[ExI] Fermi Paradox and Transcension

[Mike Dougherty]

On Thu, Sep 13, 2012 at 11:56 AM, spike <spike66 at att.net> wrote:
> On 12 September 2012 23:02, Keith Henson <hkeithhenson at gmail.com> wrote:
>>…Talking to someone on the far side of a one AU orbit has a subjective delay of a few thousand years…
>
> There is no theoretical limit to the process: adding more transmitters each
> sending the signal a shorter distance decreases the total energy needed to
> send a signal, at the expense of signal delay.

What kind of delay do you expect to exist at each node?  Range from
best case scenario for priority1 cut-through to average case of FIFO
queuing.  "Average" will also require a guess at how many messages are
en-route through any regional subnetwork.  The specialization of nodes
becomes pretty important to balance high-speed & long distance
carriers from chatty localized thinking, else you have an smooth
distribution of non-specialized nodes who are doing all things in a
suboptimal way.

> The MBrain model I suggested yesterday would have something vaguely
> analogous: each node would communicate directly with only about 5000 to
> 10000 other nodes, possibly less than that.  Then for signals to be passed a
> long distance, they would need to be received and retransmitted by other
> nodes along the path.

Have you computed a minimum bit-length for uniquely addressing each
node?  ex:  IPv4 (32 bit) isn't enough for the immediate future -
would IPv6 (128bits) addressing be sufficient for our Mbrain?
Assuming exactly IPv6 as the model for addressing/routing/etc. how
efficiently does the low-power CPU in each node get that message
repeated to the next node (i assume it has internal state rules for
determining which of it's neighbors accept the traffic to hop along)

[1] http://en.wikipedia.org/wiki/IPv6 specifically:
http://en.wikipedia.org/wiki/IPv6#Larger_address_space

I liked the Powerpoint you shared for the sake of getting the
open-minded to entertain the idea.  I'd be interested in the next
level of detail explaining how each node accomplishes common use-case
scenarios.  ...even if you're just making stuff up for the sake of
interesting pictures.  :)