[ExI] How to survive forever

[Mike Perry]

A paper I wrote (in 1986) on the possibility of infinite survival of 
records via backup copies can be found at
https://dl.dropbox.com/u/75307856/INFSURV.DOC. It considers 
individual records or chunks of information but also growing 
hierarchies of records in which every individual record can survive 
forever with nonzero probability. Records are destroyed essentially 
by radioactive decay, at a fixed assumed rate, and cosmology is 
uncomplicated and does not eventually intervene.

>On 15/06/2012 17:30, Anders Sandberg wrote:
> >
> > The main problem is likely that temperature never gets below
> > T_min=10^-19 K due to horizon radiation, and that forces a finite
> > amount of possible computation. Still, that is M_cluster/(kT_min
> > ln(2)) irreversible computations (including error corrections) - plus
> > a large number of reversible computations.
> >
>
>Ah, forgot a factor of c^2. The right number is c^2 M_cluster/(kT_min
>ln(2)).
>
>Using the Virgo cluster and asusming we use the dark matter too, we get
>10^15 solar masses, or about 10^45 kg. That will allow around 10^104
>computations.
>
>A lot, although far less than Seth Lloyds estimate that the universe so
>far has performed 10^120 operations (
>http://arxiv.org/abs/quant-ph/0110141) - but his operations seem to
>include reversible ones, which are not covered by the above estimate
>which is based on the Brillouin inequality.


We will at any rate have a long lead time for a workaround to this 
apparent eventual limit on our lifespans. If all else fails there is 
the possibility that benevolent beings in a still longer-lived 
universe will recreate us in copy form, perhaps to be followed by 
another recreation in a still longer-lived universe after that one 
becomes unlivable, and so on ad infinitum, each recreation to include 
all the beings of the previous, mortal domains.

Mike Perry