Time Travel Simulation Resolves “Grandfather Paradox” | Hacker News Reader

Time Travel Simulation Resolves “Grandfather Paradox”

scientificamerican.com

154 pointsmarkmassie124 comments

> In the presence of CTCs, quantum mechanics allows one to perform very powerful information-processing tasks, much more than we believe classical or even normal quantum computers could do

That seems like an understatement if I'm understanding it correctly.

Imagine you have a "time machine" and you want to solve some arbitrary computable problem. You try a prospective solution and see if it's correct. If not, you send the next prospective solution back to be tried on the next go around the time loop. Once you arrive at the correct solution, you send that same solution back so that the iteration stops there. Then the loop repeats indefinitely with the correct solution so that the probability of exiting the loop at the correct solution approaches infinity.

It would literally be the end of the world as we know it. P=NP. Forget about quantum cryptography, that would break all public key cryptography, all cryptographic hash functions. It would obsolete algorithmic complexity theory by effectively turning every finite space algorithm into an O(1) algorithm. It would probably bring Strong AI.

But it still couldn't break a one-time pad.

There's a chapter in the brilliant fan fiction "Harry Potter and the Methods of Rationality" where Harry gets access to a Time-Tuner, i.e. a time machine from the Harry Potter universe, and this is the first experiment he envisions and performs.

HPMOR is a really great book.

https://www.fanfiction.net/s/5782108/17/Harry-Potter-and-the...

fair warning: This, absolutely awesome and actually well written, fanfiction is by now longer than all the books combined.

It's a decent read, but you have overlook that the author is very fond of himself.

Harry Potter and the Natural 20 is also highly enjoyable.

>that the author is very fond of himself.

How? And I found it better than the original at the very least ;)

Yes, it is a decent read. But it's a self-insertion Mary Sue fic. The author is very keen on telling you how smart he is. Very.

I do like his choice not to give the idiot ball to anyone.

(Please look up the technical terms on tvtropes. But only, if you have an afternoon to burn.)

One-time pad is just a generated key used once. The "time machine" will break it by trying out all keys in the key space in the time loop. If time is not an issue, brute force method pretty much breaks any crypto.

One-time pads are still secure because the decryption process will produce every plaintext the same length as the ciphertext, and you'll have no way of knowing which one was correct.

You can't "break" a one-time pad because you can't possibly know when you've got the right "key".

Others have explained why that does not work. The way you'd use backward time travel to crack a OTP is to wait until the recipient of the message decrypts it, then your steal the decrypted message, then you go back to the time where you wanted to break the code and give yourself the message.

Another use for backward time travel is data compression. To compress a file, you simply replace its contents with a note saying when you did this. When you later want the file, just pop back to that time and get it.

This also greatly simplifies backup systems.

>Others have explained why that does not work. The way you'd use backward time travel to crack a OTB is to wait until the recipient of the message decrypts it, then your steal the decrypted message, then you go back to the time where you wanted to break the code and give yourself the message.

That's not actual decryption. You can do the same thing without time travel too, e.g by hitting the guy.

Also known as rubber-hose cryptoanalysis (https://en.wikipedia.org/wiki/Rubber-hose_cryptanalysis).

Obligatory xkcd: http://xkcd.com/538/

Here's a message encrypted by a one time pad: "AHWAM"

The key is very short, only five letters, so it should be pretty trivial to brute force. Can you tell me what the message is?

Was the one time pad HDDLC, or perhaps WJHLP or maybe BJENK or IEEOF?

What algorithm would you run on this machine to yield strong AI?

First you need a rigorous formal definition of strong AI. Then write a computer program that iterates over all possible computer programs until it finds one that meets the definition.

Or try simulating a human brain.

I can't think of any formal definition of strong AI that's very satisfying. You could certainly make a great chess or Go machine, and do a lot of AI tasks, but can you think of any reasonable formal definition of a strong AI?

AIXI-tl could use a CTC computer.

As well as performing arbitrary computations in one tick, we could also keep a CTC going as a "reset button": whenever we realise that an incorrect/sub-optimal decision has been made, we can go back to when we started.

This lets us 'brute force' our real-world interactions (ie. our IO), as well as "merely" gaining an infinite computing speed-up.

This could be exploited by something similar to the "Success Story Algorithm" http://www.idsia.ch/~juergen/directsearch/node13.html . The CTC-AI would some way of "scoring" the world (ie. a fitness function), some probability of modifying its decision-making policy, and some non-zero probability of resetting the world. If the reset probability is anti-correlated to the world's score, worlds which score higher over time are more likely, hence the machine's decisions are more likely to increase the score than decrease it, hence the policy is more likely to improve over time.

Why are there so many accidents in our "reset button" lab?

But seriously, the problem with this is that it assumes that the only resolution to the CTC is that the AI generator tries another program.

You could do AIXI: "The AIXI formalism says roughly to consider all possible computable models of the environment, Bayes-update them on past experiences, and use the resulting updated predictions to model the expected sensory reward of all possible strategies."

http://wiki.lesswrong.com/wiki/AIXI

So you could use the time loop to run through this algorithm.

But what's this about predictions? Why predict when you have a time machine? All you really need is a reward function and this algorithm:

- Get a message from the future with its best-ever reward and how it achieved it.

- Try something else

- When you get to the future, see whether you've done better than the best ever. If so, replace the best ever.

- With each iteration keep a count. If you've reached some maximum number of tries without improving the best score ever, do the best strategy one last time and quit.

It might be tricky figuring out the best span of time to use, and whether you can do mini-loops within bigger loops somehow.

Or, instead of receiving the next iteration you end up with a piece of paper with "DON'T MESS WITH TIME" written in your own shaky handwriting.

The space of possible stable outcomes is much larger than just the ones carrying the outcome of the computation.

Or maybe:

"I am the Eschaton; I am not your God. I am descended from you, and exist in your future. Thou shalt not violate causality within my historic light cone. Or else."

http://en.wikipedia.org/wiki/Singularity_Sky

It wouldn't be a constant energy algorithm though, would it? For each successive optimization iteration, you would have to expend energy. Does that energy get put back with every rewind of the clock? If so, energy is also infinite.

I wonder what the time-energy-complexity trade-offs are?

As an aside, I would gladly pay someone to walk me through the math and physics sometime. (Any grad students in Atlanta are more than welcome to take me up on that offer... I might even try Skype.)

Hah, funny, I had a similar question and I'm also in Atlanta. :)

It would probably bring Strong AI.

A CTC is effectively a computer that can process a loop instantly, regardless of how complicated it is. Given that, we could design an AI machine that could analyse an arbitrary amount of data in any depth without regard to how long it'll take. More than that, everything happens in one 'tick' of whatever is controlling the CTC, so there aren't any issues around asynchronous processing. It just 'goes away'. Such a machine would never have to wait for another process to finish. Strong AI would be pretty much a certainty - we'd be able to brute force any problem.

www.scottaaronson.com/papers/ctc.pdf

Shows that (with or without quantum computing) you can do all of PSPACE using a CTC (in polynomial time) and no more.

This is the key result in this conversation; this should be higher. Basically: if we insists on Deutsch's causal fixed point propety of CTCs, they do not give you P=NP, they give you P=PSPACE. That is, all problems which can be solved in polynomial space can now be solved in polynomial time, and vice versa.

So for a time-travel computer to give P=NP or P=O(1), it would have to violate Deutsch's causality criterion. I have almost zero knowledge of either field, but it sounds to me like it makes Deutsch's model more plausible.

PSPACE is bigger than and contains NP. You still don't get ANYTHING=O(1), though.

https://en.wikipedia.org/wiki/NP_(complexity)#Relationship_t...

Beyond P=NP: this actually gives you a constant-time solution for any problem where it's possible to enumerate candidate solutions, regardless of how long it takes to do so. So, we'd have ANYTHING=O(1).

Only if you are certain there is a correct answer. If there isn't a correct answer, you just cause a paradox. It's also more likely that the machine will spontaneously fail, or the creator will get hit by a bus before he can complete it, etc.

Using a time-loop in node.js to compute factorials https://gist.github.com/Marak/3967740

I wonder if you'd run into some sort of energy problem. Like, this time machine would require some sort of energy and incur some amount of wear and tear on during each pass. The amount of possible solutions to today's cryptography might make this brute force level of effort, even with a time machine, infeasible. I feel like nasty old thermodynamics may preserve our world still. :)

Granted, I'm way way out of my depth here. I have a Ph.D. in physics but I studied nonlinear oscillators, relating them to problems in neuroscience.

My pet theory:

You create a paradox, it just keeps looping (you go back in time and kill your grandfather -> you don't exist -> you don't build the time machine and go back -> your grandfather lives -> you exist -> you build the time machine etc). But every time around the loop there is a certain amount of actual randomness.

Perhaps that atom decays this time around, or that transistor erroneously conducts due to shot noise.

So all we see is the final "fixed point". The iteration where everything ends up going just right to avoid a paradox - perhaps the gun fired prematurely, or the time machine didn't work, or a passing airplane dropped an engine on you, or...

From the perspective of any time traveler, it ends up being as though the universe is conspiring against you if you try to do anything that would cause a paradox.

This sounds like the Novikov self-consistency principle.

Say you have a wormhole on a billiard table. It curves around and goes three seconds backward in time. You roll a ball into the wormhole, aimed such that after it exits the wormhole, it will knock its earlier self off the path so that it never enters the wormhole. Paradox.

Except when you try it, instead of emerging along the pathway you aimed, it emerges along a slightly different path, and strikes its earlier self only a glancing blow. And why did it emerge along a different path? Because it was struck a glancing blow.

http://en.wikipedia.org/wiki/Novikov_self-consistency_princi...

I suppose having a superposition of all (self-consistent) loops that collapses into a single result is equivalent to the result of what I said... Hmm.

Thanks for the link!

David Deutsch also proposed a similar resolution for closed timelike curves, which say that if you have a probability distribution of the present, all that's necessary is that when you travel back in time, you get the same probability distribution. In short, if you flip a (true!) random coin to before shooting your grandfather, no paradox exists:

http://journals.aps.org/prd/abstract/10.1103/PhysRevD.44.319...

This is described with more intuition and background in one of (always excellent) Scott Aaronson's lectures: http://www.scottaaronson.com/democritus/lec19.html

The problem with this is the universe is no longer causal. Events can happen which cause themselves. It requires you to try every possible universe and then "throw away" the ones that aren't consistent. And if you simulate a universe, doesn't it cause that universe to exist, even if you later decide you don't like it? Who is to say we don't exist in one of the many "throw away" universes vs the one where the particle causes itself to be consistent.

More on that here:

http://lesswrong.com/lw/fok/causal_universes/

If your theory was the correct one, could we use that effect to perform computation - perhaps condensing 1000s of years of computation into a few minutes, with the result encoded in which steady state it settles on?

Yep.

But you'd need absurd reliability. Say you have something that condenses 1000 years of computation into a minute. That's ~500 million / 1 time compression.

What are the chances that if you ran 500 million copies of the machine with no time compression for a minute none of them would malfunction? Relatively small, right? Well, the single machine with time compression would end up doing effectively the same thing - the chance that it gets the right answer is probably much smaller than the chance that it goes up in a puff of smoke because the power supply freakishly died or something.

In order to do practical computation with such a system you'd need absurdly reliable computers, in absurdly safe environments. If you have a time compression ratio of 1 billion to 1, well, how many 1-in-a-billion things are there to go wrong? Better hope you're not near a faultline...

I like to imagine that time travel is chancy, and the 'time tunnel' collapses each time a paradox is imminent.

Well, to be fair, any time machine is probably fairly complex. The chances of the time machine being the thing that breaks the loop by breaking/etc is probably non-negligible.

When I was younger I thought a lot about the physics of traveling back in time but I always seemed to hit a brick wall with spatial coordinates.

Where we are in the universe today is very far from where we were in the universe yesterday based on the movement of the earth alone. Add to that the movement of our solar system, galaxy, cluster, supercluster and movements I am not even aware exist and it becomes really far away. Grandpa would probably be light years away from me and my time machine.

Am I missing something here, because I've never heard this mentioned by anyone else?

If you can travel backwards in time (or even just stop time) you can be at any spatial location you desire in the entire universe (since you have essentially infinite speed).

Moreover, imagine I write down some information, then wait for exactly 6 months for the earth to take me far away from that point in space. If I now travel 6 months minus 1 nanosecond back in time with that piece of paper, but preserve my spacial position, haven't I just 'transmitted' information through space faster than light can?

FTL travel is synonymous with time travel, and vice versa.

Relativity is weird.

It's been almost 10 years, so I may be mistaken... but I think the director of Primer mentioned this in the DVD commentary.

EDIT: Right, not really part of the plot. I just think I remember him (in the commentary) talking about some of the "trickier" aspects of writing a story about time travel, and I think he mentioned the fact that nobody ever really addresses the positional aspect.

I think the model of time and space in Primer is more consistent because it's a very specific kind of time machine.

If I recall correctly, you'd turn on the machine, wait N hours, go inside the machine, and in the next N hours you'd be going back in time and come out the other end at the moment the machine was turned on. So you would have spent 2*N extra hours, the machine would have always been in the same spot, so there'd be no problem with traveling through space.

There are still problems with the paradoxes, but that's part of the plot of the movie.

It wouldn't seem to matter in the movie, since whatever you put in the device would get dragged around with the device as it traveled through space.

"Perhaps! But this problem only comes up if you're not the centre of the Universe."

http://www.qwantz.com/index.php?comic=1394

It's a valid issue. You wouldn't need to worry about any component of your motion that's "constant velocity" (so, inertial straight-line motion), since all inertial reference frames are equally valid. But you would need to keep track of curving motion (like the rotation of the Earth, orbit around the Sun, Sun circling around the galactic core, etc.).

You might be able to wave your hands and claim that the time machine follows paths back in time along natural world-lines in curved space-time (gravitational geodesics), which will at least get things like orbiting around the Sun right (and galactic motion as well). But I think you'd still need some way to keep the machine's path rotating on the surface of the Earth rather than falling back and forth through the center of the planet.

I thought it might follow backwards in space as well as time but when the book The Physics of Time Reversal by Robert G. Sachs came out in 1987 it made me skeptical of that possibility. I haven't followed the theory of CP violation for many years so this may not be a pertinent point anymore.

For most "time machines" you read about, you aren't really going to run into the "CPT" issues from quantum field theory.

Unless you're considering a model of time travel where you're still literally in the same room and just living in reverse (while still visible to all the forward-living people around you and interacting with them as you go), this isn't equivalent to CP symmetry reflection.

Most time machine ideas instead picture some sort of wormhole or "stitching" between different points in space-time, in such a way that the local "forward time direction" remains continuous for observers taking the trip (and then connects back up with forward-directed time at some point in the past).

(I'm teaching a low-level course on "Time Travel in Science and Literature" this term. It's fun, but a challenge without using math.)

Maybe time travel needs some kind of tracks, like a railroad. That would also explain why we don't see any time travelers - they can't visit us as there are no tracks that lead back to 2014.

You are not alone! I've thought a lot about how much we move in space while we "sit still" on earth. :)

The time travel show "Seven Days" from the 90s dealt with this -- sometimes the time machine would end up stuck underground or floating out in space.

Most of the time this is handwaved over, for bunches of reasons. Part of it is that there is no one valid frame of reference: you can't just say "this stays stationary" - because what does that even mean? Stationary relative to what?

If you mean "you go wherever inertia would take you in that amount of time, with no acceleration", then if you go backwards a multiple of a year you end up pretty close to Earth (quick approximation: the Sun's acceleration in its orbit in the milky way is ~2*10^-10m/s/s, which works out to ~100km in a year, ~400km in two years, etc.). In actuality you could probably get somewhat better range by letting your velocity build up to intercept Earth-that-was at the target time. (You'd need to take into account a bunch of things for best accuracy: the Earth's orbit is not "perfect", the moon's phase, the acceleration of the Sun, etc, but we know most of that.)

You probably would end up needing to make multiple jumps, pausing in-between to let Earth's (or another planet's, or even the Sun's) gravity well readjust your velocity to what is needed for your next jump.

This means that you'd need to build in re-entry capability into the machine (as you are unlikely to have the precision necessary to be able to jump into atmosphere at a sane velocity), and potentially maneuvering delta-v. Note that with a time machine you lose conservation of momentum, which makes things a lot easier. You can change the direction of your velocity vector by jumping back in time and interacting with your past self. You can potentially use this to get "free" (read: reactionless) velocity out of any gravity well.

whats your reference for coordinates?

Earth Center? Solar Center? Galaxy Center? ... Universe Center?

hmm traveling backwards along the 4th dimension by itself sounds like we are NOT traveling in the other 3.... so where would the atoms (that originated from your banana at 9:25am 3 days ago) go if we merely changed their value on the 4th axis? does their xyz stay locked at the current values or revert back to their previous values.

That is a major plot point in Spider Robinson's Callahan's Con. (Look, his first few Callahan stories were charmingly quirky, and I got into the habit of reading them. This is not an endorsement of his larger oeuvre, okay?) In that case, the cosmic microwave background was used as the inertial reference, probabilities were the main problem to solve (we'd be well beyond the relative simplicity of the three-body problem), and little details like delta-v were not merely hand-waved away.

Scott Aaronson has done some work regarding CTCs and quantum computers: http://www.scottaaronson.com/papers/ctc.pdf

Apparently, in the presence of CTCs, quantum computers are no more powerful than classical ones.

My pet theory:

Quantum mechanics is a statistically based science because we have yet to discover the constants that would unify it with general relativity, making this entire thought experiment irrelevant.

Imagine if you are watching 10 parallel strings ripple up and down at various frequencies. Viewed from the side, it would at first glance appear chaotic, but by observing the behavior long enough and recording where the activity occurred over time, you could statistically predict where the strings would be at any given time.

However, if you were aware of the number of strings you were viewing and the frequencies at which they were moving, you could formulate a precise theory of exactly what you were viewing and what you would view at every point in the future.

That is the current state of physics. Quantum mechanics is seeing without understanding the rules governing the system, so predictions and observations are made using statistics instead of constants. Once we discover the constants, the current theories will be entirely obsolete and articles like this will become relics.

yeah. I remember some ~7 years ago reading about a research team that stated that they had rewritten many of the fundamental equations of (some branch of) quantum physics to show that they could be written in a completely deterministic manner. They said that while such equations weren't likely to be useful (due to the limitations of our testing technology), it was an important result, particularly when it comes to understanding the underlying science. And they were very upset that basically no one paid their result any real attention.

Apparently, Stephen Hawking organized a time travel experiment in 2009. There was also the Time Traveler Convention in 2005 [1]. If people keep organizing such events, future time travellers may be at a loss as to which one to attend.

[1] http://web.mit.edu/adorai/timetraveler/

future time travellers may be at a loss as to which one to attend

Ahem, all of them at once?

That they can do that, technically, doesn't mean the would.

Just like we don't attend all parties available to us, they would soon get bored.

Do you really want to go to 10000 conventions for the same thing?

The best part of this event was that they turned people away. I feel sorry to those time travellers who didn't go back far enough.

Imagine you're a time traveler from the future, and you've read all about the NSA from 2009. Would you want to risk making your presence known?

What do you think the government would do if they got a hold of a real life time traveler?

The first one?

We all know no one attended the party, so maybe time travelers just didn't want to disturb that part of history?

My favorite "layman's" argument for why time travel, as it's been classically described in science fiction, can't exist is that we haven't been overrun with tourists from the future.

Sure, but CTCs only allow one to travel back to a time when the CTC existed, so I don't think that's a good argument unless there was a CTC-generating device in our neighborhood. Of course, as far as I know, all CTCs require infinities (Tipler cylinders, for example: http://en.wikipedia.org/wiki/Tipler_cylinder ), which is an argument against constructing one...

> CTCs only allow one to travel back to a time when the CTC existed

So... Quantum Leap rules of time travel then? Oh boy...

but the full question being is we = all ?

and my favorite refutation is that they've been here for a while now.

E.g. the fellow in this photo from 1940: http://www.dailygalaxy.com/.a/6a00d8341bf7f753ef01347ff60cea...

Nice fake picture :)

Consensus seems to be that the photo is actually real, just not so out of place. From the Wikipedia page "Time travel urban legends" [1]:

> Further research suggests that the modern appearance of the man may not have been so modern. The style of sunglasses first appeared in the 1920s. On first glance the man is taken by many to be wearing a modern printed T-shirt, but on closer inspection it seems to be a sweater with a sewn-on emblem, the kind of clothing often worn by sports teams of the period.

[1] http://en.wikipedia.org/wiki/Time_travel_urban_legends#Moder...

Not sure how one can confirm the photo is actually real, it's so easy to fabricate picture nowadays and to add noise to make it seem like it belongs to the original. It's be interesting to actually have an in-depth article on the methods to detect fake pictures, I have seen a couple on HN but they only focused on a few examples.

It strains credulity to believe that we do in fact have time travelers living among us and 100% of them have managed to completely escape detection. For what it's worth, there actually have been serious attempts at determining whether there are time travelers living among us, including at least one study[1] looking at Twitter for evidence of time travelers tweeting

[1]: http://www.cnn.com/2014/01/03/tech/innovation/time-travel-tw...

Who says they have escaped detection? There have been alleged time travelers like John Titor. Additionally, Titor and some other claimed time travelers have said the many-worlds hypothesis is correct, meaning that own history and experience does not necessarily provide them with the ability to make accurate predictions about our future.

Additionally, we don't actually have any way of detecting time travelers if they decide not to do silly things like leak future information on Twitter.

To be clear, I don't actually believe there are time travelers living amongst us, but studies of Twitter posts are very far from being conclusive, and there might be very good reasons why time travelers do not or cannot predict the future or just aren't very common anyway. For example, perhaps time traveling technology is legally controlled and extremely expensive, like nuclear weapons. The future of humanity is not infinite, so there's no guarantee that even if time travelers exist they should amount to any substantial number of people visiting any substantial number of times for trivial reasons like time travel conventions.

> There have been alleged time travelers like John Titor

John Titor is the only one I've heard of, are there others? And Titor's predictions, even the really short-term ones, were so off the mark that there's no reason to think it's anything other than a prank. Note that even the many-worlds hypothesis would have the short-term predictions coming true and only the longer-term predictions becoming more and more inaccurate (more importantly, any differences should be due to his presence, and his presence did not e.g. cause CERN to fail to discover the basis for time travel in 2001).

> Additionally, we don't actually have any way of detecting time travelers if they decide not to do silly things like leak future information on Twitter.

We don't have any way of detecting time travelers if they behave in a fashion indistinguishable from people who aren't time travelers. But it's pretty safe to assume that's not the case. They are, after all, from the future, and had a reason for coming to the past. That reason alone should cause observable differences in behavior. In addition, knowledge about future events should be detectable in some fashion, their lack of a real fleshed-out personal identity. Heck, even just their manner of speech should give them away.

> For example, perhaps time traveling technology is legally controlled and extremely expensive

We'd still see people breaking the rules, and such people would in fact be far more likely to attract attention. We live in a world where teenagers have successfully built nuclear reactors; no matter how legally controlled it is, someone is going to figure out how to do it anyway. If it's a matter of cost, some billionaire would figure out how to build it even without government approval. If you had billions of dollars to your name, and time travel was real, wouldn't you be tempted to build a time machine no matter what the government says?

---

As far as I'm concerned, the only two really valid reasons why we haven't seen time travelers are:

1. Time travel is impossible, or 2. Time travel is possible but you can only go back as far as the point where the machine is first switched on.

> We live in a world where teenagers have successfully built nuclear reactors; no matter how legally controlled it is, someone is going to figure out how to do it anyway. If it's a matter of cost, some billionaire would figure out how to build it even without government approval.

How many people build nuclear bombs? Or supercolliders? Nuclear reactors are an altogether different thing, particularly if you're not concerned about producing surplus energy. Despite the fact that the principles behind nuclear weaponry are known, even few governments have (independently) built them and they are very strictly controlled. I can easily imagine the engineering required for time machines is even more complex, the materials as hard to acquire, the energy expenditures prohibitive, etc. And if time travelers could change history, then they would surely be regarded as incredibly dangerous superweapons, not something Elon Musk-types can just build and hop into on a lark. Indeed one could even imagine a future in which time machines are feasible but there's a "MAD" principle at work keeping anyone from building one - so if any were built, they would be kept top secret and time travelers would act with as much secrecy and delicacy as possible.

To me the idea "we have not seen time travelers; ergo this class of time machine cannot exist" is a failure of imagination. Other reasons I can imagine: human civilizations never devote the necessary resources to engineering a time machine; some social collapse or stagnation prevents technology from progressing beyond some given point; humans or human civilization is wiped out before the invention of time machines by any number of social or natural disasters; it is more difficult to travel further back in time, so nobody goes this far back, though they can go before the creation of the time machine; the singularity happens and post-singularity transhumans don't care about traveling to pre-singularity times; etc.

> And Titor's predictions, even the really short-term ones, were so off the mark that there's no reason to think it's anything other than a prank. Note that even the many-worlds hypothesis would have the short-term predictions coming true and only the longer-term predictions becoming more and more inaccurate (more importantly, any differences should be due to his presence, and his presence did not e.g. cause CERN to fail to discover the basis for time travel in 2001).

I don't believe Titor was a time traveler either, but you can make reasonable arguments about why his predictions were incorrect: his presence changed the future in unanticipated ways, the MWI model of time travel doesn't actually send you to a world that branches off from your arrival time, CERN actually did discover time travel, etc. I don't find any of these plausible but I am also not willing to categorically assert we have no evidence of time travelers or that an analysis of Twitter is even weak evidence that time travel is not possible.

> How many people build nuclear bombs? Or supercolliders?

How useful are those to individual people? Nuclear bombs are basically only useful as a deterrent (or as leverage, e.g. making demands in return for halting your nuclear program), and that only works for countries.

Supercolliders are also incredibly specialized and only useful to a small group of scientists, and there's zero incentive for an individual person to attempt to build one. What use would a private person have with a supercollider?

> Despite the fact that the principles behind nuclear weaponry are known, even few governments have (independently) built them and they are very strictly controlled.

Sure, because of how the world reacts to countries attempting to run their own nuclear bomb programs. And because of the actual low utility value in having nuclear bombs, as covered earlier.

> I can easily imagine the engineering required for time machines is even more complex, the materials as hard to acquire, the energy expenditures prohibitive, etc.

Ah, but the upside is so much higher! Not only is it vastly more useful than a nuclear bomb, but it's also useful to individual people, as opposed to just useful to countries.

> so if any were built, they would be kept top secret and time travelers would act with as much secrecy and delicacy as possible.

They only need to act with secrecy in their home time.

> They only need to act with secrecy in their home time.

How would the USA react if, say, it discovered Russian time-travel agents from the future in our time?

> How useful are those to individual people? Nuclear bombs are basically only useful as a deterrent (or as leverage, e.g. making demands in return for halting your nuclear program), and that only works for countries.

The only reason it only works for countries is because only countries have the resources to build nuclear bombs. There are certainly other groups and organizations that wouldn't mind using them as deterrents, or companies that wouldn't mind selling them to the highest bidder.

Even if they were easy to build though, they'd still be dangerous superweapons. Like a time machine.

> Supercolliders are also incredibly specialized and only useful to a small group of scientists, and there's zero incentive for an individual person to attempt to build one. What use would a private person have with a supercollider?

Perhaps they're a component of a time machine. :) My point is even if supercolliders were not useful only to a few people, they are incredibly difficult to design and build regardless. Bill Gates or Elon Musk can't just tspend some money and get one. They would have to invest large sums of money and hire legions of engineers and scientists. We don't know that a time machine isn't similarly extremely difficult (indeed it seems more likely that it is extremely difficult to build.)

I can't emphasize the "time machines are superweapons" enough. A time machine might be useful to average individuals but you'd be insane to ever let them have their hands on one. It would effectively be suicide for everyone involved. Why would you ever let anyone have one?

> Why would you ever let anyone have one?

I think that's the real difference here. You're saying "why let someone have one?" and I'm saying "how can you stop them?"

if they screw up they get an infinite amount of chances to fix their screwup

I'm sorry I cannot give a better cite than this, but sometime in the last 20 years there was a terrific short story, probably in Analog (but possibly in Asimov's) about the first time traveller.

His first shock was when he found that the person in the past they selected to visit, Shakespeare, did not seem at all surprised when visited by a time traveller. In fact, he seemed to find it routine. The traveler at one point says he doesn't understand how Shakespeare can already know about time travelers, since he is the first. Shakespeare tells him that he may have been the first to leave, but he wasn't the first to arrive.

Shortly after, many time travelers arrive...but not to see Shakespeare. They are reporters, from all throughout the timeline, coming to try to interview the first time traveller, and Shakespeare, an old hand at dealing with time travelers, steps in to protect the first time traveller and prevent him from being overwhelmed.

"The Merchant of Stratford" by Frank Ramirez (amazing approach, really unique and ingenious).

Looks like it was in Isaac Asimov's Science Fiction, July 1979 [1].

[1] http://www.abebooks.co.uk/book-search/author/ramirez-frank-r...

The internet tells me it was in Asimov's, in the July 1979 issue, so I was way off on the time frame.

Reminds me of this classic: http://www.abyssapexzine.com/wikihistory/

"Everybody kills Hitler on their first trip."

Looks like this thread is gathering the list of good Time Travel stories. So here's one I like: "Dinosaur Beach" by Keith Laumer. A delightfully twisty and multi-layered novella.

Read: http://www.baenebooks.com/10.1125/baen/0743435273/0743435273...

Reviews: https://www.goodreads.com/book/show/584217.Dinosaur_Beach

I've always found this argument a bit bizarre, because in my mind any time travelling device would work via a pair of machines: a sender and a receiver. So nobody could come to the present simply because there's no receiver yet.

The reason we haven't been overrun by tourists because there's no stop in this time period.

>I've always found this argument a bit bizarre, because in my mind any time travelling device would work via a pair of machines: a sender and a receiver.

Were did you get that impression from? Not most Science Fiction or Physics, because neither postulates such a thing.

Actually the only vaguely plausible-ish (though by no means convincing) designs I've heard of for physically realisable time machines work like this. You do some horrendous things with spacetime curvature and you get something that can send you back in time, but no further back than the point at which you constructed your horrendously curved spacetime thingy.

Paul Davis has a short book How To Build A Time Machine which does it this way, if I recall correctly (it's been a while since I read it).

Isn't it the premise of the movie Primer?

Science Fiction is not relevant here.

As for physics, you are mistaken. The only kind of "time travel" that has any basis in physics is the one using Closed Timelike Curves (which is what the article is about) and those are limited to the time the CTC was created in the past.

So, in fact, this is a very good reason to reject the argument of "If it is possible, where are the time travelers?".

I hate this argument, not because it isn't correct as far as it goes, but because it's very often represented as much more broad than it really is.

That is, yes, we can be reasonably sure that time travel mechanisms whereby tourists definitely could and would become known in our time period will not exist (even this isn't the same as can't exist). But this is a very small sampling of time travel scenarios you could imagine, and I think it's far from the only thing described 'classically' by science fiction.

And my favourite "layman" counter argument to that is that plenty of time travel theories only allow travel as far back as to when the time travel machine was first created.

Well, the argument is that quantum indeterminism implies quantum time travel is possible, because the grandfather paradox is solved/migitated by indeterminism in which the grandfather paradox is still a valid outcome. However, quantum indeterminism itself is not yet completely undisputed; some theoretical physicists such as Gerard 't Hooft maintain that quantum mechanics can be both deterministic and consistent with current experiments.

This also raises the following question - if quantum mechanics is undeterministic, then the macroscopic world is at least probabilistic. If probabilistic systems allow the possibility time travel on the quantum scale, would the grandfather paradox also not be solved on classical scales? Since it is possible that me killing my grandfather in the past would fail with some very small probability (due to various quantum effects adding up), would the grandfather paradox then not also be solved in this case? I'm not sure if I'm interpreting the article correctly.

Two nice links for anyone who wants to nerd out about the physics of closed timelike curves:

http://plato.stanford.edu/entries/time-travel-phys/

http://philsci-archive.pitt.edu/1673/1/TMArchive.pdf

If time travel exists, then it cannot be a simple rewind of the universe's tape. Because if that was the case, then rewinding the tape to a time that your father was not born would mean that you will also not be born, and thus there wouldn't be any time travel, since neither you nor anyone else would be able to witness it.

Thus, in order for time travel to work, it means to rewind the tape and add yourself to the tape at a moment you did not exist before. Thus, if you killed your father, your replica wouldn't be born, but you wouldn't stop to exist.

Therefore, the whole 'grandfather paradox' is not a paradox at all: if time travel exists, then you would simply exist in a universe where a replica of you will exist or not, depending on how events play out.

Not so fast - who says you become a "replica" in the first place? Assuming time travel replicates things, then yes, what you said holds true.

But as I understand it you don't replicate, your time just loops back on itself. So looping back to a time before you were born and killing your father would create a paradox.

You seem to have not understood what I said. I did not say time travel would replicate anything.

You would simply be transferred to the past, without any connection to the outcome of that past.

I.e. Let's say that you are transferred back in the 50s and you go kill your father before you were born. Then nothing will happen, there would be no paradox, you will continue to exist in that universe.

I'm starting to adopt the idea of Scott Aaronson that since we have a lot of evidence suggesting that P != NP, we should treat physic ideas as less probable of being true if they make P=NP as a side effect.

Say I go back in time. Does it mean I disappear from the time I was in? And I'm in two places at a point in the past (the me from that time and the me from the future)? Now if I want to go back to the time I came from, do I join back at the exact time I left (which means no one would notice my travel), or would I have to travel back to the time I left + the elapsed time I spent in the past? And if I really did return to the exact time I left, does that really mean no one noticed my leave? Because I'm sure when I left time continued and somebody noticed.

In the form of time travel being talked about, the answers are:

a) Yes, you disappear from the time you were in.

b) Yes, you're in two places at once for the duration between your arrival and your departure

c) Forwards travel in time doesn't cause too many philosophical problems (since we're already doing it) which is why nobody talks about it. But there's no requirement to wind up exactly where you started.

Can someone enlighten me: what is this simulator they speak of? It doesn't seem answered anywhere in the article what it is, whether it's software, let alone how it works. Yet the whole article relies on it. Does it act on real protons? Or are they simulated as well?

And, most important, why do they trust a simulation device built by humans who hardly understand the matter completely, when the device is then used to make assumptions about that same matter? Is it somehow proven this simulator is correct? And how did that happen?

This is a poor article. It is reporting on a simple table top device created by researchers eager to misrepresent their work. It tells us essentially nothing about quantum mechanics or CTCs.

We don't know how quantum mechanics behaves in the presence of CTCs. There are multiple, mutually incompatible proposals for extending QM in this regime that all make identical predictions in normal, causal spacetimes (i.e. the only sorts of spacetimes we have access to). Deutch has one extension of QM, and it suffers from serious internal consistency problems, but there are others.

Since this experiment takes place without CTCs, it can't tell us anything we didn't already know. It's literally just a toy device that is sorta described by the same math as a certain system with CTCs in QM_Deutch would be, if you squint your eyes hard enough. But you could say the same thing about a computer simulating QM_Deutch in software. It tells you nothing about whether the universe actually obeys QM_Deutch in the presence of CTCs ... if they exist at all.

Similarly deceptive bits of "research" in the news often appear that claim to find quantum gravity effects on table-top experiments.

http://www.space.com/5052-black-hole-effect-created-lab.html

In all these cases, the scientists will eagerly over-represent their work to the journalist, and the journalist will eagerly gobble it up and write eye-catching headlines about new breakthroughs in physics, but the heart of the claim will be buried in the article and subtly couched in weasel words. And if you actually try to drive down at it by asking probing questions, the journal and scientist will retreat to much, much less exciting claims and you will never be able to prove they intended otherwise.

All hypothetical time machines must also be space machines, i.e. they must take into account the movement of the planets, the galaxy, the universe to avoid materializing their passengers in the middle of space or the centre of the sun. My question is what frame of reference would they use?

This is exactly what always bothers me in time travel movies.

You never hear about the poor time traveler who got the calculation slightly wrong and ended up deep in space or inside of a mountain.

Only when you find their fossils.

You can use a time machine as a very good space drive.

"All" you do is jump back / forwards in time repeatedly so that gravity pulls you in the direction you want to go. (In actuality: good luck plotting a flight path)

Want to get to 20% of the speed of light? Sit in freefall near Jupiter for 5 weeks (assuming 20m/s/s acceleration), jumping forward / backward in time as appropriate to stay in roughly the same spot relative to Jupiter.

http://en.wikipedia.org/wiki/Closed_timelike_curve

"In mathematical physics, a closed timelike curve (CTC) is a world line in a Lorentzian manifold, of a material particle in spacetime that is "closed", returning to its starting point."

Now, go to http://en.wikipedia.org/wiki/Virtual_particle#Vacuums and look at the picture "One-loop diagram with fermion propagator"

It seems to me that virtual particle pairs satisfy the CTC criterium.

I do not think that there can be a Grandfather Paradox. If one goes back in time, he/she will simply be removed from the state of the universe as it is now and be added to the state of the universe at the target past date. He/she could then go on killing their father before they were born, but nothing will happen to them except not meeting with themselves as babies.

I've enjoyed thinking about the following:

It must be almost impossible to time travel back in time. But it could be possible, given unlimited time. Thus, given that at some point in the unlimited future, time travel would occur, how can we increase the chances now of being visited by these future time travellers?

> But it could be possible, given unlimited time.

Just because time is infinite, does not mean that all things are possible.

My understanding is that the universe is in a bounded random walk. As space-time moves, the unexplored potential state space increases more quickly than can be explored.

If there are multiple universes, then each universe could explore one of these states. So if there are more states than universes, not all things are possible. But if there are more universes than states, all things would be possible.

Though there are infinite potential states and infinite universes, I feel like the potential states expand faster than the number of universes. This means that our likelihood of living in a universe where something like like time travel is possible, is incredibly low.

It's quite possible that the problem with paradoxes stems not from them being 'impossible', but from our limited cognitive tools failing to equip us to think about them.

Sort of like people in ancient times believing you can't predict the motion of the planets because they didn't have calculus yet.

It's also quite possible that the whole idea of time travel seeming probable stems not from being actually possible, but from human brain cognitive bias of memory. In our head, past events are almost as real as current events, therefore creating obvious illusion of possibility of traveling in reality just like you travel in memories.

I would like to see a unified explanation that describes how our current modes of time travel work.

How does the matter that comprises my car and myself successfully get from Palo Alto to The Mission every morning?

If you had a program that generated a random number, and somehow send a hash of said number to the past to be used as a salt to generate itself. Would the new random number be truly random?

Time to dust off my old flux capacitor again...

lol, can't believe this gets down-voted. God forbid there is humor on HN!?!

Reply on news.ycombinator.com