The Challenge of Relativistic Spaceflight | Hacker News Reader

The Challenge of Relativistic Spaceflight

icarusinterstellar.org

78 pointsuser_23571186 comments

TL;DR: Chemical propulsion is a dead end. But, the alternatives require matter/energy of ridiculous scale (hundreds of times the power output of human civilization, launching hundreds of millions of tons of matter into orbit). We're not going anywhere, any time soon.

That's not true. Project Orion would have been propelled by detonating small nuclear bombs, which have huge amounts of energy. It would be absurdly expensive, but technically feasible today.

Personally I would prefer alternatives that didn't involve "detonating a shit ton of nuclear weapons in the atmosphere" or "sending lots and lots of uranium and/or plutonium into orbit on rockets".

I believe they never figured out the necessary shock absorbers, so manned flight wouldn't be feasible today.

Not just magnitude but duration. Name one machine that works without major overhaul for more than 100 years. For bonus points, name a machine that's significantly more complex than a clock.

Name one machine that's designed to work for over 100 years.

Many of NASA's spacecraft have survived well beyond their intended death dates, because they were constructed so well, and because space is cold, static, and good at preserving things.

There are plenty of simple, well-made machines designed to work indefinitely -- e.g. doorknobs, window assemblies. Even simple, robust, well-maintained machinery tends not to last a very long time. There's a stone bench somewhere that Thomas Jefferson supposedly used to sit on that has become curved over time.

Are you seriously going with "a stone bench" as your example of the durability of "robust well-maintained machinery".

I'm pretty sure he's taking the opposite view, that (a) building elements have indefinite design lives but don't last that long, and (b) Even stonework doesn't endure indefinitely.

I'm not sure I really agree with the second point in this context - stonework tends to endure. We have Norman cathedrals that are looking good at nearly a thousand years old, comparable to the timescales under discussion to spaceflight. More to the point, there's no reason to believe that the basic structural elements of, say, a hollowed nickel-iron asteroid wouldn't be durable enough for spaceflight.

I live in a mostly original (at the time I purchased it) 1950s home, and I can assure you that doorknobs and window assemblies do not last indefinitely.

http://www.wired.com/gadgetlab/2011/06/10000-year-clock/

Hence my bonus points ;-)

You're unlikely to see any project designed to last more than a few generations simply because the original stake holders won't live to it's end. Also, there's the opportunity cost -- technology usually advances which makes the investment retroactively not worthwhile.

What sort of major overhaul do you have in mind? A spacecraft on a multi-century flight would presumably have continuous ongoing maintenance and repair, but would need to function without a spell in space-dry-dock halfway through its journey.

So to answer your question, what about the London Underground? It's a pretty complex machine that has been running for 120 years, with various overhauls, but of the frequent short shutdown type rather than total closure and redevelopment.

The challenge for a spacecraft is being a closed system, but it's just an engineering problem.

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

Solves the "anywhere", but not the "any time soon".

The numbers only become science-fictiony large if you want to go fast. Why not go slowly?

The only reason humanity is obsessed with speed is because of our short life spans, but that is a solvable problem.

All we have to do is to eliminate aging and it's associated diseases, and then we can take our time going to the stars.

The most difficult part may be accepting limits upon our appetites, so that we are not driven to consume everything and we can live sustainably. This is an issue aboard a starship, where we will have to take a chill pill on consumption of resources, just as it is an issue on starship Earth.

Ignoring the whole "we'll just cure mortality...how hard can it be?", people tend to go batshit crazy after spending a couple of months in confined conditions looking at the same faces day after day. Think they're going to do better mentally when they look forward to the same for a 100 years.

Then we have to address this newly pinpointed problem - to somehow make people not „go batshit crazy”. Maybe practicing some kind of meditation? I admit that it's not as exciting as some technical overhaul we all expect, but still...

What's the point of these huge generation ships? It's not like there's not enough atoms on other planets. You should send a tiny fast robotic ship, maybe with an uploaded mind if your AI tech isn't good enough. It will mine materials at the destination and build a receiver antenna. Then you can send more uploaded minds by radio, which will arrive at the speed of light with low energy cost.

Another reason why you shouldn't send human bodies in tin cans, or even grow them from eggs on the spot, is that human bodies are not adapted to the conditions in space or at the destination, so they would need an unreasonable amount of life support equipment. If you really want a colony of biological creatures with human minds for some reason, bioengineer the right kind of bodies on the spot and download minds into them.

If we're going to stipulate fantastic concepts like "uploaded minds" then let's just invent a hyperspace engine too and solve all the problems at once.

In the real world, we want to send human bodies because that's what human minds are. Cartesian duality is a religious/philosophical concept for which no scientific basis has yet been found.

>If we're going to stipulate fantastic concepts like "uploaded minds" then let's just invent a hyperspace engine too and solve all the problems at once.

We don't know if breaking the light-barrier is possible regardless of a civilizations level of technology. Brain uploading will certainly be enormously complex and a fantastic feat of technology, but it's certainly possible based on everything we know.

Yes it's possible there is more to consciousness than just the physical brain, but that's such an absurd and outlandish assertion, it isn't remotely worth the time to worry about it.

I think you have "fantastic" and "real world" backwards. The Blue Brain project is planned to have an uploaded rat brain this year (2014). Uploading human minds is very likely to be closer than capturing the whole energy output of the Sun, which is discussed in the OP.

Though uploading will likely cause so many changes that space exploration will seem insignificant in comparison, see Robin Hanson's "The crack of a future dawn" [1] and Carl Shulman's "Whole brain emulation and the evolution of superorganisms" [2]. In general I think most people don't take this scenario seriously enough, the authors I mention are the rare exceptions.

[1] http://hanson.gmu.edu/uploads.html

[2] https://intelligence.org/files/WBE-Superorgs.pdf

The Blue Brain project will have a rat brain model. It reproduces some of the responses you'd expect from a real. That's orders of magnitude of technical progress away from what an upload would require. We don't even know for sure if it's physically possible to perform an upload (due to quantum mechanics), much less have a device which can come close to measure a significant part of the quantum state of a working brain (i.e. the technical problems beyond physical feasibility).

Can you point to some introductory reading about the quantum mechanical effects at work in the brain? I've read The Emperor's New Mind but was left with the impression that it was more conjecture than research.

I'm afraid you fell for the hype. We're still unable to understand the puny brain of Caenorhabditis elegans, or how the bees achieve all that they can do with as little a brain. Rat brain is several orders of magnitude harder.

Uploaded minds is hardly an argument depending on cartesian duality. Arguing that there's something irreproducible about the human brain is quite close to cartesian duality.

I am not arguing that uploading minds is any more feasible than a generation ship, but it is at least more plausible than hyperdrive -- based on what we think we know right now.

I guess I disagree that uploading minds is more feasible. One is more of an engineering/physics challenge, and uploading minds is dependent upon a vast set of unknown unknowns. We aren't even out of the door on the idea of biological computing let alone at the mailbox.

One key difference is that we know brains can exist. We might not know the entirety of how they work, but the important thing is we already know they can be reproduced in some fashion.

Hyperdrive or any other form of FTL travel is purely in our thoughts. Nothing currently exists that even proves it is possible in our current reality. For all we know we could spend an infinite time attempting it and never achieve it.

True, however a key distinction i'm making here is knowing something exists and knowing how it works and being able to replicate that in another system could be just as much of an infinite timesink.

I'd rephrase the "nothing currently exists that even proves it is possible" to, nothing has been observed to demonstrate its possibility.

That and things like warping space could throw a wrench into the "we can't travel faster than light". But maybe we can end run around it. Though the same can be said of the brain, we really know we don't know a lot more than what we don't know about physics.

We have no evidence at all that a given brain can be reproduced. We know that new brains can be made and developed (i.e. procreation), and even then we don't really know how that works.

But the post above said "mind" not "brain" which typically signifies that the author draws some sort of distinction between the two. I'm just saying: I don't know of any physical evidence for that distinction.

Human brains are no more difficult to exactly reproduce than any other piece of matter: extremely difficult. The best we've done to date is a few previously entangled particles.

Edit to add: what I'm talking about here is quantum teleportation of the matter of the brain.

Typically when people talk about "uploading a mind" they mean that the mind is software running on the hardware of the brain--like Windows running on a PC. I'm not aware of any scientific evidence that the mind and brain can be considered separately in that way.

The question comes down to "could you run a sufficiently accurate simulation of a brain on a sufficiently powerful computer that it could pass for the mind of the person whose brain was being simulated"? I'd say the belief that the belief this is impossible is a variant of the cartesian duality.

The only problem is when you send ships to mine other planets, there may be eggs with face-huggers in them on those other planets.

I heard about this interstellar communication silver-bullet before. You're just taking communications on earth for granted - need more bandwidth between points A and B? Just throw in another independent physical communication channel between them! Well, that's fine here but it doesn't work the same way at creating data links between Earth and another celestial body inside our solar system and definitely not between the solar system and another star system. Think how much info can we theoretically send at a time to another star and how much time will it take to send a brain of knowledge! I think it will be more feasible to send those brains of knowledge in some physical form (and guess what the most intuitive way for doing that would be).

Hmm. Theoretically it seems like lasers should support speeds of terabytes per second, so you should be able to transmit a human mind in minutes to hours (maybe faster with compression), and that should take less energy than sending an actual brain at reasonable speed (rocket equation). If you want more bandwidth, use multiple lasers located close to the solar system, but still far enough that they can be distinguished from another star. At longer distances, use multiple intermediate stars in parallel, like the internet. Or is there some theoretical limit that I'm missing?

"is there some theoretical limit that I'm missing?"

Well, considering that the closest star (Alpha Centauri) is 4.367 light-years away, that only means that if something's wrong with the transmission itself, we can figure it out only after almost nine years. Other than that, I don't really imagine any feasible solution for communication on interstellar scale other than laser-based one, so my initial point was put forward exactly in that use-case. If there ever will (and I hope it will) be a galactic-scale communication system, humanity will have to:

- Improve the current laser technology to the point of sending laser beams without much dispersion over light-years distances.

- Improve the optical sensor technology in order to receive (and distinguish) from (multiple) artificial sources of light orbiting close to a noisy star.

- Set up the laser system(s) taking in account: the time-frame of the moving celestial bodies - both stars and the orbiting emitter(s)/receiver(s), the gravitational time dilation factor, other currently known factors that I don't know of (I'm not a professional physicist), and whatever any other new additional subtle effect that may be still discovered in such setting! I think this fine calibration will be the cake's top-cherry, BTW.

So yes, theoretically all is just an engineering problem. The development speed cycles of such systems though, are already dictated by the laws of physics - years (and in human time - generations). The mere reliable communication on such scale will be a great success, but I don't dare to hope of sending brains of knowledge['] in mere hours.

['] http://www.scientificamerican.com/article/what-is-the-memory...

Master the art of "the transporter" with information being sent via optics. Send a ship out into space. Aboard will be an environment for humans and a big receiver on the side. Send the ship. Wait until it reaches it destination. "Beam" humans to the ship. If boredom ensues, reverse the transmission. Leave the "Stargate" behind. :)

Edit: see quantum entanglement

Edit 2: Thinking aloud about quantum entanglement. I understand that we can slow and stop light by passing it into a Bose-Einstein condensate. Can we thereby suspend two entangled particles, each in its own BEC? That way, we can send our ship out into space, unfreeze light (time?) and proceed to transmit Data (heh).

Edit 3: Transmit, that is: faster than the speed of light. (thinking not) :)

Edit 4: Now, I want to know what happens when a BCE containing stopped light travels at or near the speed the speed of light.. Ah never mind. I'm not a physicist. Cold and fast don't go together easily, either. :)

Edit 5: Do watch: https://www.youtube.com/watch?v=-8Nj2uTZc10

Edit 6: By the way, if Edit 2 is used, no big receiver would be required. Hmm.. Send the quantum receiver to a Class M planet.. Skip the on-ship environment. Terraform to taste.

You can't use entanglement to do FTL communication.

http://en.wikipedia.org/wiki/No-communication_theorem

Glad to learn something new. Thanks.

Edit 7: Learn what all those words really mean and how they relate; drop the whole thing a get a beer.

Thanks for the tip. Ever thought about going into teaching?

Yeah. And I do.

I think we'll need fusion to make intra-solar system travel (really) possible, where we just transport stuff through-out the entire system, with a speed of at least 1/100 of the speed of light. Achievable in ~50-100 years.

For inter-stellar space travel, say to the closest 20-100 star systems, we'll need either an anti-matter powered ship or a black-hole powered one (a bit more likely) that can travel to these stars in only a few years to a decade, and then another few years to a decade back, with the speed of 30-50 percent of the speed of light. Achievable in ~200-300 years (remember technological progress is exponential not linear).

For whole galaxy exploration we'll definitely need something like a warp engine. When we'll get this is harder to predict, but it will probably take at least 1,000 years to do it.

Things that could definitely help with space exploration: figuring out immortality against aging, which could also make us more interested in 100-year old projects, transferring our consciousness to machines/more resilient avatars not just for space travel, but also for living on new dangerous planets, and quantum Internet so we can rapidly communicate through long distances.

If we do the last one, we might not even need to travel ourselves or transfer into other bodies, to go to those places. We'd just send robots that we control absolutely and very precisely, just like we were there (Avatar/Surrogates-style).

> (remember technological progress is exponential not linear)

No it is not. For example, airliners fly about as fast as they did a generation ago, and if you buy a desktop computer today it won't be noticeably faster than the ~3 year-old one you're using now.

Do you really think the next TV you'll buy will be exponentially bigger than the one you have today? Prepare to be disappointed.

warp-engine would require FTL travel, which is not possible. How can you travel a distance less than zero? At C all distances are zero.

DNA is already a pretty good way to travel the stars, a human ovum (the largest human cell) has a low enough mass to get out of Earth's gravity well using only electrostatic or solar-pressure forces, no fuel required. If it were somehow interstellar-vaccum-hardened and able to modulate the energy gradient it uses to change direction, it could travel to the stars.

I suppose you have misunderstood quantum entanglement, as it does not offer any way to communicate instantaneously over vast distances. It's more like when the light-standard in front of you turns green, you know INSTANTANEOUSLY that the "entangled pair" light-standard perpendicular to your view is red.

It's not quite valid anymore to say warp drive is flat impossible, given all the theoretical work on the Alcubierre drive. Whether it will ever actually be practical is still a big unknown.

you mean the drive that if it were used it would annihilate your destination in a immense-burst of gamma energy? Not to mention we have no idea how to effectively modify space-time. It is interesting, but the drive is basically "if we had MAGIC, we could do this".

FTL is also exactly the same as time-travel. If you could travel 25,000 light-years in an instant, you could turn a BIG telescope back at earth from your ship and watch early humans hunting mammoths.

I'm not saying it will ever work, just that those sorts of objections are quite different than "relativity says it's impossible, full stop." Modifications have improved the practicality from the original design, eg. by reducing the energy requirement from Jupiter mass to the mass of a small car.

The way to modify space-time would be to use negative energy, which exists in theory but maybe not in any practical way. We don't know.

Regarding negative-energy; I think you mean "nothing in our current understanding prevents something like what we describe as negative-energy from having existed at some time in the life of the Universe". Very different from "exists in theory".

Well, that is just the light. Does it count as time travel when we look at stars that are thousands of lightyears away?

yes. We see the stars today as they were thousands of years ago. It's thousands of light years away from our perspective, but from the perspective of the photons the trip took zero time as it was a zero distance. SR is weird like that.

I saw a lecture from the researcher that is working on the project and he explained why there isn't a burst of deadly energy. Also they are working on those experiments, so they have at least some idea how to do it. You should really update your data.

wow, someone tell the Nobel people that this guy has experimentally tested space-time modification and shown it does not accumulate bow-energy as it covers distance. I kid, but those that claim the bow-energy problem also have to prove it for it to be a real problem, and your comment seems to indicate the lecture was much more of a grant-money pitch.

How would the ovum develop and be nurtured? Sympathetic alien surrogates?

Why not seed more simple lifeforms, such as genetically modified bacteria, upon hundreds of "candidate" planets so that by the time our generation ships arrive there's a good chance some basic terraforming has already occurred.

It was more the thought exercise itself than a practical approach. I just wanted to consider the benefits of very low mass offers to space travel.

Sounds like you're talking about panspermia, which does seem very likely to me as well. http://en.wikipedia.org/wiki/Panspermia

> How would the ovum develop and be nurtured? Sympathetic alien surrogates?

You could send along artificial wombs and some kind of computer/AI to raise + educate the first generation born.

Never been a parent, eh?

\shrug\ It's all sci-fi tech anyway. Artificial wombs probably aren't that far off: http://www.popsci.com/scitech/article/2005-08/artificial-wom...

There's a clear demand for advanced reproductive technology, and it would offer the secondary advantage of neatly sidestepping the current abortion debate.

You're missing my point (which I suppose I really didn't make, so my bad). A child raise by machines is likely to be one seriously emotionally messed up child.

Well, all kids are raised by machines, just usually machines called Mom and Dad. The question is whether the machines we could make to replace us as parents are better or worse. It's hard for us now to imagine artificial machines being better at it than we are, but, you know, so is relativistic spaceflight.

Oh, yeah, probably. Still, there wouldn't be any frame of reference. Maybe the first generation of kids would all be weird and emotionally unstable by our standards, but they're be "normal" for Alpha Centaurians (or whatever)

I must say that the idea of a single human cell traveling through space at relativistic speed, with the end goal of landing on a planet and growing into a fully formed human is one of the wildest and most unique ideas I've heard in a long time. Perfect sci-fi material. Totally awesome.

Agreed, although not entirely original. Inspired by actual work in this area - http://www.spacecraftresearch.com/MII/MII_overview.html

I was getting my head around what "100µm" and a "few micrograms" meant, and I found that the human ovum is ~3.6 micrograms and around 100µm in size.

One efficient way to convert mass to energy is using the Hawking radiation of artificial black holes. This seems to by allowed by the laws of physics according to this paper:

http://arxiv.org/abs/0908.1803

What I got out of that is first company or nation to manage useful self replication technology in space will gain immense power.

Doesn't your spaceship get ground down or blown to bits by interstellar dust?

Use a very tiny, highly reinforced spaceship about the size of a standard 330 ml can.

I think he may have meant 1/12th rather than 1/2th of the speed of light, but I must say I'm very tempted now to start calling one half a "tooth" instead.

No, he meant 1/2 the speed of light. The speed of light is 299,792,458 m/s, so about 300,000 kilometers per second. Half that is the speed he mentioned.

Maybe not "tooth", but why do we use the phrases "a third" and "a fourth" for fractions, but not "a second" for half?

Perhaps because of the potential confusion with "a second" being used to mean "another" i.e. 2X rather than 1/2X.

e.g. I could eat a second pizza

Do I mean another pizza or half a pizza?

[OK it should be phrased as "a second of a pizza" - but that is rather close to "a second pizza"]

Not that I have a better answer, I believe the phrase "a third of a pizza" is just as confusing.

You are, of course, quite correct.

I should make a mental note to comment on HN while actually focusing on something else..... :-)

Third can be used both as a noun and an adjective.

Noun form: "A third of a pizza".

Adjective form: "A third pizza".

You can say "a third of a pizza" and not be confused though, so there's no difference in saying "a second of a pizza" if that were indeed an accepted use of the word.

Secondth?

Because we should be saying "one third", "one fourth", etc., but it's a little bit faster to say "a third".

Is continually accelerating at 9.8 m/s^2 necessary? Sure, that's what's needed to get off the ground, but it seems excessive for a whole interstellar journey.

Well if you accelerate slower, then you travel slower, then the relativistic effects of your speed are exponentially slower. So it's good to accelerate as quickly as possible.

One challenge to relativistic spaceflight is computing. If you put a CPU into a vehicle, which goes close to speed of light, it may not work properly, because the energy pulses in the device will become unsynchronized when going in different directions.

Special relativity says the speed of light is the same in all reference frames. In the reference frame of the moving CPU, it would operate correctly.

While the speed of light is the same, relativity also has a lot to say about accelerating reference frames.

That said, if you were accelerating fast enough to affect the behavior of a CPU, you'd be experiencing G-forces comparable to the tidal forces of a black hole; computation would be the least of your problems.

Hmm, suggests a vaguely related question to my mind. Generally speaking, do black holes have any implications for computations? E.g. does having a black hole at your disposal let you solve any problems you wouldn't be able to otherwise?

What you are describing is a form of hypercomputation. It is not yet known whether hypercomputation is even really a "thing." A hypercomputer, if it existed, would be able to solve the halting problem.

http://en.wikipedia.org/wiki/Hypercomputation http://en.wikipedia.org/wiki/Malament-Hogarth_spacetime

I'm not a physicist but this sound weird. if this was true we could tell at which speed we are going and this would mean that not all inertial frames are indistinguishable, which is at the heart of relativity

Nope. All different parts of the starship are at rest relative to each other. No such effects would occur.

You seem to believe in the existence of an absolute frame of reference that all speeds are measured against. There is no such thing. "Speed" only exists as a relative measure.

Cool. I wonder if this holds true for quantum computation, as we theorize it anyway.

Reply on news.ycombinator.com