were Platonic Ideals, existing in some master form within a central communications server. In this rejected scenario, individual users would be sitting at their computers manipulating (dumb) proxy objects. Like Platonic Shadows, these proxy objects would get their statefulness sent to them by a central server. And since the conversations between this central server and the machines on the network concerning the state of all objects are going on continuously, it can really tax the bandwidth of the WAN.(Failure of the objective)
A user's change change-message will travel round-trip across the network to its farthest reaches so that all the replicas in the system independently commit to the change order. Only then will a new screen display be generated for everyone on the network, one that incorporates the change (viewing processes are only allowed on stable versions of the object).(Creation of the objective through the interdependent subjective - i.e., an objective compromise and the personal communal)
Consequently, the TeaTime approach is to provide an architecture that is synchronous to the degree that I/O is synchronized, but at the same time allowing for adaptation of computational strategies. The key idea for I/O coordination is that input and output events (to interactive devices) are synchronized with global universal time, which is coordinated among all sites involved in a computation.Time is perhaps the one absolute we share objectively, despite our varying subjective abilities to pay attention to that absolute. Attention is the limited natural resource of the Internet. While OpenCroquet seems to have achieved that absolute objective nirvana through ultimate subjectivity (P2P), I'm not seeing any rewards for the subjective point of view in the system.
More later.
