[ExI] Wrestling with Embodiment

[Eugen Leitl]

On Sat, Jan 28, 2012 at 06:56:28AM -0800, Ben Zaiboc wrote:
> Eugen Leitl <eugen at leitl.org> declared:
> 
> > On Fri, Jan 27, 2012 at 11:49:04PM +0100, Stefano Vaj
> > wrote:
> > 
> > > One Egan's novel where intelligence seems actually
> > disembodied is
> > > Permutation City.
> > 
> > Unfortunately, most of Egan's scenarios are incompatible
> > with current knowledge of physics, with some having even
> > more serious problems (e.g. dust).
> > 
> > Hence most of Egan fills me with disappointment, since
> > I cannot engage sufficient suspension of disbelief.
> 
> 
> To be fair, though, this is fiction.  The function of 

Pedagogical fiction, only slightly less so than Vinge's.

> fiction is to inspire, not to thrash out the nitty-gritty of real-world problems.  

I have no problems with suspension of disbelief in fantasy,
but if I see something obviously impossible in science fiction
it's just jarring (call it uncanny fiction valley, if you wish). 
Like obviously fake physics in Hollywood, it disrupts one's 
enjoyment of the ride. 

There's very little novelty in Egan to any seasoned
transhumanist, but it's certainly good to see this pushed
out to the wider muggle audience. Albeit this can backfire,
it's hard to explain feasibility of machine-phase to those
already exposed to a Hollywood version of it via science
fiction movies. Once misclassified, it's an uphill battle.

> I'm sure there are practical problems with his vision of a 
> galactic network of routers, where people travel as gamma 
> ray laser beams over thousands of years.  

No, that part is straightforward. Wouldn't do it with
gamma, though, longer-wavelength are more practical.
Purely photonical routing is actually closer than most
people think

ftp://chacha.hpcl.titech.ac.jp/IEEE-ST.ppt

We have InterPlaNet (DTN), NASA is experimenting with
line of sight laser signalling to augment its deep
space network so we'll get a system of optical line of
sight relays in this solar system soon enough.

> But isn't it *inspiring*?

Not really. I found the impossibility of dust a harder
nut to crack. We know Hash Life works (though I'm not sure
what the simulation-internal observer would see, that would
take some simulated signal tracing), why doesn't dust work?