Post

I have always hated this diagram. Two reasons: (1) It attempts to rename a concept that has already been established in academia: "distributed" computing is very often the figure (b), (2) due to power law mechanics, (a) or (b) is almost always the case and I have never seen (c) in any network.

Agree, the diagram makes no sense. In today's terminology, I'd say all three are distributed, (a) is client-server, (b) is a 3-replica cloud service (the triangle in the top half) that some clients access via caches, and (c) is a mesh network (short-range wireless links chained together).

Whether something is decentralised or not is a function of the administrative control of different parts of the system, not a function of the network topology.

28 reposts163 likes

So decentralization always refers to a measure of power distribution? I think there’s some entanglement because power is, or at least can be, exerted spatially. So naturally topology plays a role

Also is orthogonal to technological networks. Decentralized economic/cultural/political/whatever networks follow a lot of the same relational flows. People say “Distributed” when they mean “I can see the wires” and “Decentralized” when they mean “you can’t tell me what to do” but neither is real.