4
$\begingroup$

Yesterday I was talking with some of my friends about ideal/utopian worlds and came across a potential infinite loop problem. My ideal world was that of a transhumanist vision. Humans still look and even feel the same, but their insides are mechanical and computerized.

Possibly Non Necessary Information

This ascended humanity is able to live and colonize most worlds without too much concern as they are largely independent of their environment. They also do not need to terraform planets, as they can just alter their sensory inputs to make the planet look however they want to. They can live out their lives for as long as they desire, free from work as lesser robots and AI would perform all the manual labor necessary for them. Even if their new bodies decay, they can be brought back into a new one from a backup that is constantly being copied and stored on servers constructed wherever they go. They are still allowed to permanently kill themselves if they wish as well as work to their heart's content.

Necessary Information

Ascended humans have the option of leaving the real world entirely and can retreat into their own mind or on servers and live out life in a virtual reality. This last part, however, is where the potential problem occurs.

If for instance an ascended delved into their own mind, but decided for the sake of immersion to block out or even delete parts of their memory, they could end up going into an infinite loop. If they had immersed themselves into a world where a transhumanist utopia was also getting ready to be developed, they could end up doing the same thing over and over again infinitely, just with slight variations. Each iteration would lead to the creation of a new personality for the original person as each one would have grown up in a different environment and had different experiences.

I imagine that this would quickly begin to take a toll on the servers as even one person would consume a lot of data and computer space. So if one person or, worse, thousands or hundreds of thousands started to essentially reproduce asexually we would see the computers and servers become overloaded and run out of memory. I also doubt that the robotic servants would be able to keep up in creating the required space and computers necessary for these new personalities or people. I don't really know exactly, but if it caused too much overload it might even cause the servers to crash and result in the extinction of the ascended.

Even if the computer space was not a problem though, there's also the potential problem that instead of the personalities being separated and treated as separate individuals, they become one confused mess and drive the person insane. Thus leading to a group of unpredictable, psychotic, spacefaring androids.

Question

Regardless if you think a transhumanist utopia is possible or not, how do you avoid/prevent/fix this infinite loop in this transhumanist scenario?

Note: I would prefer it if free will was kept intact as much as possible.

$\endgroup$
  • 5
    $\begingroup$ "Thus leading to a group of unpredictable, psychotic, spacefaring androids." That actually sounds fun. $\endgroup$ – PyRulez Jun 28 '16 at 13:12
  • 1
    $\begingroup$ Nice question Devin $\endgroup$ – James Jun 28 '16 at 14:13
  • $\begingroup$ Have you watched Battlestar Galactica? It's not exactly the same, but they explore a similar theme. $\endgroup$ – Maya P. Jun 28 '16 at 14:30
  • 1
    $\begingroup$ How is this opinion based? $\endgroup$ – James Jun 28 '16 at 17:38
  • 1
    $\begingroup$ A nice reference for this might be Permutation City (by Greg Egan). Without spoiling too much - the initial part of the book discusses virtual environments which have to deal with (computational) resource shortages in the real world. $\endgroup$ – Jamie Stevenson Sep 13 '16 at 15:08
5
$\begingroup$

I think you need to consider how your technical garbage collection would work. At what point would a memory or personality go "out of scope" and no longer be needed, and therefore cleaned up by a garbage collection process, freeing that memory for other use. If the user is editing their memories and personalities, at some point a lot of data leaves scope and gets cleaned up. Does it really matter what Personality #12 did on its tenth birthday when you're onto personality #4125?

The better, and more sinister, answer is if you remove all references to personality #12, would the person even miss it? Would they be able to know that they had personality #12 at all? You could have a situation where you keep the "base" personality safe and sound, and just update some references, so that when "base" personality went into the recursive loop (and created personality #2) you edit that memory so that instead of waking up as personality #2, the person recalls waking up as personality #x, where x is the earliest recursion you are keeping for memory purposes. Since you can estimate the amount of storage a typical lifespan would require, you can allocate and keep y number of personalities, deleting and updating references to old ones, keeping them in a personality linked list of sorts.

$\endgroup$
  • $\begingroup$ Thank you for your answer. Your comment has also made me consider the possible moral dilemma, regarding that scenario. Would terminating a previous personality be like killing another human being? If they are too different from the original they could be considered people in their own right. And each one would still have the desire to live. Hmm. Would make for an interesting story. Sort of reminds me of Unwind. $\endgroup$ – Devin Jun 28 '16 at 17:44
  • $\begingroup$ better yet, what happens during a critical system failure and the person is restored to a "higher" recursive loop? The loop "below" them probably didn't get properly cleaned up, so they would have visages and ghosts of data from that layer. Maybe iteration 12 remembers everything about iteration 13's daughter, but iteration 12 never had kids. $\endgroup$ – Marshall Tigerus Jun 28 '16 at 17:59
3
$\begingroup$

The problem is that nested virtual universes will consume undue public resources. The simple technical, if uninteresting, solution is to deprioritize individuals who consume too many resources.

Give each person running the reality simulator using a shared computer their own account. The device running the simulation will have a system to detect when any one user is consuming an unacceptable amount of computational power, as determined by the material constraints of your world. The device would then simulate both the reality and the user's conscious at a reduced rate, thus making the expensive world progress more slowly than the real world, but not seem out of sync for a user still in the world.

$\endgroup$
2
$\begingroup$

From the perspective of the servers running the virtual reality, there would not be an infinite loop. One person could descend into any level of simulation, but only the "current" one would need to be simulated. The rest could be suspended or abandoned and regenerated by the system when the person exits a layer of simulation.

Likewise, there would not be infinite (active) personalities, just one very fragmented personality, depending on what the person decides to take with him into the next layer and then take out again.

$\endgroup$
2
$\begingroup$

Your question could be shorten to: How do I fit infinite to finite?

Answer is: No way, no way.

Exponential growth, which you have discovered for your self in that situation, at some practical point indistinguishable form infinity.

Is that problem new, or unique for your world only?

No, it's not. I'm pleased to inform you, we are (as we, I refer to any species currently available to find living, on Earth) exponential growth capable creatures. We live that way for million of years and most free will isn't harm in that time.

As free will isn't ability to do anything anywhere, I suggest you solve that simply as is done on any hosting services, limit space for copies for one individual - Sorry man, you run out of space for new copies, free the space or place efforts to extend our civilization capability's and get some space for free.

$\endgroup$
2
$\begingroup$

Return Conditions

If a transhuman gets bored and wonders, "would I do it all over again?" They can block off the memory that they are transhuman, and the memory that they have blocked off memories, and imagine a world where they are about to become transhuman, or join a server running that simulation. In either case it's possible the memories may fill the available storage, either after 1000 years or if everyone keeps asking "would I do it all over again?" and blocks off more memory. By including some standard return conditions like 'wake up whenever the server memory is at 90% or wake up if my memories are more than what can fit in my own head' you can ensure this doesn't happen. If you let them block out chunks of memory without some return condition, they may as well be deleting it. If you want to make sure they don't spend the rest of eternity asleep they can wake up whenever x time passes, or whatever.

Simulation Depth

Alternately you could have some notification between simulations, "By the way Ted, you're already in a simulation, you can't add another level. Would you like to wake up now or keep going?" Which would only be revealed upon trying to pass into a deeper level of simulation.

$\endgroup$
0
$\begingroup$

Considering that it is a virtual world, even if they keep on creating new personalities - it would be in their mind only. According to them (the human), they have 5 different personalities according to the type of situation, but for a AI simulating that world, there is only one person to consider. Some extra data will be required to show that there are more personalities, but it won't overload the servers or anything.

$\endgroup$
0
$\begingroup$

If I understood you correctly, you assume that after being "digitized" a person might go the same path, to the same end - being "digitizing" again in virtual reality.

However, there must be a reason behind that decision to be "digitized". For instance, you want to defy laws of physics, use backups, control space and time, etc. That's why you go to the virtual reality.

When that purpose is achieved, though, there're no point of being "digitizing" again. Your once "digitized" transhuman must find new goals, another reasons to exist.

$\endgroup$
0
$\begingroup$

Same way that we have dealt with all resource-related problems in the pre-transhuman world: make limited resources cost something.

In this digital world, we'd be dealing with data space instead of physical land and electricity instead of food, but the principle is the same. People who created new servers in the physical world (by controlling robots that acquired resources and then built the servers) could sell or rent that space to other people. We do the same thing today - server space isn't free! People could make money by producing anything that is a limited resource in the digital world - art, new ideas, more efficient programs, physical structures for producing energy, electricity to keep the servers running - and selling it to people who want it.

Creating data "clones" is basically the same thing as making children - make sure you have enough resources (server space) available to support them before producing them. The fact that they can be produced almost instantly doesn't really change anything - if you make a million clones when you only own enough data space to allocate to five and your private server crashes, that's your own fault.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.