Hacker News new | past | comments | ask | show | jobs | submit login

Time loop logic is a hypothetical system of computation that exploits the Novikov self-consistency principle. In this system the computer is able to send the result of a computation backwards through time and rely upon the self-consistency principle to force the sent result to be correct.

I've written about this before. Instantaneous data transfer with zero lag is theoretically possible through use of quantum entanglement. You can use this same technique to build a doomsday clock similar to the film "Tomorrowland".




> I've written about this before. Instantaneous data transfer with zero lag is theoretically possible through use of quantum entanglement.

I hope you wrote a correction wherever you wrote this. Quantum entanglement can not transfer information by itself (because of the no-communication theorem) therefore any data transfer protocol that uses entanglement must also use some other form of communication (e.g. classical communication), and therefore is at most as fast as your other communication method is.


Maybe in general, but this is in the context of "time loop logic". Ordinarily, you can have entangled states where measuring both parts will produce the same (random) outcome. Unfortunately the outcome cannot be influenced by either party, so it's useless for communication, as you say.

Now suppose that the first party resolves to go back in time and kill his grandfather unless the bits he measures are precisely the message he wishes to send. The universe can't tolerate the paradox, so it is corralled into the only non-paradoxical outcome: both parties read the desired message. Something like that, anyway.


Ah I guess I misunderstood what was being claimed. I thought the claim was "you can use entanglement to instantly send messages", but it's actually "you can use entanglement and this model of how time travel might work to send messages instantly".

Obviously these are rather different. In fact I feel like the entanglement doesn't even help in the time-travel protocol. If you want to send messages instantly and you can send messages back in time, you can "just" send yourself the message distance/c seconds in the past and send it then. Some messing around let's you arrange for it to arrive at exactly the time you want to send it in the future.


Assuming what I wrote above is correct (I don't actually know anything about closed time-like curves), it's actually weirder than just going back in time and mailing a letter early so it arrives when you want. Specifically, once you set up the entangled state, no other communication between the parties is necessary.


Yeah I understand the proposed protocol using entanglement. I'm just saying that if you're assuming we have access to the ability to send messages back in time there are much simpler protocols to achieve instant communication /without/ having to do anything involving entanglement.


IMO this is like those discussions where someone says FTL communication is possible by using a very long solid rod, jiggling one and and monitoring the other... except on closer inspection it'll only work if the rod is infinitely rigid, which is already a violation of many of the same underlying physical laws.

So it's a kind of circular logic, or at best an observation that having one kind of supernatural magic power would let you cause another kind of supernatural magic outcome.

In this case, if only I had a time machine I could break the speed of light. (Or possibly vice-versa.)


https://en.wikipedia.org/wiki/Novikov_self-consistency_princ...

> The principle asserts that if an event exists that would cause a paradox or any "change" to the past whatsoever, then the probability of that event is zero. It would thus be impossible to create time paradoxes.

Computer says no.


> Instantaneous data transfer with zero lag is theoretically possible through use of quantum entanglement.

No, that's a common myth about quantum stuff.

The entangled relationship can't be used to transfer information on its own. Whatever you sample from the "instant parts" can only be checked and understood after additional context arrives in a conventional way--limited by the speed of light and the flow of time.

https://www.space.com/41968-quantum-entanglement-faster-than...

There are also other quantum things that the youth of today need to be educated on:

https://www.smbc-comics.com/comic/the-talk-3


Ok human. You can read the Hans Moravec paper on CTC and time loop logic to understand Novikov self-consistency principle is formalized, yes you must run the computation for as long as it would have already taken or result will never arrive.

11 years ago I wrote some examples of how password cracking would look like using such setup. https://marak.com/blog/2013-05-13-time-loop-software

I say through "through use of quantum entanglement" as glossy term for reader. Crystalline quantum circuits are not really existing yet so is all scifi.

I won't be able to repsond to further posts in this thread. Cheers.


Novikov makes some big-ish assumptions; the no-communication theorem scores a bit higher on the ol' razor for me...


The salient bit from _Marak_'s article:

Time-loop logic does not violate causality. We are able to retrieve the answer instantly because we have committed to spending sixty seconds in the future calculating the answer and sending it back.

Maybe quantum superposition is just the universe compiling.


The blog-post asserts causality is not violated, but the author contradicts it elsewhere by claiming the hypothetical system would grant other benefits like "zero latency" gaming, which in turn means arbitrary data (e.g. human decisions) can somehow travel back-and-forth faster than light.

Then FTL communication opens up an enormous can of causality-worms.


That's a point, and I let them expand their definition of causality to include a chain of events flowing in a reversed arrow of time. I think they hand-wave away all those worms with the precept that any attempt to cause a paradox or other inconsistency would fail in practice. But I'd love to hear some examples that come to mind of problems that crutch doesn't fix.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: