One of the themes of current scientific progress is getting more and more information out of tiny amounts of data. Who'd have thought that we could learn so much of distant and recent biological history from DNA, and so much about distant planets, stars, galaxies, and the cosmos from tiny differences in very small amounts of light?

Pratchett's death puts an extra edge on the question-- to what extent can people be re-created from what they've left behind them, especially if they've written novels which include a lot of their personality?

Any thoughts about theoretical limits of how much can be figured out from small amounts of data?

New to LessWrong?

New Comment
16 comments, sorted by Click to highlight new comments since: Today at 8:49 AM

One of the themes of current scientific progress is getting more and more information out of tiny amounts of data.

But the other theme is the selection effect. We call it the Stone Age not because stone tools were the prominent fixture of their lives, but because the stone tools are the things that survive. How much of a caveman's life was spent in a cave? Well, that's the part of it that we can see. We might as well call them "bone people," except every now and then we find a mummy to make it clear there was so much more to them.

And so, like ancient people, it seems like your best bet for persistence is to fall into a glacier.

We call it the Stone Age not because stone tools were the prominent fixture of their lives, but because the stone tools are the things that survive

Woah, is this common knowledge? I feel silly for not realizing that myself.

Woah, is this common knowledge? I feel silly for not realizing that myself.

Among archaeologists, I believe yes; I think when talking to the general public, though, they emphasize what we do know rather than what we don't.

[-][anonymous]9y00

If you want something really fun...

It's becoming likely that around the time of the European contact with the Americas, there were large societies with high population densities living throughout what is now the Amazon rainforest. As parts of the jungle are cleared large geoglyphs are being discovered, along with apparent raised causeways between settlements in a number of Western basins. Furthermore, there appears to be artificial soil called Terra Preta all over along the rivers, full of charcoal and pottery fragments that collectively ameliorate the problems of low nutrient retention that hamstrings agriculture in rainforest conditions. There are vague reports of high population densities in the amazon basin that were not believed for centuries by a conquistador named Francisco de Orellana that may have been at least somewhat more accurate than they have been given credit for. What appears to have happened is that after the various European diseases (smallpox, hepatitis, etc) killed off most of the population, almost all material evidence for these people other than geoglyphs, soil engineering, and pottery shards would have rotted away. In the deep amazon, there was no stone. All building and tool materials would have been biodegradable in the very wet, warm, full-of-living-things climate.

[-][anonymous]9y130

When my father died, I realized in the last years we hardly ever talked about anything serious, neither before much, he did not hand me over his life experience, advice, philosophy, knowledge, basically the best part of himself. What we have remaining is photos the memories of fun had but that is not the best part of a person.

I realized that everybody who has children owes them to write a book. Dumping everything in it, less-obvious knowledge, experience gained, what worked for me what not, life advice, philosophy, values. Basically an extract of the best part of you, a manual to life, a "what would mom/dad do in this situation" type of book and the best kind of memory left behind, a practical memory, a problem-solver type, not just photos about how we went skiing together.

To put it differently, part of the issue is that when your children realize it is a good idea to listen to you, it may be too late. I did not realize the old mans experience is relevant until my own child was born and that was a week after he gone.

People who have no children but are in any ways interesting or high-achieving should also do this. It would be really better if we had a book from Prathett full of stuff like how he sees the world, values, philosophy, ideas, writing techniques. That way the best part would be preserved.

In the longer run, "uploading" will solve it, but as of now this sounds like a great idea, even better than cryonics, for I am not sure how useful I will be to people 2000 years later when they revive me, but such a mind-extract can be useful for my daughter in 30-40 years. It is selective uploading into the Gutenberg-computer.

There is one sort of information almost everyone leaves behind them, namely the memories of people who knew that person when she was alive. By definition, there is no information theoretic bound preventing reconstructing a person with such precision that everyone who knew her won't be able to tell the difference. Similarly, there is no information theoretic bound preventing reconstructing a person that writes novels which the original version could have written, as far as the readers can tell. So, if we launch FAI tomorrow we can get a version of Pratchett sufficiently close to the original to "fool" his wife, daughter and fans. On the other hand, there might be stronger complexity theoretic bounds: reconstructing a person to fit a given set of "output" is a problem which is a priori only known to be in NP.

[-][anonymous]9y20
[-][anonymous]9y10

One of the themes of current scientific progress is getting more and more information out of tiny amounts of data.

I think that if the universe was once a single spot and everything we see comes from there, then the information about everything is everywhere. If we had knowledge of the mechanics and enough computing power we could understand what has happened in the universe since the start just by observing the current state of one atom.

This is because, if we could measure them with full precision, the current position, direction, and speed of an atom (and all other measurements if we could do them physically) are only possible with one and only one specific history of everything else in the universe.

...to what extent can people be re-created from what they've left behind them...

I think us or any other intelligent species that continues after us will control every aspect of our environments. Just like now we are starting to understand the full "language" of DNA and we are in the first steps of cloning, AI, etc., we will also control the planetary weather, solar system level direction and orbits of planets, etc.

In this context it will be very easy to replicate past living beings. The problem is that because they will be operating on a different set of materials (h2o, salts, carbon, etc.) it will not actually be that original person even if it's an exact replica.

Unless there is some sort of entanglement possible, I think that If a person is copied, the copy is not the original person unfortunately, so when we die we will not be back unless its on the same set of materials, which is possible, but very improbable.

EDIT: The above phrase:

This is because, if we could measure them with full precision, the current position, direction, and speed of an atom (and all other measurements if we could do them physically) are only possible with one and only one specific history of everything else in the universe.

Replaced the original sentence:

This is because the current position, direction, and speed of an atom (and all other measurements that can be done physically) are only possible with one and only one specific history of everything else in the universe.

To reflect, as observed in the comments below by lesswrong.com/user/asr/, that "You can measure those things to only finite precision".

This is because the current position, direction, and speed of an atom (and all other measurements that can be done physically) are only possible with one and only one specific history of everything else in the universe.

This seems almost certainly false. You can measure those things to only finite precision -- there is a limit to the number of bits you can get out of such a measurement. Suppose you measure position and velocity to one part in a billion in each of three dimensions. That's only around 200 bits -- hardly enough to distinguish all possible universal histories.

[-][anonymous]9y10

I agree, I should have written a conditional:

This is because, if we could measure them with full precision, the current position, direction, and speed of an atom (and all other measurements if we could do them physically) are only possible with one and only one specific history of everything else in the universe.

I will edit above.

Other than our ability to measure these dimensions I think that their current state is only possible with only one history of the universe since the beginning.

That is not at all true; for example, see the inverse problem (http://en.wikipedia.org/wiki/Inverse_problem). Although the atom's position is uniquely determined by the rest of the universe, the inverse is not true: Multiple different states of the universe could correspond to the same position of the atom. And as long as the atom's position does not uniquely identify the rest of the outside universe, there is no way to infer the state of the universe from the state of the atom, no matter how much precision you can measure it with. The reason is that there are many ways that the boundary conditions of a box containing an atom could be arranged, in order to force it to any position, meaning that there is a limited amount that the atom can tell you about its box.

The atom is affected by its local conditions (electromagnetic and gravitational fields, etc), but there are innumerable ways of establishing any particular desired fields locally to the atom.

This causes challenges when, for example, you want to infer the electrical brain activity in a patient based on measurements of electromagnetic fields at the surface. Unfortunately, there are multiple ways that electrical currents could have been arranged in three-dimensional space inside the brain to create the same observed measurements at the surface, so it's not always possible to "invert" the measurements directly without some other knowledge. This isn't a problem of measurement precision; a finer grid of electrodes won't solve it (although it may help rule out some possibilities).

[-][anonymous]9y20

Thx so much for the link above!

Since I am not a scientist I didn't know about the inverse problem, but I had thought about that intuitively.

I think it's true that a current state of an atom, in a 3 dimensional context, may be achieved through multiple histories or even manipulated by a living being who wants to "cheat the system".

But in a 4 dimensional context, where the coordinates on a base field (maybe Higgs field), which could also be called time, are included, that would eliminate the possibility of recreating the same position for different stories since time happens only once, in a forward motion, and thus making that state unique and irreplaceable.

I think a position of height x, width y, and depth z is replicable the way you say above.

But, is a position of height x, width y, depth z, at time w (considering a single time line since the big bang) replicable?

If yes then the inverse problem refutes my original idea in this thread, if not IMO it is still possible to reverse-understand the universe that way.

Ah, hmm.... Maybe! If you include the entire history of the atom, then I'm not actually sure. That's a tough question, and a good question =)

If you can only measure them to some finite precision, they may not have more precision than that.

If you want to jump into the deep-end of the sci-fi pool, it's convincingly arguable that with a complete enough grasp of the physical laws, and sufficient processing power, you can reconstruct a past state up to Xs into the past and with Y% accuracy. (Of course "accuracy" is a pretty arbitrary measure).

For example, using the current trajectory of a baseball, it's fairly trivial to deduce where it came from 1.5 seconds in the past with, say, 99% accuracy. Now just extend that process to the entire Earth. It would be earth-shatteringly difficult, of course. And of course there are a hundred and one open questions about the nature of consciousness, whether a person reconstructed in such a fashion would be the "same" person, and of course the problem that describing a system isn't the same as reconstructing a system in reality.

But it's not impossible given our current understanding of the universe.

EDIT Somehow I missed the above post. Which covers all this.

[This comment is no longer endorsed by its author]Reply