6 ms·
It doesn’t matter, it still implies time travel. Use the wormhole acceleration-return thought experiment as an example. If you can leave point A and arrive at p
by GW150914 8y ago
It doesn’t matter, it still implies time travel. Use the wormhole acceleration-return thought experiment as an example. If you can leave point A and arrive at point B before light could, it doesn’t matter if you took a shortcut.
- lainga 8y agoOK, touche - it's true that an Alcubierre drive could be used for time travel, but it's not necessarily implied, whereas it is for an object actually travelling beyond the speed of light. The discoverer of the method itself [0] (Miguel Alcubierre ) believed that chronology protection [1] would prevent time travel, as the spacetime bubble would destroy itself somehow. [0] https://en.wikipedia.org/wiki/Alcubierre_drive https://en.wikipedia.org/wiki/Alcubierre_drive [1] https://en.wikipedia.org/wiki/Chronology_protection_conjecture https://en.wikipedia.org/wiki/Chronology_protection_conjectu...
- GW150914 8y agoIt’s a cool idea, and admittedly would be a sort of loophole, but it might just mean that such a drive would require an arbitrarily high amount to work. Maybe Stephen Hawking was right and as soon as you build a machine capable of time travel, turning it on blows it up, because of QM effects. It does seem a little neat if how we intend to use the machine is the deciding factor in how it works.
- AnimalMuppet 8y agoIt also could blow up if it accidentally causes time travel...
- sls 8y agoBack in his collection All the Myriad Ways, Larry Niven wrote a brief essay called "The Theory and Practice of Time Travel" that explored ideas like these. One excerpt: > GIVEN: That the universe of discourse permits both time travel and the changing of the past. > THEN: A time machine will not be invented in that universe. > For, if a time machine is invented in that universe, somebody will change the past of that universe. There is just too much future subsequent to the invention of a time machine: too many people with too many good motives for meddling with too many events occurring in too much of the past. thus > Niven's Law: If the universe of discourse permits the possibility of time travel and of changing the past, then no time machine will be invented in that universe.
- jerf 8y agoI actually spent some time sketching out the math for an aborted science fiction world in which I had FTL that was prevented from time travel by a cosmic censor. You don't really have to "work it out" because the numbers are quite plain as day. It turns out that if you use any remotely sane numbers on the speed of an FTL craft, it is extremely difficult to travel in time. To get time travel, you have to accelerate on the other end to very significant fractions of the speed of light, which itself takes so much time that it's hard to recover it via the FTL drive again. You have to put in absurd numbers for the speed of the drive and your ability to accelerate (and your ability to carry energy for conventional acceleration!) before you can get time travel out of most FTL drives, because bending your light cone enough to put the past into your FTL drive's path takes a lot of energy. The only fictional drives that have serious time travel possibilities are the ones in universes where accelerating to significant fractions of the speed of light is easy-peasy, and the FTL drive is very, very fast (some are instantaneous) and can be arbitrarily steered. It's actually not that long of a list; a lot of FTL drives can hypothetically be used for time travel but the engineering in the universe still doesn't practically permit it. Just adding a cosmic censor post facto would preserve the story just fine. Many of the universes that meet those criteria are already so fantastical that they do in fact have time travel, so... those work out too. (There's also several video universes where it's nominally possible, but the video wildly contracts the documentation. Both Star Wars and Star Trek have documentation that their ships can get to pretty significant fractions of the speed of light in small amounts of time, with ludicrous amounts of acceleration available to them, but on screen we only see them moving with absurd levels of acceleration like 50 or 100G, but which still takes a long time to get to light speed as a result, and with refueling being necessary often enough that it's hard to believe they could actually make it.)
- dragonwriter 8y ago> Both Star Wars and Star Trek have documentation that their ships can get to pretty significant fractions of the speed of light in small amounts of time, with ludicrous amounts of acceleration available to them, but on screen we only see them moving with absurd levels of acceleration like 50 or 100G, but which still takes a long time to get to light speed as a result, , and with refueling being necessary often enough that it's hard to believe they could actually make it.) I don't know about Wars, but Trek strongly implies that impulse drive reaches significant fractions of light speed with near-zero acceleration time; there is certainly nothing to indicate that the on-screen non-warp acceleration is limited to 50-100G, and there is little on-screen indication that refueling (as opposed to drive system maintenance, and replacing non-fuel consumables used onboard) is regularly needed.
- SubiculumCode 8y agoAny progress on that idea (Alcubierre drive)? Always seems the most interesting. I do a Google Scholar search, but I don't know the physics journals, so its hard for me to know legit from crank abuse, without laboriously checking each result for journal impact factors.
- TheOtherHobbes 8y agoThe debate continues. E.g. https://www.hindawi.com/journals/isrn/2013/482734/ https://www.hindawi.com/journals/isrn/2013/482734/ I suspect using GR to design a warp drive is like using a spoon to build a freeway. Without a theory of quantum gravity - which appears to mean not just a theory of quantised space time, but also of the non-locality and complex distributed relationships that haunt QM - you're probably not going to get very far.
- fooker 8y agoWhy does this imply time travel? You could look backwards in time, sure, but we can already do that.
- surge 8y agohttps://www.youtube.com/watch?v=HUMGc8hEkpc https://www.youtube.com/watch?v=HUMGc8hEkpc
- flatline 8y agoDoesn’t the wormhole scenario require bending the thing back on itself? I always thought that just because something could violate causality doesn’t necessarily mean it would, but IANAP and maybe that distinction doesn’t matter.
- ummonk 8y agoThat assumes that you can accelerate a wormhole mouth...
- monochromatic 8y agoHow does that imply time travel? Yes, you get there before the photons, but so what?
- aristus 8y agoIf you go from A to B before the photons can, from B's reference frame you are arriving from the future with perfect knowledge of events that have not yet "happened".
- SimbaOnSteroids 8y agoYes however if you were to pop back to A you wouldn't have any future knowledge at A (minus the 'future' knowledge of B).
- monochromatic 8y agoBut you can’t go back to A and change those events before they happen. That’s not time travel. You could get the same effect by just blocking the photons before they make it to B, and then traveling at conventional speed.
- trentlott 8y agoThat's always struck me as sort of inconsequential, even though light is the apparent "speed limit" of the universe. The same thing happens with sound regularly. A thing happens, and we receive light before we receive sound. Some third faster thing that gets to us before light would just be some other thing. The distance of travel requires time relative to both reference frames. If a star explodes at A and I travel to B at 2x the speed of light to tell folks at B, so what? Some time has elapsed and those two frames are still chugging along with causality. Maybe there's some physics I've not learned that makes this obvious, but I've never grasped exactly why violating light speed prevents entropy from continuing. Even if at two different rates. If we perfectly timestamped everything happening on two sides of the world and sent a telegram at twice lightspeed from A to B, it would still arrive after it is sent.
- 8y ago
- hinkley 8y agoThe thing that always bugged me about time travel fiction is that the reference frame is just assumed to be the surface of the earth. How fast are we wizzing through the universe? Why wouldn’t I reappear somewhere out past Saturn when I travel forward or backward in time? Similarly, if wormholes are real, and you construct one in a gravity well, what happens to gravity? Especially if both ends are in a gravity well?
- hossbeast 8y agoDoes the influence of gravity reach through the wormhole?
- hinkley 8y agoIf a wormhole is a 'fold' in space, instead of a magic teleportation device, why wouldn't gravity and light and EM fields travel through the wormhole?
- andrewflnr 8y agoIt seems that the usual thing to do is put the wormhole in orbit around your planet or whatever. That's based more on science fiction than physics, albeit scifi by smart knowledgeable people. I'm not sure what to think about gravity "reaching through". Gravity and the wormhole are both just curved spacetime. Maybe the larger body distorts the wormhole throat in a way that looks like "reaching through"? Certainly, you need to have some effect that causes energy conservation when you make a perpetual water wheel by suspending one mouth over another and running water through it...