jack: (Default)
[personal profile] jack
I reread Permutation City recently. I think it's probably an injustice to say it made more sense this time, but knowing what was going to happen let me follow a few of the events more easily, and consider the main idea better.

Can this be a spoiler? I've joked before about things like the Titanic sinking not being a spoiler for any story, as it's intendedly part of the background. However, if a story is based on a philosophical/mathematical fact/observation, that fact has to be true beforehand as well -- you might plead ignorance that no-one had told you that the titanic famously sunk, but the fact that, eg. travelling round the world, resetting your watch at each timezone, gains you an hour, is an inescapable consequence of how astronomy and calenders work.

However, I'm only teasing. Obviously most such facts can be true but not obvious, and it's a spoiler that they're important *here* whether you knew of them or not. And many are *not* obvious even if true. And some might be literally true, but only seem relevant with the appropriate emotional impact.

Thus, the rest of this post contains no spoilers for the plot as such, but rather for the underlying idea. In this case, I do think you benefit a bit from discovering what's going on as you go along, but if you don't mind, never mind.

Permutation City (Spoilers)

Wow, the preamble nearly tired me out itself.

To rephrase the basic idea of Permutation City, this universe (and any other, virtual or real universe) can be considered a set of data. ie. this atom (or probability function) here at this time, then this one here at this time, etc. As an analogy, imagine this data is the memory of a computer.

If you plot this data in order of time, you observe some (fairly) simple natural laws that describe it, gravity, electromagnetism, etc, describing the relationship of the data at one time to that at another time. In our analogy, this would be analysing the memory and observing that it seemed to form a coherent video. (For the moment, lets ignore that the computer memory analogy is quantised and maybe finite, but the real universe may be more complicated.)

However, this ordering of the data is arbitrary. It seems simple to us, but is that necessarily a guarantee it's the only "correct" way? In our computer memory analogy, if we'd examined it differently, we might have discovered not video, but instead "well, I can't make out the rest, but this pattern of every 101th byte spells out a decypherable text stream!" (And what if the pattern isn't as repetitive as every 101th bytes, or even a pattern at all?)

That might be the result of pure chance, in that a large enough video stream, some words will be formed by patterns of bytes. But which is important is a matter of perspective, of viewpoint.

It only just occurred to me that since this idea is presented in a fiction book, the question of viewpoint is interestingly directly relevant. Viewpoints are always arbitrary in books. An author invents a whole set of people, a whole universe of people, and you read about one or two of them. Egan is particularly good at considering this sort of question -- for instance, if a human mind is recreated in a computer simulation, which is real? If the computer simulation is moved about in memory, which is real? We can't say what it means to be real, but we can say what it means to be "able to be the narrative viewpoint for a novel"!

If our text stream represents one narrator's viewpoint, and the video another, is one valid? What if the narrative of the text stream is more internally consistent, does that make that the valid one?

In Egan's example, humans are routinely copied into computer simulations. Then someone proposes that these simulations being run in this universe is arbitrary, simply the potential being there for the data somewhere to describe the simulation, given the initial conditions and rules, is enough.

There are lots more refinements I've left out. I find it hard to convey the idea. The first time I read it I dismissed it as basically sophistry; now I find it hard to criticize it coherently, so valid or not, I feel I should be able to understand it (possibly understand how it's flawed or what it reveals about our ways of thought, but understand it).

In the book, a set of copies (human simulations) create such a simulation, based on simple turing machines, but containing copies of themselves, and all the data, libraries, etc they want, and run the initial stages of it, and consider that it has a life of its own, and this continuation, invisible to them, is as valid a continuation of them as the version they may have left running in computers on earth.

The second half describes the continuation of the life in the simulation. Which at first seemed gratuitous -- if the originals don't know, how can we? But the author is certainly at liberty to describe it, the concept exists, whether or not, if the experiment were performed, that world "would" exist or not.

Of course, does that or does that not raise the idea that *all* possible universes are described this way whether we deliberately initialise them or not? And is what I perceive as me more me than a parallel universe coincidently-congruent version? If it were, could I opt out of life, confident of being continued in a universe where I spontaneously appeared, defining that as me, not my death here? With parallel universes we (or I) generally say "no" -- for whatever reason, I am loyal to *this* one, since any other way lies madness. But this gedankenexperiment is perilously close to raising the exact same question without any physics at all.

I'm going to cut short there. I may try to rework these thoughts into a coherent statement at some point in the future, though I'm very curious to know simply what other people's thoughts about Permutation City were.

Date: 2008-01-14 09:41 am (UTC)
chess: (Default)
From: [personal profile] chess
While the underlying concept was interesting, I thought Permutation City was mostly about trying to answer the question 'what kind of eternal life would people make if they can make any kind of eternal life' - what do people *do* post-scarcity, when they don't even have scarcity of time and can remake themselves utterly? And gave us a variety of pictures of what various people did with their eternity (and some heartbreaking moments which are very much spoilers).

Date: 2008-01-14 09:55 am (UTC)
From: [identity profile] atreic.livejournal.com
I forget books very quickly (my memory of Permutation city is now very fuzzy). My brain is dimly remembering that one of my objections was that the universe wasn't infinite enough for the idea to work. But as my brain cannot remember the mind bending idea in enough detail, this may not be right!

Yes, every time I try to think about permutation city my brain dribbles out my ears, so I should stop trying and do some work. I have no idea what me-ness is. I am a consciousness that is glad to be experiencing the universe and doesn't want to stop. If I could copy myself so there was an identical consciousness enjoying experiencing the universe, that wouldn't mean I would want to give it up and sink into nothingness. I don't care about there being a Sally Vernon, I just care that I am here having fun.
ext_8103: (Default)
From: [identity profile] ewx.livejournal.com

If every mathematically describable universe exists simply because it is mathematically describable (not necessarily by anything you could construct inside this universe) then we ought to imagine that our universe is one such.



We should consider that the majority of describable universes might, instead of having an apparently consistent history back to a coherent (if hard to analyze) starting point, start from some random strange state but then evolve according to fixed laws.



But our universe is consistent, at least according to a hypothetical recent random starting state that happens to include the Earth. For instance, the inbound photons apparently originating from distant galaxies and quasars show us the same picture of the universe from one moment to the next.



So either our universe is implausibly unlikely or there is some meta-law governing the subset of universes that can physically exist amongst those that are mathematically possible. (Or the whole theory is bunk, of course.)

From: [identity profile] cartesiandaemon.livejournal.com
Someone else recently linked me to a a body of theory on this thought, http://cosmicvariance.com/2006/08/01/boltzmanns-anthropic-brain/
ext_8103: (Default)
From: [identity profile] ewx.livejournal.com
That's a pretty clearly written article for its field l-)
From: [identity profile] cartesiandaemon.livejournal.com
:) I don't have great experience in cosmological ontology (or that theory in particular), I was indeed pleasantly surprised it made some sort of sense (even though it *sounded* like sophistry). I was *linked* to it from a newspaper article screaming "we're all brains in space! BRAINS!"

Date: 2008-01-15 01:33 pm (UTC)
From: [identity profile] gareth-rees.livejournal.com
Greg Egan has a Dust Theory FAQ which you might find interesting.

Date: 2008-01-17 12:18 am (UTC)
From: [identity profile] cartesiandaemon.livejournal.com
Thank you, yes indeed!