The internet is an engine of centralisation

NOTE HXA7241 2015-09-20T10:02Z

The ‘flatness’ of the internet just makes it even more subject to be shaped by the inevitable non-flatness of our uses of it.

0: Introduction: the myth and its wrongness

In the early years of the internet a vague beguiling myth emerged of a new social order. The network would dissolve hierarchy and bring a more democratic structure, organised from the interconnections between people – a decentralised world.

But it was all wrong – the internet has made power even more concentrated than before. And no-one seems to know why, or quite what to do. How can this be? Could it have been predicted? Perhaps: here is a quick explanation.

1: The internet is structurally unbiased

The internet is implemented in a decentralised/flat/distributed/network/peer structure, but does that mean its use will be too? No. The abstraction it presents for use is structureless: it merely enables any point to connect to any other.

Consequently the internet does not induce a distributed/flat network structure in its uses, nor in fact any kind of structure. It equally supports any kind – which might include even completely, singly, centralised.

2: The gain of informational-leverage means centralisation

The basic structure of the internet does not favour or disfavour centralisation, but something else does – the uses we choose for it.

The power of information is in its ‘informational-leverage’: creating it costs something, but copying it is near unlimited and free, and each copy yields the full value of the original.

Each piece of information is a centralisation of the leveraged gain of its many copies, and so the desire for such gain pushes toward creating such centralisations.

3: The internet increases the range of desired structures

The internet adds more nodes/people/etc to connect to each other, and better connections between them; by this it increases the range, scale, and kinds of possible informational structure.

Since the internet can provide more of any structure, and there is a drive to centralisation, this means more instances, kinds, and scales of centralisation – more centralisation, and bigger centralisation.

4: Conclusion: a matter of proper control

The internet is a gargantuan and anfractuous engine, and we drive it disorganisedly toward centralisation. But this is not truly, entirely a bad thing after all. Centralisations are good when done well.

The problems of cooperation are those of concentration, filtering, abstraction, of information. Being for or against de/centralisation per-se misses the point. Cooperation is centralisation. The right centralisations, because of the gains of informational-leverage, are tremendously valuable, but which centralisations shall we have? And how and where? The important question is more like: how are they controllable by individual and rational feedback.