[ExI] Space Gnats

[Ben Zaiboc]

BillK <pharos at gmail.com> wrote:

>On Fri, May 31, 2013 at 3:06 PM, Ben Zaiboc wrote:
>> Indeed, and one of the best things about it is, once you're solid-state, and once the
>> infrastructure is in place, you can literally travel at the speed of light.
>>
>> That will be an interesting and strange scenario:  Matter being s-l-o-w-l-y shifted around
>> the solar system, like elephants hauling felled logs, while minds can flit between them
>> like gnats.  Space gnats.
>>
>> Is this feasible?  Or have I been reading too much Greg Egan?  (would the power and
>> other requirements be too high to make beaming minds around the place, encoded
>> on lasers, a practical thing?)
>>
>>
>
>I like the idea, but solid-state doesn't mean you can travel at the
>speed of light.
>Solid means a solid block with no moving parts, except the electrons
>(photons?) inside.
>
>So you still have to find the power to drive a block up to high speeds
>and slow it down at the far end.
>
>But your idea of intelligences travelling at light speed might still
>be feasible if the computronium block was sent on ahead and waited for
>the mind to be lasered across the solar system into it. If your
>standard Type 1483 mind bricks were scattered around, then minds might
>be able to connect to empty bricks and move at light speed, leaving an
>empty brick behind for someone else.


That's almost what I mean, but I'd assume that rather than the 'mind-bricks' only being able to hold and run a single mind, they'd be capable of running millions.  My analogy was that the mind-bricks would be the elephant/logs component, and the minds could be encoded onto lasers for transmission between them.  Obviously, as you say, the travel-at-the-speed-of-light bit would only be possible once the infrastructure was in place at the destination.  Spreading the mind-bricks over the solar system (or galaxy) would be the slow(-est) part.  Greg Egan describes it better than I can, in books like 'Incandescence' (which incidentally annoyed the hell out of me for not having an ending.  It just has a beginning and a middle.).


Bart <B.Mesman at tue.nl> wrote:

>I think what Ben is referring to with 'solid-state' is semiconductor devices. Chips. Neural chips. No matter whether "you" are hard-coded in the chip-floorplan, or as a neural configuration on a "general-purpose" neural chipset,
>
>both can be described in a binary file, containing synaptic strengths on O(10^12) synapses. That's a lot of data, but it's finite. Now as a description, space 'travel' would actually amount to communicating data, via the laser. No physical mass is moved in the process. Necessary requirement is that the neural substrate (the chipset) on sender and receiver side are compatible. The neural configuration on the sender side does not 'disappear' in the process. So if you want to have only one unique copy of you active at any time, a wise protocol would inactivate the sender configuration.


Alan Grimes <ALONZOTG at verizon.net> wrote:

>The fastest travel time using that method is a bit worse than C/3
>
>
>Here's a reasonable protocol:
>
>1. Transmitter broadcasts a fuck-ton of data into the ether.
>
>2. Receiver computes a checksum on each block and broadcasts it back to
>the sender to prove that it's radio was switched on and that it actually
>received something usable.
>
>3. Depending on the reply, the transmitter either goes back to #1 or
>broadcasts a "Clear to continue" message.
>
>4. The receiver begins computations. Conceivably, you could shorten this
>by including the checksums with the initial transmission but, given the
>payload, it seems more reasonable that the sender have as much control
>over the process as possible.


Good points, both of these.
Naturally, for this kind of travel to be possible, the hardware would have to be capable of suporting /any/ mind, not just one specific one, and I envision an entire ecosystem at each node, not just a single brain-equivalent.  It would be more like beaming into a simulated body/brain, created on demand in a virtual city.  The nodes would be huge, in terms of the vertual space and machinery inside, although physically small, and capable of supporting millions of minds and all the associated support.

Re. error-checking, yes, that's a reasonable assumption, we may have a system much like the current Internet Protocol, but even so, travelling at even 1/10 the speed of light is waaaay better than anything we have now.  The catch, of course, is that you can't take anything with you except information.  Everything physical you might need has to already be there, but I don't see that being a problem, with the level of technology that would be needed.

But this is all background, taken as given.  What I was asking was, is the idea workable, from a physics pov?  Making various assumptions about the size of a 'mind-file' (from a few magabytes to many terabytes, depending on how compressible the data would be (can a mind be encoded as variations from a standard mind-template, etc.)), how much power would be needed, would this method of travel be prohibitively expensive, only doable for the tiniest of minds, or would it be no problem, and could the solar system easily be teeming with space-gnats?

Come to think of it, could the galaxy /already/ be teeming with them?  Another solution to the Fermi Paradox?


Ben Zaiboc