variable specific impulse magnetoplasma rocket interstellar capabilities?

[-Velocity-]

In that case, why bother with humans at all?

That may in fact be the case. It's hard enough to know where technology is heading, but at least it has to follow the laws of physics, and we probably already know all the laws we can practically make use of. Predicting how society will evolve is MUCH harder, especially because it is reactive to the technologies that may or may not be innovated. And predicting today how society will evolve when/if non-humans join it is pretty much impossible. Who knows. It may be that "we" lose the drive to establish colonies around other stars. I doubt we will lose the motivation to explore however, that is probably inherent in any intelligent being; curiosity is a necessary trait for survival, and all reasonably intelligent animal species have it to some degree. But establish colonies?

I'd argue that we're already seeing a push against the drive to expand; people want large tracts of land left as preserves, we punish countries that try to expand their boundaries by force or subterfuge. It's hard to imagine that we will never want to expand to another star system, but what if there are lifeforms already there? By what right do we have to move in and massacre an alien biology and replace it with our own? 100 years ago, that would have been morally acceptable, but today, not so much. We've come to recognize that the oppression or destruction of the weak by the strong, while practiced by nature, is a social evil. Just because nature favors it doesn't make it right; nature is neither right nor wrong it just is, and trying to draw moral lessons from it is foolish.

A lot of people will argue that humans are still just humans and don't change, but that's NOT the case when the influence of society is factored in- we're different than we were thousands or even hundreds of years ago because we grow up in a society that favors different moral values, and are "shaped" differently by it. And again, society may include non-humans in the not-that-distant future.

So who knows. Maybe, we'll settle the solar system, and come to feel that that is enough. I doubt it, because there should be systems where life cannot evolve but that we could still settle. But, even if we do still want to establish extrasolar colonies, not only does the technology have to get to the point where it's possible, the desire to establish colonies has to exceed the cost of doing so. We probably could have gone to Mars decades ago, but the public's desire to do so never exceeded the cost of going there.

And going back to your original point, yes, you might wonder why bring humans at all? Maybe you settle humans and other Earth lifeforms on an extrasolar planet simply because humans want to settle another world. That's reason enough, and no better or worse than any other reason. All drives and desires are ultimately arbitrary.

Edited November 25, 2014 by |Velocity|

[NERVAfan]

Tell will tell of course, but I think that sitting here today, in the midst of a continuing revolution in computer technology, in the midst of us seeing AI get ever stronger and stronger, you cannot reasonably believe that it is unlikely for us to ever achieve a truly intelligent machine.

Sure I can. Electronic computers are not brains and do not function in the same way at all (and I am skeptical that quantum computers will ever be very practical due to temperature/decoherence issues). Maybe you could simulate a human brain... but I don't think that will be very workable at the level of fidelity needed (largely because I don't expect the current "revolution in computer technology" to last much longer).

Technologies tend to start slow, explode, then slow down/plateau again.

The late 19th/early 20th century saw an explosion in land and sea transport (trains, ships driven by engines instead of sails, cars, trucks etc.). The early-to-mid 20th century (post WWI to say 1970) saw an explosion in aircraft technology and the development of orbital launch all the way to Apollo.

The late 20th century to now is seeing an explosion in computer technology... but it probably won't last forever, or even all that much longer (we've already gone to multi-core computers due to practical limits on CPU speed, IIRC cooling based...)

I think the next revolution(s) will be

-biotech (arguably it's already happening with things like DNA sequencing getting way cheaper... but the big explosion of consumer applications like we had with computers hasn't happened, and probably won't while people are irrationally terrified of GMO food -- I do expect it to come eventually, but probably out of Asia or maybe Latin America, not US/Europe). But I expect that revolution to be much more limited than some would hope, though maybe more by regulatory/social concerns than actual technological capabilities (I can't imagine any practical way to get longevity drugs by the FDA without decades of testing, for example).

-nuclear fusion and everything that leads to

Nanobots, maybe someday, but ... by definition at that scale (a couple of nanometers is the scale of large molecules like proteins - atoms and chemical bonds are on the order of 0.1 nm) it's more chemistry than machinery, and so will probably have many of the same limitations as life (needs a medium such as water and specific substrates/"food"/raw materials, narrow temperature range, etc.) If we ever get serious nanobots I expect them to have those limitations, probably be bigger than "classic" nanomachines (more virus or even small bacteria sized), and be derived more from advanced biotech than classic "mechanical" technology.

A more reasonable question is whether creating such a machine is possible and practical with silicon microfabrication technology or its derivatives. If not, they would likely take a much longer time to be realized, or at least, miniaturized.

Well, yeah. Humans prove that intelligence is possible, sure. But for the interstellar question, what's relevant is if we can have intelligence without the lifespan and life support requirements of humans -- "biological computers" or fragile quantum computers may not qualify.

But as we're trying to predict the future here in this thread, we should consider the ALL technologies that are most likely to be developed, not just one. You should NOT try to predict the future by only projecting forward progress in one field of technology, which is what the "generation ship" assumes- mostly it represents only progress in space utilization, really.

Not really - I assume such a ship would use advanced biotech to create a new ecology that's far more energy-efficient and efficient at supporting humans. But you could build a (larger) generation ship without that.

Instead, when you look at the total SUM of technologies that are most likely to be developed, the idea of generation ships becomes silly and obsolete, because the VASTLY more efficient "seed ship" becomes possible with advanced machine intelligence and genetics/biology. In fact, the seed ship not only becomes possible it appears to become somewhat practical too, assuming we spread out into space and utilize the vast resources out there in the asteroids.

The thing is that seed ships need a major fundamental breakthrough (not just AI, but AI "human" enough to raise functioning human beings). Every step needed to build a generation ship could be spelled out right now - all we lack is money and will. If we'd done Project Orion in the 60s, we probably could have built a generation ship in the 80s or 90s.

(I think the "hardest" problem is ironically super-long-lasting on-board power/lighting to feed the photosynthesis.)

EDIT: And again, I'm skeptical of getting a seedship capable of reliably* surviving interstellar travel in a package too small to support a worldship (especially given biotech to "engineer down" the size of the ecosystem needed to support a couple of hundred humans).

*because if the AI is "smart" enough to raise humans, it's a person, and therefore its life is valuable too, so it shouldn't be a matter of sending tons and hoping some get through.

Edited November 26, 2014 by NERVAfan

[magnemoe]

We could not have sent a generation ship today even if we had an engine and dirt cheap access to space.

Lots of problem first keep the ecosystem running for 500-2000 years, second keep the ship system running for this time.

Yes you must have repair facilities up to the level of replicate any part of the ship, this will be an one way colonization mission anyway so you need this functionality.

This will increase the weight of the structure a lot at least unless you had good nanotech.

Regarding dust in interstellar space we don't know much of how common it is, makes sense that the solar system has magnitudes more than interstellar space.

Using the stage you used to accelerate as shielding from you stopped accelerating until you release it and start breaking would help a lot.

--

I doubt systems without life would be interesting enough targets for even probes.

[cantab]

The ship would most likely have a general purpose workshop, and the design made so that as far as possible all components could have replacements built in such a workshop - including the shop's own tools. Compared to the scale of the whole ship it would be a fairly small impact.

[-Velocity-]

Sure I can. Electronic computers are not brains and do not function in the same way at all (and I am skeptical that quantum computers will ever be very practical due to temperature/decoherence issues). Maybe you could simulate a human brain... but I don't think that will be very workable at the level of fidelity needed (largely because I don't expect the current "revolution in computer technology" to last much longer).

It's really hard to know how far we can take our current computer technology. We keep running into "barriers" that we find ways around. I am doing my dissertation on silicon MEMS and working on some carbon nanotube field emission stuff on the "side", so I have some background knowledge- but the field is so huge and I don't work on silicon transistors. But I have heard of some new radical ideas that are currently being explored to help keep extending Moore's law even further.

However, eventually, given the fact (unless you believe in souls, which if you do, that's fine, I won't argue) that we have seven billion working examples of intelligent sapient minds, I just find it almost inevitable that we will discover a way to synthetically create one. The first machine intelligence will not be small, but we should be able to eventually make it so.

What IS less certain clear to me is whether or not we can make one that can survive a 1000+ year journey. But maybe we can make one that can last long enough AND still be smart enough to do the job of a seed ship MI.

Nanobots, maybe someday, but ... by definition at that scale (a couple of nanometers is the scale of large molecules like proteins - atoms and chemical bonds are on the order of 0.1 nm) it's more chemistry than machinery, and so will probably have many of the same limitations as life (needs a medium such as water and specific substrates/"food"/raw materials, narrow temperature range, etc.) If we ever get serious nanobots I expect them to have those limitations, probably be bigger than "classic" nanomachines (more virus or even small bacteria sized), and be derived more from advanced biotech than classic "mechanical" technology.

Yes, that's what my beliefs are, and as I said, I actually study and work in the field of MEMS. There's a lot of folks that still believe that MEMS will scale something like transistors do with Moore's law as time goes on. But that's simply not the case. Solely mechanical devices simply no longer work when you get to the nano and sometimes even micro scales. For example, stuff likes to stick to each other due to van der Waals forces and the Casimir effect. Moving parts don't work nearly so well as they do on the macro scale. You might find that you drop your iPhone and when you pick it back up, it no longer is able to recognize which side is up. You've pulled the MEMS accelerometer into contact with its mechanical stop, but with enough force that so much of it made contact with the mechanical stop that the van der Waals forces pulled it into permanent contact. Stuff like gears- FORGET about it. Spinning parts- forget about that too. MEMS gyroscopes are made by making a mass oscillate back and forth on springs and measuring lateral displacements. (I saw some research on electrostatically levitating a mass and spinning it to make a gyro, but I think the effort was ultimately met with failure.) And it only gets worse the smaller you go, as many of the machines we use on the macro scale utilize physical relations that scale with volume (radius^3), while there are limits to performance or opposing forces we can ignore up at the macroscopic that become dominant in the microscopic because they scale with surface area (radius^2) or just radius.

ANYWAY, to make a long story short, yes, nanotech will almost certainly be largely chemical in nature. The "classic" nanomachine- picture the ludicrous image of a propeller-driven, multi-armed, metallic-looking robotic submarine drilling away at arterial plaque while a red blood cell looms large in the background- is a fantasy. Real nanomachines will resemble life forms, or even BE life forms. In fact, as I've said before on these forums, I believe that life and the living cell itself is naturally emergent nanotechnology. We are nothing but a vast colony of nanomachines working together.

Well, yeah. Humans prove that intelligence is possible, sure. But for the interstellar question, what's relevant is if we can have intelligence without the lifespan and life support requirements of humans -- "biological computers" or fragile quantum computers may not qualify.

Yes, but I still feel it's far more likely that we'll master an intelligence smart enough and long-lasting enough to pilot a seed ship than we'll ever build generation ships. And the level of machine intelligence needed for a seed ship may not be much higher than what we've already got today- of course, our current computers are not anywhere as radiation hardened and reliable enough yet though.

Edited November 27, 2014 by |Velocity|

[NERVAfan]

Yes, but I still feel it's far more likely that we'll master an intelligence smart enough and long-lasting enough to pilot a seed ship than we'll ever build generation ships.

Well, I guess the reason I disagree is that I think that if technological civilization lasts another 100 years or so, we'll have significant space colonization, and once you have that I think generation ships are pretty much inevitable eventually - they strike me as a natural extension of space colonies.

And the level of machine intelligence needed for a seed ship may not be much higher than what we've already got today- of course, our current computers are not anywhere as radiation hardened and reliable enough yet though.

Something that would be able to raise a vaguely functional human being would be not only enormously more capable than anything we have today, but a fundamentally different kind of capability.

Generation ships don't require any fundamental breakthroughs; seed ships do.

[78stonewobble]

Well, I guess the reason I disagree is that I think that if technological civilization lasts another 100 years or so, we'll have significant space colonization, and once you have that I think generation ships are pretty much inevitable eventually - they strike me as a natural extension of space colonies.

Something that would be able to raise a vaguely functional human being would be not only enormously more capable than anything we have today, but a fundamentally different kind of capability.

Generation ships don't require any fundamental breakthroughs; seed ships do.

Not to mention the ethical dilemma of keeping a sapient being (if we're talking AI) in more or less isolation for hundreds potentially thousands of years. AFAIK, isolation can be viewed as a form of torture.

Then again it might not have to be a true artificial intelligence, just complex enough.