Another quick and dirty Clique Space(TM) description.

I wrote the following in a letter to someone, and I think it does a good job at giving the reader a basic understanding of my concept. The description is very quick, and, unlike this one, perhaps forsakes a lot of detail in an attempt to keep it within reach of the reader's attention. Here it is...

********

I'll give you a run-down about what Clique Space is supposed to be. A Clique Space is a cluster of Agent Devices. Agent Devices talk to each other by opening channels - forming logical synapses - between themselves. An Agent Device can accept connections to different external devices depending on whether the Agent Device can support the medium that the external device uses. Any device at all, so long as it can exchange state information with another device, is a candidate for connecting to an Agent Device.

Every device is a device in Clique Space. This includes Agent Devices; they're nothing special in terms of what Clique Space is supposed to model. At any instant in time, any device that is collaborating with one or more other devices is modeled in Clique Space as a Participant in a Clique. A Clique can therefore have two or more member participants. One participant is the Clique's owner.

This is what I think is going on in real time in our brains (neurons form Cliques which grow, shrink and disband and move like pseudopods throughout one's whole nervous system) and my Clique Space hypothesis, if I can prove that it at least works, may go on to show how one can get devices (or, rather, clusters of devices) to behave like people [10 April 2013 edit: who are really just clusters of devices otherwise known as cells]). I think the Agent Device is a synthetic equivalent of a neuron; and will demonstrate the functional necessity of the neuron, the synapse, the neurotransmitter, and various other structural features of biological nervous systems we know exist.

********

Whatever description I could put in writing, I don't think I could cover the concept as comprehensively as I could if I disclosed the code. Yet, so far, the code I have is incomplete, and so the code does not even comprehensively cover the concept as it exists in my mind. Maybe sometime, I'll be able to present an implementation, and therefore prove the concept works.