[ExI] proto-bitcoin

Eugen Leitl eugen at leitl.org
Tue Apr 23 06:44:18 UTC 2013


On Mon, Apr 22, 2013 at 06:06:26PM -0400, Mike Dougherty wrote:

> > Virtualization assumes shared resources, and saving and restoring
> > state. In case of provably optimal computers there is nothing shared,
> > so saving and restoring N bits of state requires twice the number
> > of hardware, and considerable delays.
> >
> > Which means there will be no virtualization, and no cloud.
> 
> Then what the heck were you talking about when you said, "Oh it'll be
> cheaper and easier to have fantasy worlds"?  You didn't _really_ mean
> people would be playing pretend, did you?

You notice I never say virtual worlds, but artificial reality.

The reason is that virtual has a specific meaning in CS --
it means time-sharing of a given hardware resource, which
implies saving and restoring state which only works when
there's very little state to keep track of.

In case of solid state minds, it's all state, so virtualization
buys you nothing in terms of time-sharing hardware but introduces
a large number of complications. Which doesn't mean you can't
checkpoint, serialize state and reinstantiate on a different
chunk of bare computronium iron.

I guess technically you could call it a cloud, as long
as it's the equivalent of today's pixieboot on bare iron.
 
> >> someone is still running physical machines, yeah?
> >
> > Nobody is running anything. No more than you are running your
> > body. You *are* your body. Nothing is going to change about that.
> 
> Ok, you ARE your body... and you might get high on drugs and not care
> about it for a while... but that's hardly sustainable.  You will have
> to take care of it; even if you've externalized the cost of that care
> to pharmaceuticals or medical technologies that you didn't
> discover/invent yourself.

My point precisely. Solipsism is never good for your health.
 
> I don't understand how one might BE physically bare-bones hardware and
> unconcerned about the processes of physical reality.

We're definitely on one page here. Just a communication problem.



More information about the extropy-chat mailing list