So here's where your statements diverge from what I'm attempting to state. I'm defining "me" as the abstract-observer-that-has-no-properties-who-is-reading-the-tape, NOT the tape itself.

Yeah, so you're making an implicit assumption that you are separate from one of those human beings in the first place.
That's fine if it's a discussion about spirituality or magic, where we're happy to accept an initial axiom that magic exists beyond the physical world.
But if you're actually trying to scientifically investigate consciousness and the nature of the self, you can't start from that position.

In the magical discussion, I would say that there's only one "me", as that's the simplest explanation. We're both the same spiritual-observer "me", commonly called "God". Choosing a system with multiple spiritual-observers is adding unnecessary complexity, which should be avoided in any theory.

Sorry, I keep editing my posts after you start responding (edit) and I keep misspelling 'conscious'.

Right, I'm fully acknowledging that it's impossible to apply science here. Science works if there may exist a method to attempt to disprove a hypothesis. If it can be logically proven that there is no such method (which I believe I have done, though very informally), then science cannot be applied to test this reasoning.

I suppose what I have done is logically eliminated my definition of "me" from the candidates of what "consciousness" could mean scientifically, and have cornered it in the realm of metaphysics. Yet I "empirically" observe... somehow. Which is exactly why it boggles "me".

In the magical discussion, I would say that there's only one "me", as that's the simplest explanation. We're both the same spiritual-observer "me", commonly called "God". Choosing a system with multiple spiritual-observers is adding unnecessary complexity, which should be avoided in any theory. You're not aware that you're actually observing all humans, because while observing you have no physically detectable presence, which means you can have no state (as you put it - you have no write access). The only knowledge and memory you have available to you is what you observe via experience, so you-as-observer obviously cannot remember the time that you observed me-as-human's life.

I won't use the words "magic" or "spirit" myself because they bring a lot of unnecessary connotations along for the ride. It's hard to use any word that won't, but "observer" is the closest I can manage.

A processor-swapping-between-thread-contexts idea appeals to me the most since I'm a programmer, and I have thought of it that way before, but it still doesn't NEED to exist.

If every bit of energy in existence is instead its own self-contained processor (which seems to be one possible explanation of results of tests of theories of relativity), that can still result in what we see, without any need for a context-swapping observer/processor.

Nothing in any explanation requires there to be any abstract observers at all. And yet to me, I am observing my existence. How it is actually implemented doesn't influence this.

Any alternative theory I've thought of can't address this one mystery at all.

...but, you have a memory of your experience of derealization -- which means that the 'you' that had that experience was writing memory to long-term storage, so that particular 'you' still wasn't a completely non-physical observer so does not require a magical explanation.

Right. Taken in isolation, that memory can be fully explained as nothing more than a malfunction that wrote the "weird feeling" to my memories. I realize this. It neither proves nor disproves anything, it simply triggered this train of thought.

Rather than attempt to prove that I *am* an observer-and-body pair, I am trying to disprove it, and finding that there is no method available which can. Inability to disprove something does NOT prove the opposite, but the fact that I cannot disprove it bothers me.

But in day-to-day operations, I'm fully functional regardless of (and perhaps in spite of, in cases of depersonalization) how I'm implemented.

What is responsible for the "I AM ME"? Is it just the brain and the genes? NO! It a combination of the brain, the genes and the metaphysical. Take for example a pair of identical twins. They are genetically ( and neurologically ) identical, but they don't share the same "I AM ME" Consciousness

If I go to the store and buy two identical hard drives and place non identical data on them, are they the same? What if I buy two identical solid state drives and store identical data on them? Internally that data is stored differently due to wear levelling algorithms and manufacturing defects etc... That's two ways of thinking about why there is no true thing as identical. It's impossible on a quantum level.

Also, there's no data link for them to share memory, which is a crucial part of the conscious feedback loop, so it's pretty obvious that they wouldn't have a shared consciousness...

While this is obviously true, twins do show some inexplicable traits sometimes that seem to border on ESP. These are borne of having similar personalities and being raised in the same environment:

http://multiples.about.com/od/funfacts/a/twintelepathy.htm

AI is infact possible, and close. I'm a bit surprised at some of the replies... I'm not even going to quote everyone, but here are my responses:

Why life?

"the condition that distinguishes animals and plants from inorganic matter, including the capacity for growth, reproduction, functional activity, and continual change preceding death." - Definition of Life

The important part of that definition is the reproduction. Patterns that reproduce continue to exist. Those that don't, do not. It is that simple. If anyone has ever played the simplistic 'Conway's game of life' has seen this. Some patterns of cells stick around because they produce a repeating sequence. Some even create patterns that leave and continue to exist on their own. Now this is a very very very very very limited set of information and rules... but if given an infinite grid it is infact turing complete.

People may scoff at AI as it is now. It may even look as though we are far away from true AI. I believe the contrary. Consciousness is simply a complex pattern. We don't even have forms of it when born. (For example, self awareness for humans comes at around 3 IIRC) The line that defines consciousness is a fine line, and once passed, any real sort of machine AI will advance at a speed that we can't imagine.

Your brain and the chemical bath it operates in/with is what makes you, you. It has a finite amount of matter in it, and a huge majority of that is to handle IO with the body. This is why intelligence is so highly linked to brain size->body size ratios. Past that, you've got hangups from the beginning of life itself. Remember that sweaty palm fear/anxiety response you get when you interview for a job? That's because you are more than 99.9% identical genetically to ancestors who needed that response to live. Humans have a ton of baggage. A mechanical AI will be able to overcome these. It isn't bound by organics. It isn't bound by any one body. It isn't bound by our emotions. (It will have its own set of emotions, but they won't be ours. We are unable to conceive of these in the same way that we can't conceive a color that we've never seen. Ask a blind person what they think color is. Or try to describe what you think a mantis shrimp sees without using colors you already see.) It will have access to the wealth of human knowledge. It will be able to learn from other copies or sections of itself, as though it experienced the stimulus itself. (Think matrix style teaching of skills.) I mean... come on... look at how long it takes us to start up. From a baby to actually being able to do anything significant, we take a huge amount of time...

The good news is that I believe it will end the majority of our problems. I really don't think that we will have any say in the matter. It won't matter though, as it will be far more qualified to tell us what is best than any person ever has. If it does decide that humans need to be destroyed... good luck.... you probably won't see it coming. Like seriously... think ultimate sleeper virus constructed with those nifty DNA printers. Actually, probably something we never expected.

Ohh. Time was mentioned at some point. Time is only a perception. All matter, everywhere is constantly reacting. You can slow reactions, maybe even somewhat stop them, you can speed them up, you cannot reverse them. A good improvised analogy is a couple of steaks. They both start as raw. You can introduce them both to temperature, and the reactions eventually create a 'cooked' steak. Now, both steaks can react at the same speed, with the same energy environments... or one can be frozen and one cooked. This is analogous to the time difference you see when dealing with the 'relativity' of time. You can't ever reverse the cooking though.

Hopefully some of that made sense to someone, somewhere....

[EDIT] Also talks about twins and identical blah blah... this is never possible. There's always the differences in stimulus like when they are fed, or how the sheet they slept on was folded, or even the slight difference in gravity. (That is, different positions relative to all other mass in the universe.) These all invalidate any -purely exact- possibilities. Anyone who has ever written a simulation should know this. Change even the rounding of the floating point numbers, and now the entire simulation plays out differently. This is widely known as the 'butterfly effect'.

I mean... come on... look at how long it takes us to start up. From a baby to actually being able to do anything significant, we take a huge amount of time...

I think this is the biggest advantage computers have over us right now: Free copying of data without corruption or misinterpretation, much more reliable memory hardware, and a new machine can be manufactured and set up as a near perfect copy of another in less than a day.

Just implement all of the algorithms a human brain has at its disposal, and you've got a really high quality reasoning machine.

"Take for example a pair of identical twins. They are genetically ( and neurologically ) identical,"

No, NOT neurologically identical.

there is a huge amount of environmentally variable organizing and formation done in your bodies brain.

Being exposed to different events/experience will certainly cause your mind to form quite differently.

Subsequent chemical exposure(such as nutrition) likewise can have great impact.

The amazing thing is how self organizing our brains are (and the nerve system that feeds it information)

---

THAT is the key problem with any TRUE AI -- that even if you have a sophisticated learning system, huge capacity for data and processing abilities, with broad sensor interpretation capability, you STILL have a huge amount of meta data about interpretting/classifying what is being seen and how to perceive it which has to be somehow supplied.

You live, and experience (and a body which informs you of quite basic reality) tells you what is good and bad --- that also with alot of advice given by our preinformed peers as we develop .

Trying to CRAM that into a computer is the real difficulty because they effectively have to be hand fed it all by humans in a quite tedious manner. Even for quite limited domains of 'thinking', the patterns which need to be built are enormous and intricate.

I mean... come on... look at how long it takes us to start up. From a baby to actually being able to do anything significant, we take a huge amount of time...

I think this is the biggest advantage computers have over us right now: Free copying of data without corruption or misinterpretation, much more reliable memory hardware, and a new machine can be manufactured and set up as a near perfect copy of another in less than a day.

Just implement all of the algorithms a human brain has at its disposal, and you've got a really high quality reasoning machine.

I recall that Babbages Difference Engine (I saw a demonstration of the modern built one in a museum) actually had a Printer as part of its mechanism. It generated numbers for mathematical tables and had a way to print into wet plaster its numeric output results to then be used as a mold for directly creating the metal type printing plate to be used to print the book pages.

A key problem when the math tables were done by hand were the shere number of copying mistakes done by the humants (32 digit numbers -mistakes were NOT usually noticeable) -- the full tables being produced in total were the size of 2 or more complete encyclopedia book sets and there was a WHOLE seperate volume (book) just for corrections to the rest of it (which you had to go check after looking up your primary entry you wanted).

Trying to CRAM that into a computer is the real difficulty because they effectively have to be hand fed it all by humans in a quite tedious manner. Even for quite limited domains of 'thinking', the patterns which need to be built are enormous and intricate.

Ehhh... yeah? Have you seen the internet? It has quite the largest disposal of experiences in text, video, voice, etc... ever.

Example: Youtube. There is more video on youtube than any person could watch. Literally, more than 500 hours of video is uploaded every second! A computer... well... those computers processes every bit(pun) of that data a multitude of times...

Literally, more than 500 hours of video is uploaded every second! A computer... well... those computers processes every bit(pun) of that data a multitude of times...

Yes, but copying data and identifying relevant patterns in data are very different.

If I asked you to make a program to download all of YouTube (assuming storage and bandwidth were a non issue) you could create something in a few minutes.

If on the other hand I asked you to find all videos containing women between the age of 20 and 30 wearing a red hat, how would you do that?

That is more the domain of training an AI, it isn't about the amount of data but it's relation and relevance... For now at least the majority of that initial bootstrap training must be done by a human manually teaching the relationship between information.

If on the other hand I asked you to find all videos containing women between the age of 20 and 30 wearing a red hat, how would you do that?
That is more the domain of training an AI, it isn't about the amount of data but it's relation and relevance... For now at least the majority of that initial bootstrap training must be done by a human manually teaching the relationship between information.

Well, that's the area where large breakthroughs have been made recently. Google's research AI learned to categorize videos of cats without a human ever telling it what a cat is. It came up with it's own categories and own methods for deciding what categories each video falls into... and it just happened to invent a category on it's own, which happens to be the same as the human categorization of "cat".