**From:** Anders Sandberg (*asa@nada.kth.se*)

**Date:** Tue Feb 19 2002 - 08:01:13 MST

**Next message:**Eliezer S. Yudkowsky: "Special difficulties of AI terminology"**Previous message:**Simon McClenahan: "Re: AI:This is how we do it"**In reply to:**Louis Newstrom: "Re: Hole in a box"**Next in thread:**neptune@mars.superlink.net: "SPACE: Cycler orbits"**Reply:**neptune@mars.superlink.net: "SPACE: Cycler orbits"**Reply:**Louis Newstrom: "Re: Hole in a box"**Messages sorted by:**[ date ] [ thread ] [ subject ] [ author ] [ attachment ]

On Tue, Feb 19, 2002 at 07:32:44AM -0500, Louis Newstrom wrote:

*>
*

*> > Anders wrote:
*

*> > > Assume we can store x bits / kg of matter. One stored bit is
*

*> > > worth c^2/x J of energy if its substrate is transmuted.
*

*>
*

*> I disagree with your premise. As Godel pointed out, a single integer can
*

*> store an infinite amount of information:
*

*> Digitize the information into discrete bytes a, b, c, ...
*

*> Encode a single integer as 2^a * 3 ^b * 5^c ... using prime numbers for the
*

*> base.
*

*> Any unique infinate data will have a unique result.
*

*> Then take only TWO atoms and store the number as say the number of inches
*

*> betwen them.
*

*>
*

*> Thus, any two atoms can store an infinite amount of information. (The only
*

*> limit in the scheme I have given is the size of the universe in inches.)
*

*> Given cleverness in how to store large integers, I don't think there is a
*

*> limit based on the amount of matter in the universe.
*

You can only move slower than c. Storage and retrieval of data will

take exponentially increasing time as it gets longer.

Storing a one megabyte text would require a distance X where log X =

127*(log(2)+log(3)+...+log(1e6/(log(1e6)-1))) ~= 127 sum_1^7.803e4

log(n) ~= 1.0172e8 (here I used the prime number density approximation

pi(n)=n/(log(n)-1)). exp(1e8) is a *very* large number. For

comparision, ln [100 billion lightyears] ~= 64.417. So that megabyte

text is far too large to be encoded in the visible universe. If you

settle for a 100 billion lightyear distance, then you can't use any

character higher than 92 (backslash in ascii) if you want to fit it in.

Gödel numbers are great in mathematics, but useless in megascale

computation.

*> The real energy shortage will be in how much energy it takes to read/write
*

*> information. All reading and writing of data seems to require energy.
*

*> Thats where the energy cost will lie.
*

Sure, it is the kTln2 terms I have been using all along.

-- ----------------------------------------------------------------------- Anders Sandberg Towards Ascension! asa@nada.kth.se http://www.nada.kth.se/~asa/ GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y

**Next message:**Eliezer S. Yudkowsky: "Special difficulties of AI terminology"**Previous message:**Simon McClenahan: "Re: AI:This is how we do it"**In reply to:**Louis Newstrom: "Re: Hole in a box"**Next in thread:**neptune@mars.superlink.net: "SPACE: Cycler orbits"**Reply:**neptune@mars.superlink.net: "SPACE: Cycler orbits"**Reply:**Louis Newstrom: "Re: Hole in a box"**Messages sorted by:**[ date ] [ thread ] [ subject ] [ author ] [ attachment ]

*
This archive was generated by hypermail 2.1.5
: Fri Nov 01 2002 - 13:37:40 MST
*