4 ms·

I'm a bit astounded that the author has a PhD and is seriously proposing that time travel is the source of the correlation in an Arduino-based "time machine" an

I'm a bit astounded that the author has a PhD and is seriously proposing that time travel is the source of the correlation in an Arduino-based "time machine" and hasn't really explored any other options like the pseudorandom number generator being inaccurate. How does this get on HN and does anyone actually believe it? EDIT: I'm looking through the preprint PDF and it looks like they selected the future action (how long to expose for) using the last bit of the previous observation as a random seed (see page 9). That will of course create a correlation from the past event to the future action, and they use that correlation to try to claim that the future action is impacting the past observation. It's 100% bogus and terrible science. I don't know how they could possibly not see an issue with using the very measurement they're trying to find a correlation with as the source of randomness for deciding the future action instead of literally any other pure RNG available. It really seems like they were maliciously creating that correlation for academic clout, job prospects, etc.

314 5y ago
Responding to your edit: that is not what it says in the preprint. The last bit is used to decide the rounding mode for converting the rand in the interval 0-1 to an index of the shuffled array. They do not specify how the seed was picked at all.
- jmossbridge 5y ago
  That's correct. In fact, I give the code in the paper. FYI -- the same effect has recently been replicated by a Berkeley physicist using a totally independent random number generator. But I understand the concern here; that's why I asked him to try to replicate it using better equipment. As to the concern about it being Arduino-based -- it's easy to criticize experiments because they don't use sophisticated equipment. However, what actually matters is if you can find an explanation for why the student-level equipment would result in the artifact(s) you suspect are there. That's more difficult to do, and in fact no one has been able to do it yet. We will see how this shakes down, but I've been doing the experiment continously for two years and the effect is robust. Thanks to everyone for reading the article. The argument is good to read. It's a delight to me that people are thinking about these issues, since as should be clear from the article, I think they are important to think about. And, of course, I certainly don't have all the right answers. But science isn't about that -- it's about allowing yourself to be brave enough to find questions that you are curious to answer (even if no one else is) -- and rigorously trying to discover the answers. Or at least, further questions. -- Julia
  - jmossbridge 5y ago
    P.S. This book just came out that should be interesting to a few folks here: https://ndpr.nd.edu/reviews/time-travel-probability-and-impossibility/ https://ndpr.nd.edu/reviews/time-travel-probability-and-impo...
  - walterbell 5y ago
    Thanks for posting on HN! I'm still hoping to see more papers on this much neglected topic. Science depends on experiments, inspired by good questions. Many software developers work daily within infrastructure (e.g. git source revision) with features to model and manage event sequences/history. If a software developer works on multiple projects, or variations of one project over an extended time, they quickly find themselves negotiating with past selves (software archaeology) and future selves (technical debt). Another petri dish for time travel concepts is the finance industry, which has developed countless financial instruments to estimate the present-time value/risks of future-time assets. Compute infrastructure is used to model financial futures, most of which never happen, so we can collectively encourage a small subset to become real. The lives of some HN participants have been changed by just one of those instruments (stock options) and their time-travel (tax, vesting, bonus, cliffs, retention, signaling) incentives. Cryptocurrency "smart" contracts can commoditize math-based social engineering of present behavior in pursuit or avoidance of future events. What's the game-theoretic Prisoner's Dilemma equivalent, during a mental negotiation with hypothetical past, future or even concurrent selves?
    - jmossbridge 5y ago
      Super interesting points, IMHO. And it brings up the point that while I've talked with many folks in different fields about time-travel thinking, I have never found a field where it's not useful to bend one's mind in this way. I've come to the conclusion that we have to shift the way we think about ourselves in time, so we can navigate more successfully as a species. I suppose that's obvious from the article. Anyway, to address your question re: Prisoner's Dilemma equivalent within a time-travel paradigm -- isn't that what virtually every Dr. Who episode is really asking? But more recently -- to my view the movie Arrival successfully solves a time-travel-inspired Prisoner's Dilemma by modeling the Present Amy Adams as capable of negotiating with her Past and Future selves (as the captors) and bringing actionable tactics into the Present. So she not only succeeds in being released from prison, she saves the world too. So it's communication and collaboration with the captors (the non-Present versions of ourselves) that can turn them (us) into advocates. Meanwhile, the other prisoners are the aliens, who have access to the same strategy over time (and use it to teach Amy Adams' character). There can't be any clear communication between them, but they can circumvent this by connecting in the future and being aware of that connection in the Past/Present. So it's a brilliant solution to the prisoner's dilemma -- just because you can't communicate in the present doesn't mean that after you get released you can't share how you did it! The interesting thing about the future, in general, is it's way less crowded than the past and present -- because the past and present are where most people think you have to do things to get them done.
bpiche 5y ago
Robert Anton Wilson or PKD might have liked this. I think it's adjacent. Reminds me of the Pi film. Pretty crazy.
mrjangles 5y ago
These crackpot papers appear on the archive all the time. I know a few physicists who rejoice in reading through them for a laugh. I'm quite sure that would be why it was posted... I hope.
JKCalhoun 5y ago
I guess I read it as part tongue in cheek, part meditation on life and self (and past selves), and part criticism of modern physics.

314 5y ago

Responding to your edit: that is not what it says in the preprint. The last bit is used to decide the rounding mode for converting the rand in the interval 0-1 to an index of the shuffled array. They do not specify how the seed was picked at all.

jmossbridge 5y ago
That's correct. In fact, I give the code in the paper. FYI -- the same effect has recently been replicated by a Berkeley physicist using a totally independent random number generator. But I understand the concern here; that's why I asked him to try to replicate it using better equipment. As to the concern about it being Arduino-based -- it's easy to criticize experiments because they don't use sophisticated equipment. However, what actually matters is if you can find an explanation for why the student-level equipment would result in the artifact(s) you suspect are there. That's more difficult to do, and in fact no one has been able to do it yet. We will see how this shakes down, but I've been doing the experiment continously for two years and the effect is robust. Thanks to everyone for reading the article. The argument is good to read. It's a delight to me that people are thinking about these issues, since as should be clear from the article, I think they are important to think about. And, of course, I certainly don't have all the right answers. But science isn't about that -- it's about allowing yourself to be brave enough to find questions that you are curious to answer (even if no one else is) -- and rigorously trying to discover the answers. Or at least, further questions. -- Julia
- jmossbridge 5y ago
  P.S. This book just came out that should be interesting to a few folks here: https://ndpr.nd.edu/reviews/time-travel-probability-and-impossibility/ https://ndpr.nd.edu/reviews/time-travel-probability-and-impo...
- walterbell 5y ago
  Thanks for posting on HN! I'm still hoping to see more papers on this much neglected topic. Science depends on experiments, inspired by good questions. Many software developers work daily within infrastructure (e.g. git source revision) with features to model and manage event sequences/history. If a software developer works on multiple projects, or variations of one project over an extended time, they quickly find themselves negotiating with past selves (software archaeology) and future selves (technical debt). Another petri dish for time travel concepts is the finance industry, which has developed countless financial instruments to estimate the present-time value/risks of future-time assets. Compute infrastructure is used to model financial futures, most of which never happen, so we can collectively encourage a small subset to become real. The lives of some HN participants have been changed by just one of those instruments (stock options) and their time-travel (tax, vesting, bonus, cliffs, retention, signaling) incentives. Cryptocurrency "smart" contracts can commoditize math-based social engineering of present behavior in pursuit or avoidance of future events. What's the game-theoretic Prisoner's Dilemma equivalent, during a mental negotiation with hypothetical past, future or even concurrent selves?
  - jmossbridge 5y ago
    Super interesting points, IMHO. And it brings up the point that while I've talked with many folks in different fields about time-travel thinking, I have never found a field where it's not useful to bend one's mind in this way. I've come to the conclusion that we have to shift the way we think about ourselves in time, so we can navigate more successfully as a species. I suppose that's obvious from the article. Anyway, to address your question re: Prisoner's Dilemma equivalent within a time-travel paradigm -- isn't that what virtually every Dr. Who episode is really asking? But more recently -- to my view the movie Arrival successfully solves a time-travel-inspired Prisoner's Dilemma by modeling the Present Amy Adams as capable of negotiating with her Past and Future selves (as the captors) and bringing actionable tactics into the Present. So she not only succeeds in being released from prison, she saves the world too. So it's communication and collaboration with the captors (the non-Present versions of ourselves) that can turn them (us) into advocates. Meanwhile, the other prisoners are the aliens, who have access to the same strategy over time (and use it to teach Amy Adams' character). There can't be any clear communication between them, but they can circumvent this by connecting in the future and being aware of that connection in the Past/Present. So it's a brilliant solution to the prisoner's dilemma -- just because you can't communicate in the present doesn't mean that after you get released you can't share how you did it! The interesting thing about the future, in general, is it's way less crowded than the past and present -- because the past and present are where most people think you have to do things to get them done.

bpiche 5y ago

Robert Anton Wilson or PKD might have liked this. I think it's adjacent. Reminds me of the Pi film. Pretty crazy.

mrjangles 5y ago

These crackpot papers appear on the archive all the time. I know a few physicists who rejoice in reading through them for a laugh. I'm quite sure that would be why it was posted... I hope.

JKCalhoun 5y ago

I guess I read it as part tongue in cheek, part meditation on life and self (and past selves), and part criticism of modern physics.