In part II, the first sims are created. If this happens (and I'm presuming this will precede the creation of human-like minds, for reasons I'll elaborate later), I can imagine a few people will rush right up to the plate - even though it isn't total transformative uploading (which will have philosophers agonizing over it by itself).

I hope it would be pretty clear that even the basic capabilities of an emulated mind would dwarf that of natural human. These might include (for starters):

1. mental snapshot backups. shut your mind down, and spawn a process that makes a copy of your state vector for reactivation in case of catastrophe. with proper technology, the frequency of backups could constantly be increased until their happening in real time for all practical purposes.
2. mental speed control. human minds are limited in how fast - or slow - they can think. A sim can continuously take advantage of accelerating computing power to think and experience the world faster. The world would appear relatively slower. The opposite is also true, except that there would be no limit to how fast you could make the world go. You could shut yourself off and hibernate for as long as you could feasibly keep control of the machinery holding you.
3. self-programmability. so long as a computer will be able to keep track of all the vast minute details of your brain, all of that data will be available for the owner to examine and analyze. furthermore, all of those patterns will be accessible for the owner to influence, subject to one's ability to comprehend it's complexity. Some behaviour or sensational modifications would be simple, others intractable.
4. reduced cost of living. Initially, the computers might be expensive, but these things are always going down, at rates faster and more predictably than the myriad needs of natural human upkeep. give a sim enough hardware, some OS software, and electrical power, and they can live forever. Give them connectivity and access to a market full of high tech toys, and things look pretty damn exciting. Throw in some extra gear, and the security against violence is much better than the average naturla human.

I think abilities like these would seem pretty attractive to a fair number of people.

The trouble with the proposed operation, of course, is that there's no guarantee you'll have these abilities when you undergo the procedure. The chances are split 50/50, pretty much. Either you'll experience life as the sim, or as you were. The process represents a momentary split in your consciousness that is not immediately recoverable.

In Part III, I try to take into account the effects of this kind of irreversible splitting on society. What I didn't consider when I originally wrote it was how forethought and time might change how people implement the procedure. I assume that each fork would simply be a free person; that each would not want to be bound by the other. Considering how similar the forks are, however, I don't think it would be too much to expect that one would be able to make contracts with oneself prior to forking, such that the agreement is binding on both parties after the result had been achieved. One such contract might be that the sim would be responsible for the economic well-being of the natural sibling for a time - or permanently! This would solve the capitalization of labor problem that swells throughout the story, and this is hinted at in Stuff is Stuff.

In The Great Showdown, much is made of the conflict between voting naturals, who withhold suffrage from sims for fear of being outnumbered, and wealthy sims, who have no democratic power, but hold nearly all of the society's wealth (I will leave the rather sticky and depressing question of how a poor government can govern merely by virtue of heavy taxation for another time). It occurs to me that the personal question of how one will relate, cooperate, or compete with one's opposite fork is this precise conflict, rendered in miniature. The difference is that, with a personal contract, the ambiguity of non-censensual "social contracts" that have bothered libertarians for centuries can be done away with.

This leads to a nice question which lends itself well to a story. It might make a good expansion to part II. It would resolve around the question: can you trust yourself to keep your own word? What kind of contract would be too much? Such an agreement is peculiar, because you will know damn well whether your intent is to bolt on your end of the bargain depending upon which end of the deal you come out. King Solomon's dilemma ain't got nothing on this.

Some interesting contracts that might come up:

• The sim is hereby charged with the economic well being of the natural sibling. The story makes mention of the natural being the sim's parent, a position that's anomalous considering the (at first) equal competence of both parties. With an agreement like this, the sim is the one that begins to resemble the parent!

There's a different concept of uploading that I was first introduced to by John Smart of ASF, though the conecpet is not entirely original. It's a form of "weak" uploading called personality capture, whereby a simulation of you, fed by an increasingly accurate feed of all the sensory inputs of every moment of your life, coupled with an analysis of all of your behaviours in the same period theoretically develops a model of you so precise that it can mimic you in the same way a sim in cyos might. A digital you would have the goal of resembling you as closely as possible, so that when you died, it could assume your identity, property, rights, and relationships with a minimum of disruption. This might be an interesting condition to put into the contract, as well, except that it would encourage your sim patron to murder you, or neglect your health. Beside that, longevity treatments might make it unlikely for you to die at all!

• The sim's obligation toward economic support of the natural ends upon, or shortly after, the permanent uploading of the natural to sim status. Once uploading technologies can transform brains into sim minds without stranding or killing the original patient, it has all of the same potential the original sim had (although it would be a little late to the party), and would not require as much economic support.

What contracts would you make of yourself before you forked a sim?

— Nato Welch 2006/01/10 01:49

Incorporating would be an obvious step for a sim to take. Even if law had not caught up with the implications of machine intelligence, corporate personhood is well-established, and has been for more than a century. American precedents support the entire Bill of Rights for corporations - artificial persons. The only thing they lack is the vote.

I don't think that granting voting rights to a class that can perfectly repreoduce faster than rabbits is a good idea. Humans generally don't reproduce for political power, because of the great effort involved, and because children don't reliably retain their parents' political views. Sims would have neither of these limitations, as they are only limited by exponentially expanding computing power, and they can make perfect mental copies of themselves. For this reason, I doubt democracies will choose to extend suffrage to machine intelligences of any kind.

Corporations, of course, in their present form, require human owners - stockholders, proprietors, etc. It will be up to a natural human to perform this role. The upload might be thought of as an employee, or it might be thought of as property owned by it, or even as being the corporation itself, but the law may not recognize an autonomous corporate person.

This dependency can be exploited by natural humans. A person who forks a sim and incorporates it can then contract with it, becoming its sole shareholder. Financial support could be offered in the form of dividends. The working class, by way of forking and incorporating sims committed to supporting their originating shareholder families, might finally be converted into the capital class, just in time to be made obsolete in the marketplace. The key point of leverage will be the intimate familiarity and progenitor status with regard to the sims they spawn, creating a consensual entitlement to the fruits of the labor they produce.

Like Ray Kurzweil, I think it will be possible to simulate brain functioning to a sufficient degree for personal operation by the "brute force" approach of throwing more computing power at the problem and working at it from the bottom up, rather than attempting the task of trying to understand the entirety of complex human mental operation from the top down. Just because it would be possible to have a computer work and act like the natural person it emulates, doesn't, of course, mean that it's mental processes will be any less mysterious. We will, however, have vastly more - and safer - access to that person's mental operations. an emulation can be observed at any level without disturbing or jeopardizing it.

In Bill Joy's Wired magazine piece, Why the Future Doesn't Need Us, he makes a case for a policy now widely known as "relinquishment". The idea is that the technically advanced nations of the world should give up pursuit of highly advanced and dangerous technologies in order to avoid their accompanying nightmare scenarios.

The trouble with this analysis is that universal relinquishment is impossible to enforce. As a result, only socially irresponsible nations would develop the technology. That's not only bad, that's worse. When it comes to hi tech, also it's generally the case that the knowledge required for offense is the same knowledge required for defense. You can't have one without the other.

Having a branch of CYOS dedicated to the relinquishment issue, where the choices result in 1) an unstable arms race or 2) domination by technically supoerior foreign powers. would highlight this nicely.