Kahalachan's Desert Island Hypothesis

[Kahalachan]

I have not heard of this analogy before, but if so I would love to hear who had a similar idea as me.

To get my point across, I shall first paint a picture of you living on a desert island with a pair of binoculars.

This island can sustain you comfortably but is indeed very small and isolated. It has a cave for shelter, but this cave has no useful minerals to be mined and is virtually barren of any other tools aside from sharp rocks. You have one tree, and only enough room for this one tree, which happens to be the sole source of all your fruit and vegetation. You may be able to catch a few crabs and shellfish to suppliment your food as you can use a rock from the cave to crack open shells.

Based on all your observable calculations, if you were to chop down the tree to make a boat, you wouldn't have enough space on the boat to fill it with the amount of food you would need to travel. You also would not have enough food to sustain you if you were to create a boat. By all you are capable of observing, you know that there will still be only a vast ocean by the time the food runs out, which would cause you to starve to death if you leave this island. You are stuck on this island and cannot find another island or continent.


Drawing on this analogy, the Desert Island hypothesis states that space travel will forever be limitted due to the lack of resources nature has bequeathed upon us here on earth, so far isolated from all other possible resources that can be used for space travel and survival.

Also to add a bit of social irony, if aliens were to ever visit the earth, our only hope of further exploration would be in the hopes that they were hospitable and allocated resources to us in our desire to explore. Just as you on your desert island will be completely subject to the hospitability of someone else who has the resources to go to your island. The social irony comes from an apprehension Steven Hawking had about aliens visiting us; because our history is wrought with civilizations having greater resources supplanting those with fewer resources.


Do you agree or disagree with the Desert Island Hypothesis?

[Jeff Corey]

Huh?

[Apathia]

I saw that movie about the Fed Ex guy! It sucked.

[Uncle Vanya]

Also to add a bit of social irony, if aliens were to ever visit the earth, our only hope of further exploration would be in the hopes that they were hospitable and allocated resources to us in our desire to explore. Just as you on your desert island will be completely subject to the hospitability of someone else who has the resources to go to your island. The social irony comes from an apprehension Steven Hawking had about aliens visiting us; because our history is wrought with civilizations having greater resources supplanting those with fewer resources.


Well, then you're making the assumption that the aliens somehow had more resources to begin with, or that they could somehow carry more resources with them in their interstellar travels. If aliens can visit us, why can't we visit them? I mean, we're aliens to them.

Perhaps many many years from now we'll be visiting distant inhabited planets and kidnapping random alien farmers and probing various bodily orifices they have. Then one of our ships will crash in the middle of a desert on this planet, and the alien government will try to cover it up.

[ail]

Essentially do not agree with your hypothesis.

I would use the binocular lenses and the suns rays to create a signal fire with some brances at regular intervals. Eventually I should get noticed.

We humans are a resourceful lot. I believe, given time and if we can get along, a way to discover whether we are alone in the universe can be achieved.;)

[Wowbagger]

I would use the binocular lenses and the suns rays to create a signal fire with some brances at regular intervals. Eventually I should get noticed.
We don't know what the equivalent would be, for getting noticed by aliens.

I think the analogy has merrit. There could be upper limits placed on any planetary system, that disallows interstellar travel to take place.

The largest planet might not be convertable to enough fuel to do it.

[athon]

Do you agree or disagree with the Desert Island Hypothesis?

The analogy might have use, depending on the situation.

Unfortunately the variation is in what we consider a 'resource'. Currently, according to present technology, what we have access to cannot provide us with the means to reach other inhabitable systems. That does not mean that we will not a) have access to other resources in the future (we could use those of other local planets and bodies which we mine and/or terraform), b) make better use of the resources we currently have through improved measures.

This is where the analogy falls apart. A better analogy would be to include more resources that he may or may not ever learn to use effeciently to his advantage. For instance, there is driftwood and bits of flotsam that has washed on the shore. There is no way he knows of at present which could use it to provide a solution to getting off the island, yet that does not remove the possibility of it.

Athon

[sinclairmcevoy]

Huh?

Ditto squared.

[Kahalachan]

The analogy might have use, depending on the situation.

Unfortunately the variation is in what we consider a 'resource'. Currently, according to present technology, what we have access to cannot provide us with the means to reach other inhabitable systems. That does not mean that we will not a) have access to other resources in the future (we could use those of other local planets and bodies which we mine and/or terraform), b) make better use of the resources we currently have through improved measures.

This is where the analogy falls apart. A better analogy would be to include more resources that he may or may not ever learn to use effeciently to his advantage. For instance, there is driftwood and bits of flotsam that has washed on the shore. There is no way he knows of at present which could use it to provide a solution to getting off the island, yet that does not remove the possibility of it.

Athon


Well I suppose the best definition of resource would be usable energy and matter that can be manipulated. So the sun + all the energy and objects availabe on the island/earth.

I agree with the analogy of objects drifting to shore being useful.

If we could imagine someone being omniscient but floating around in space, it seems like they could not do much. If they lacked any ability to manipulate anything, someone omniscient still couldn't do much to escape. It would appear that there is a necessary bare minimum of matter or energy to manipulate for them to be able to put it to use in order to move from their current position.

In short, I am asking, not considering things that could drift to our planet, will there be a limit as to how far we can travel through space due to pure physical limitations and a lack of resources? Similarly to how our island straggler can only go so far in the ocean before they must return.

Edit: The Desert Island Hypothesis is one that ignores the possibility of any help coming from the heavens; as in alien assistance or useful objects drifting through space and reaching our maximum travel radius.

[Kahalachan]

Well, then you're making the assumption that the aliens somehow had more resources to begin with, or that they could somehow carry more resources with them in their interstellar travels. If aliens can visit us, why can't we visit them? I mean, we're aliens to them.

Perhaps many many years from now we'll be visiting distant inhabited planets and kidnapping random alien farmers and probing various bodily orifices they have. Then one of our ships will crash in the middle of a desert on this planet, and the alien government will try to cover it up.

I don't mean to promote woo here. I do not think aliens visitted us. I am wondering if there are limitations that we can never break that prevents us from exploring space up to a certain distance.

Similarly to being stuck on an island and not being able to travel past a certain point, regardless of how smart or knowledgable the castaway is, simply because there is not enough for our castaway to work with.

[Floyt]

If we ever achieve anything even distantly approaching matter to equivalent energy conversion, there will be plenty of energy to shoot any number of colonists and the kitchen sink to the stars. Whereas your castaway could probably not hope to smelt a speedboat from the island's sand even if he thinks really hard on how to do it :)
So no, I believe we are in a more hopeful position.

[El Greco]

In short, I am asking, not considering things that could drift to our planet, will there be a limit as to how far we can travel through space due to pure physical limitations and a lack of resources?

Of course there will be a limit. With what we currently know about the size of the universe, there is either a limit or, by the time those limits have been raised, such questions are meaningless because our intellect will have progressed too far for our current understanding.

But the original question, whether we have limited resources to reach an alien civilization, is one that nobody can answer with any degree of certainty, so I'd say is meaningless.

[drkitten]

Well, then you're making the assumption that the aliens somehow had more resources to begin with, or that they could somehow carry more resources with them in their interstellar travels. If aliens can visit us, why can't we visit them? I mean, we're aliens to them.

Oh, that one's easy. The aliens came from a more "hospitable" environment.

For example, we have only one habitable planet in our solar system; the aliens may have come from a system with several habitable planets, enabling them to develop cheap, short-distance space flight as a testbed for the development of longer-duration flights.

Similarly, the aliens may have come from a part of the galaxy where the stars are closer together and so are able to harvest resources more effectively from the parsec-radius sphere around their home star than we are. For that matter, the aliens may simply have come from a world where unobtainium is common enough that they can use a more efficient method of travel (put some flax on this island, so that I can build effective sails for my boat, and I have just doubled the distance I can safely travel).

-----

The problem that I see with this analogy is that it a) ignores the possibility of developing or discovering new resources, and b) ignores the possibility of accumulating (renewable) resources over time. I can build an arbitrarily large boat given enough time by simply harvesting branches but leaving the tree itself alive, or by planting new trees as I (plan to ) chop down the old ones.

[Kahalachan]

The problem that I see with this analogy is that it a) ignores the possibility of developing or discovering new resources, and b) ignores the possibility of accumulating (renewable) resources over time. I can build an arbitrarily large boat given enough time by simply harvesting branches but leaving the tree itself alive, or by planting new trees as I (plan to ) chop down the old ones.

Well for b, there is only enough room on the island for one tree. I am assuming that when we reach a maximum stage of being able to renew our own resources, that still won't be enough.

If our island inhabitant could die of starvation but not old age, they would have all the time in the world to study the island but even if they completely understood the island, they might not be able to accumulate the actual materials needed to travel that far.

Perhaps water can be harnessed in some way for him due to its most basic properties that water possesses. So the vacuum of space or the very fabric of space-time can be used in some way should we not have enough resources within our vicinity. The very basic properties of the universe could be useful in some way.

So your objection with point a, is understood and we may have to hope for developing or discovering some kind of resource.

But the main point of this Desert Island scenario is that we may forever be limitted.

Is the maximum radius in which we can travel by purely optimizing all energy and matter at our disposal, less than the minimum radius needed to obtain more energy and matter?

Edit: This hypothesis doesn't seem answerable today of course, but perhaps we will reach a maximum of what we can learn and use on our planet and be stuck.

I don't mean to treat the earth or solar system as a closed system isolated from the universe. But I certainly am conjecturing a possible limitation on our end as to what can actually be used by humans for space travel.

[blutoski]

Drawing on this analogy, the Desert Island hypothesis states that space travel will forever be limitted due to the lack of resources nature has bequeathed upon us here on earth, so far isolated from all other possible resources that can be used for space travel and survival.

Why employ an analogy?

Just calculate how much energy you need to travel to a neighbouring star, and see if it's obtainable.

[Fnord]

"Kahalachan's Desert Island Hypothesis" is interesting, in an academic way, but have you any experimental data to back it up?

You know, something that could turn it into a theory for peer-group review.

[Ladewig]

Every decade we can see more and more details of nearby stars. I think the appropriate analogy would be the tree growing higher and higher, thus allowing the castaway to see farther and farther. He might eventually see another island and therefore communicate with distant islanders.

[Kahalachan]

"Kahalachan's Desert Island Hypothesis" is interesting, in an academic way, but have you any experimental data to back it up?

You know, something that could turn it into a theory for peer-group review.

No experimental data.

It seems like large spaceships would have to be assembled in space. Because for some huge ship with a usable biosphere would require too much energy to take off.

So we would have to pool all usable materials for this, assuming this to be a collaborative project of all nations.

We would have to see how much we can stockpile on this giant ship for propulsion through space.

It seems like we should make something so big as to be able to have some gravity for biological organisms that are used to some gravity, in order for the biosphere to be successful.

Space would literally be like the new ocean where a huge ship would have to be docked and then cast off from outside the earth.

[PixyMisa]

It seems like large spaceships would have to be assembled in space. Because for some huge ship with a usable biosphere would require too much energy to take off.
And for practical engineering reasons as well.

So we would have to pool all usable materials for this, assuming this to be a collaborative project of all nations.Uh, no.

NASA currently receives 0.5% of the US federal budget, about 0.1% of US GDP, about 0.02% of world GDP. If we really wanted to build big spaceships - if there was value to it - we have the resources. We have heaps of resources.

Let's say our ship is ten times the size the aircraft carrier USS Nimitz, say about a million tons, excluding fuel and payload. And let's say that it's mostly made of steel. World steel production is in excess of a billion tons each year.

It's not resources per se, it's the cost of getting stuff into orbit. So what you do if this is a long-term project is you launch a million tons of mining, refining, and engineering equipment, and then go get stuff where the gravity well isn't so deep. The asteroid belt is a good candidate.

We would have to see how much we can stockpile on this giant ship for propulsion through space.Steel is not in short supply. I'd presume we'd be using nuclear power and ion drive; that's not in short supply. Food production is in the billions of tons per year. Water is everywhere.

It seems like we should make something so big as to be able to have some gravity for biological organisms that are used to some gravity, in order for the biosphere to be successful.Ugh. No, not even remotely. Your ship would have to be the size of Mars. You just spin the ship, or parts of it.

Space would literally be like the new ocean where a huge ship would have to be docked and then cast off from outside the earth.Um, yeah, I guess.

[PixyMisa]

Is the maximum radius in which we can travel by purely optimizing all energy and matter at our disposal, less than the minimum radius needed to obtain more energy and matter?
This is the key question. And the answer is no, not by a few orders of magnitude. That's without even getting working fusion power, and if we do that, we get another few orders of magnitude. After all, we can get to Jupiter easily enough already, and Jupiter is basically two septillion tons of fusion fuel.

[BenBurch]

I think the real answer is that starships are hellishly expensive, ridiculously slow, and of uncertain destiny once launched. In a galaxy our size I doubt there are more than two starships in use at any time.

The speed limit in most parts of the galaxy is around 1% of the speed of light; Erosion from collision with space dust and gasses is the limiting factor. And you'd better never hit a rock. And the rocket equation will tell you of the immense amount of energy and reaction mass required to achieve even that. Add to this the problems of running a civilization in a can for centuries, and you have a recipe for disaster. Likely this is only tried by sentient species who have found home to be so inhospitable that survival is in question.

So, that is my answer to the Fermi Paradox; They aren't here because they never were able to get anywhere.

[PixyMisa]

I think the real answer is that starships are hellishly expensive, ridiculously slow, and of uncertain destiny once launched.
Expensive compared to what, though?

In a galaxy our size I doubt there are more than two starships in use at any time.
There are a lot of assumptions built in to that statement. Let's say that once Earth gets into the interstellar transport business, it launches one ship every twenty years. Let's say that we're limited to 1% of the speed of light, and the average trip is ten light years. That makes for fifty ships active at once, based on the output of one small planet.

The speed limit in most parts of the galaxy is around 1% of the speed of light; Erosion from collision with space dust and gasses is the limiting factor. And you'd better never hit a rock.
Yep. Fortunately, there aren't a lot of rocks in interstellar space.

And the rocket equation will tell you of the immense amount of energy and reaction mass required to achieve even that.
You'd be using ion drives or similar, so you'd have a specific impulse ten to fifty times that of chemical rockets. And you could beam power from the solar system using large laser arrays for the acceleration, and keep your fuel for the deceleration.

Add to this the problems of running a civilization in a can for centuries, and you have a recipe for disaster.
Not my problem, I'm in engineering! ;)

Likely this is only tried by sentient species who have found home to be so inhospitable that survival is in question.
You have an even more cynical view of sentient species than I do.

So, that is my answer to the Fermi Paradox; They aren't here because they never were able to get anywhere.
Eh. The galaxy's a big place.

[drkitten]

This is the key question. And the answer is no, not by a few orders of magnitude. That's without even getting working fusion power, and if we do that, we get another few orders of magnitude. After all, we can get to Jupiter easily enough already, and Jupiter is basically two septillion tons of fusion fuel.

And the asteroid belt, which is even closer, is how many billion tons of structural materials?

As far as I can see, the only thing we would possibly be missing is a few billion tons of carbon, which we could harvest if necessary from the atmosphere of Venus.

[geni]

And the asteroid belt, which is even closer, is how many billion tons of structural materials?

Total mass is 3.0 ± 0.2 ×1021 kg however about a third of this is Ceres.


As far as I can see, the only thing we would possibly be missing is a few billion tons of carbon, which we could harvest if necessary from the atmosphere of Venus.

C-type asteroids are hardly uncommon.

However the astroid belt takes some getting to. Probably best to start with NEAs.

[BenBurch]

Expensive compared to what, though?

Compared to the total production of our planet, and I assume any other. We are talking about an immense thing that absolutely has to be manufactured in space. You need to be mining asteroid material and refining it and smelting it and casting it in space. Reaction mass also has to come from comets or asteroids or a gas giant planet and some scooping technology we can only imagine. Specialized goods come up a gravity well.

If fission-powered it would take up a sizable percentage of the planet's available nuclear fuel.

In other words, bloody expensive no matter who does it just from a basis of raw materials.

There are a lot of assumptions built in to that statement. Let's say that once Earth gets into the interstellar transport business, it launches one ship every twenty years. Let's say that we're limited to 1% of the speed of light, and the average trip is ten light years. That makes for fifty ships active at once, based on the output of one small planet.

I believe we could maybe build a total of two or three before not being able to maintain the level of technology required due to resource exhaustion and the general lack of any regard such an all out effort would require for the sustainability of the parent civilization. Hence I believe that nobody would do this unless doomed.

Yep. Fortunately, there aren't a lot of rocks in interstellar space.

Only takes one.

You'd be using ion drives or similar, so you'd have a specific impulse ten to fifty times that of chemical rockets. And you could beam power from the solar system using large laser arrays for the acceleration, and keep your fuel for the deceleration.

Assumes technologies not in evidence. Right now the best we can design is that you build a simply immense nuclear power station that is only used when the main propulsion is online, which might be the first 100 years or so and the last 100 years or so.

Not my problem, I'm in engineering! ;)

Well, this is an engineering concern. The people are not just payload, but have to actually rebuild the ship as needed underway. To do this you need an intact social infrastructure.


You have an even more cynical view of sentient species than I do.


No, I simply believe that you underestimate of the size of the job and its uncertainty of success is rather large. Making a starship would for us be a civilization-ending quest. And for any imaginable single planet would at least be as expensive as building a fleet of space habitats of similar scale, which would usually be judged a safer species survival alternative unless you were having problems that required you to leave your region of space.

Eh. The galaxy's a big place.

That's the problem.

[geni]

Well, this is an engineering concern. The people are not just payload, but have to actually rebuild the ship as needed underway. To do this you need an intact social infrastructure.


Too much like hard work. Just freeze everybody. If you need an awake crew freeze everybody with slightly more people than you need and wake new people up as the old ones die.

[PixyMisa]

Compared to the total production of our planet, and I assume any other. We are talking about an immense thing that absolutely has to be manufactured in space. You need to be mining asteroid material and refining it and smelting it and casting it in space. Reaction mass also has to come from comets or asteroids or a gas giant planet and some scooping technology we can only imagine. Specialized goods come up a gravity well.
The specifics are all true, but it's still not significant. That million tons of steel for the space industry systems are 0.1% of Earth's present annual production. Since it would take maybe 20 years to produce and launch all that stuff, we're looking at 0.005% of our annual production.

Now, launching the thing into space, that would take significant effort. Using Saturn V launchers, we'd need daily launches for the entire 20 years to get a million tons into LEO.

If fission-powered it would take up a sizable percentage of the planet's available nuclear fuel.
Hmm. I'll have to do the calculations on that. And you can resolve that by travelling more slowly. Fuel requirements rise as the square of the speed, so at 0.1% of the speed of light (assuming your calculation is accurate) we'd only need 1% of available fission fuel. And we can power the acceleration externally and use a staged deceleration, letting the bulk of the ship continue on as an interstellar probe while the colony pod slows down.

I believe we could maybe build a total of two or three before not being able to maintain the level of technology required due to resource exhaustion and the general lack of any regard such an all out effort would require for the sustainability of the parent civilization. Hence I believe that nobody would do this unless doomed.
Well, you're wrong about everything except possibly for the fuel issue, and that can be solved anyway.

Only takes one.
Not meaningful. If your likelihood of hitting one is 0.1%, it's 0.1%.

Assumes technologies not in evidence. Right now the best we can design is that you build a simply immense nuclear power station that is only used when the main propulsion is online, which might be the first 100 years or so and the last 100 years or so.
Nope. It just needs a bunch of powersats and a bunch of lasers. Both can be build using present-day technology.

Well, this is an engineering concern. The people are not just payload, but have to actually rebuild the ship as needed underway. To do this you need an intact social infrastructure.
It might be an engineering concern, but it's a problem of psychology and sociology. Or biology. Just keep 99% of the people in some form of hibernation. (Now, that does assume technologies not in evidence.)

No, I simply believe that you underestimate of the size of the job and its uncertainty of success is rather large. Making a starship would for us be a civilization-ending quest. And for any imaginable single planet would at least be as expensive as building a fleet of space habitats of similar scale, which would usually be judged a safer species survival alternative unless you were having problems that required you to leave your region of space.
You haven't shown this, though.

That's the problem.
No. There are on the order of 200 billion stars, and perhaps a trillion planets, in our galaxy. It might take 100 million years for a space-faring race to spread out across the entire galaxy, but it's eminently achievable. Maybe no-one's bothered... After the first billion planets, do you really need to expand further? But it's not a satisfying answer to the Fermi Paradox.

[PixyMisa]

Okay, there are between 5 and 40 million tons of known uranium reserves, depending on the economics of extraction. And another 4 billion tons in the ocean, which can be extracted economically.

Let's go with a million-ton ship, and a laser-array boost for the outward trip. To decelerate a million tons from 1% of the speed of light takes, uh, 9e21 joules if I haven't screwed this up. And according to this page (http://en.wikipedia.org/wiki/Orders_of_magnitude_%28energy%29) that is roughly ten times world uranium resources without using fast (breeder) reactors, or 5% of world resources with fast reactors. So you were definitely in the ballpark there.

However, that appears to be based on the low-end estimate of reserves (http://www.iaea.org/NewsCenter/News/2006/uranium_resources.html), ores extractable for less than $130 per kg. Total available uranium is about a thousand times that. So we're looking at 0.005% of world uranium resources per launch if we want to bring the entire million-ton craft to a halt.

On the other hand, we'd need to bring along 250,000 tons of uranium oxide. On the other other hand, it's easy enough to dispose of. And if we settled for 0.1% of the speed of light, and a 100,000 ton colony pod, we'd bring it down to as little as 250 tons.

So, easy-peasy.

[BenBurch]

A million tons does not begin to cover it.

You need enough people onboard to have a large enough society to train first rate professionals in every imaginable field. Doctors, Psychologists, Biologists, Botanists, Engineers, Physicists, and even Kindergarten teachers. Heck you'll even need law enforcement and lawyers. And a jail. This is why I said that society is an engineering concern. You need to have at least a large city, and enough farm to feed them, and enough materials processing to clothe them for what will be a really, really long trip; 51 Pegasi is 50 ly away, or 5000 years at 1 percent of the speed of light.

And they not only need nuclear energy to run the drives, but also a whole civilization for longer than any terrestrial civilization has ever endured.

Also, don't assume you CAN colonize the place once you get there, either. You better give them either the option of taking on more reaction mass and moving on, or of becoming residents of a self-sustaining O'neil-like colony.

They would feel right at home doing that, but their forbearers could have done that here much more cheaply!

And are you certain that a power beam would be switched on at the right time to power deceleration? We might be extinct here by the time they need that power. Or we might have forgotten all about them. Or we might know they will need it, and, well, too bad its nobody's job to make that happen, huh?

[PixyMisa]

A million tons does not begin to cover it.

You need enough people onboard to have a large enough society to train first rate professionals in every imaginable field. Doctors, Psychologists, Biologists, Botanists, Engineers, Physicists, and even Kindergarten teachers. Heck you'll even need law enforcement and lawyers. And a jail. This is why I said that society is an engineering concern. You need to have at least a large city, and enough farm to feed them, and enough materials processing to clothe them for what will be a really, really long trip; 51 Pegasi is 50 ly away, or 5000 years at 1 percent of the speed of light.
You will be in communication with Earth the entire time. You'd still need hundreds, perhaps thousands of people.

The complement of the 100,000 ton USS Nimitz is 5,680. The Nimitz isn't self-contained the way a spaceship would be, but it was built in 1968.

And they not only need nuclear energy to run the drives, but also a whole civilization for longer than any terrestrial civilization has ever endured.That depends on whether you're willing to rely on the laser array for energy. If so, you only need the nuclear energy for deceleration.

Also, don't assume you CAN colonize the place once you get there, either. You better give them either the option of taking on more reaction mass and moving on, or of becoming residents of a self-sustaining O'neil-like colony.You'd send out probes first. A probe would only mass a few tons - possibly much less (http://en.wikipedia.org/wiki/Starwisp) - and it would be unmanned, so it could travel at far higher speeds.

They would feel right at home doing that, but their forbearers could have done that here much more cheaply!Their forebears would not have the resources of an entire new solar system at their disposal.

And are you certain that a power beam would be switched on at the right time to power deceleration? We might be extinct here by the time they need that power. Or we might have forgotten all about them. Or we might know they will need it, and, well, too bad its nobody's job to make that happen, huh?Deceleration has to be powered either by internal sources, or by an array pre-constructed by robotic probes at the destination. Either way, you don't have to rely on the power beam for that part.

One note - in my previous post I accidently assumed 100% efficiency for both power generation and the drive system. This seems unlikely, so multiply fuel requirements by a factor of, say, 3.

[BenBurch]

Communication with Earth is worthless if you have a patient in the ICU. Or if your main water condenser fails. Or if the AI in the ship's computer goes insane. You need local experts. Not third-raters with training. And to have real experts, people who shine at what they do, the three-sigma-smart ones, you have to have a population. And one that encourages the sciences. Because there is simply no way even in principle we can understand the failure modes of a 5000 year old machine or a civilization in isolation.

It's a huge risk with people's lives, people you will never meet because they are born in transit.

[PixyMisa]

Communication with Earth is worthless if you have a patient in the ICU.
If the doctors on board can't cure them, the patient dies.

Or if your main water condenser fails.You have forty more, and parts for forty more beyond that, and machine shops to produce parts, and machine shops to produce machine shops.

Or if the AI in the ship's computer goes insane.Well, then you're just screwed.

You need local experts.Yes. And you have them. Thousands of them. That's why we're launching a million-ton ship.

It's a huge risk with people's lives, people you will never meet because they are born in transit.That's entirely irrelevant to the question of whether it can be done or not.

[geni]

A million tons does not begin to cover it.

You need enough people onboard to have a large enough society to train first rate professionals in every imaginable field. Doctors, Psychologists, Biologists, Botanists, Engineers, Physicists, and even Kindergarten teachers. Heck you'll even need law enforcement and lawyers. And a jail. This is why I said that society is an engineering concern. You need to have at least a large city, and enough farm to feed them, and enough materials processing to clothe them for what will be a really, really long trip; 51 Pegasi is 50 ly away, or 5000 years at 1 percent of the speed of light.

Note romotely. You active and awake population should never pass 100.


Also, don't assume you CAN colonize the place once you get there, either. You better give them either the option of taking on more reaction mass and moving on, or of becoming residents of a self-sustaining O'neil-like colony.


By the time we are looking at interstella travel I suspect astroids will be viewed as good colonisation targets.

Probably the first attempt would be made by one of those colonies fitting engins too their astroid and freezeing most of the population (we will assume that artificial womb technology is posible by this stage or colonisation is slightly pointless).

[Roboramma]

Why even send people at all? Just send robots capable of doing what the people would do themselves - build and reproduce.
Then, after 500 years, or 1000 or whatever, they start making productive use of their new solar system in ways that are valuable for us here. For instance, computing power, or sending beams of laser energy our way to be used here, or sending an interstellar ship with whatever valuables (anti-matter? Uranium? whatever) they've produced in that time back to us.

This eliminates the need for "experts", life support, etc. to be carried along on the journey. The cost is thus hugely minimized.

Of course, I'm assuming "technologies not in evidence". But anyone who thinks that technology will not advance in 50-100, or 500 years is kidding themselves.

[PixyMisa]

Why even send people at all? Just send robots capable of doing what the people would do themselves - build and reproduce.
Then, after 500 years, or 1000 or whatever, they start making productive use of their new solar system in ways that are valuable for us here. For instance, computing power, or sending beams of laser energy our way to be used here, or sending an interstellar ship with whatever valuables (anti-matter? Uranium? whatever) they've produced in that time back to us.
Or at least, send them ahead to scout things out and build up infrastructure beforehand.

[geni]

Or at least, send them ahead to scout things out and build up infrastructure beforehand.

Not really required. Human genetic samples have a very low mass. Trivial to send enough to start a civilisation with them.

[BenBurch]

This eliminates the need for "experts", life support, etc. to be carried along on the journey. The cost is thus hugely minimized.

Of course, I'm assuming "technologies not in evidence". But anyone who thinks that technology will not advance in 50-100, or 500 years is kidding themselves.

You then need an AI that is an expert in everything that it needs to be to complete the mission.

[BenBurch]

Note romotely. You active and awake population should never pass 100.

Assumes technologies not in evidence. We have no idea whatsoever how to suspend people for more than a few minutes.

And 100 is not a stable breeding population if that is what you meant. I'm fairly sure there is enough social diversity in a crew of 100 to keep them from seriously loathing one another after a few years.

[BenBurch]

Not really required. Human genetic samples have a very low mass. Trivial to send enough to start a civilisation with them.

Again, assumes technologies not in evidence. Just because we can imagine a technology does not make it possible.

And I assert that a machine cannot be built that will function in 5000 years. So any such machine has to either be 100% self-maintaining, including the ability to make its own engineering changes to itself, or it requires a crew of humans to do those functions.

And remember, please that 5000 years is 50 ly. Our chances of finding a place that can be colonized within 50 ly appears to be small.

[geni]

Assumes technologies not in evidence. We have no idea whatsoever how to suspend people for more than a few minutes.

We can do hampsters indefinetly. We will work out how to freeze humans at some point.


And 100 is not a stable breeding population if that is what you meant.


Any human population countianing educated women knowlage of contraception and lack of relgious reasons to have large families appears not to produce a stable breeding population. For colonisation to be posible you need the technology for artifical wombs and probably child rearing.


I'm fairly sure there is enough social diversity in a crew of 100 to keep them from seriously loathing one another after a few years.

100 is about the limit of your monkeysphere. You don't really want to exceed that with an awake crew.

[geni]

Again, assumes technologies not in evidence. Just because we can imagine a technology does not make it possible.

I would argue that we are pretty close to being able to turn such material into babies without the need for other humans. Your problem is turning those babies into a functioning society.


And I assert that a machine cannot be built that will function in 5000 years. So any such machine has to either be 100% self-maintaining, including the ability to make its own engineering changes to itself, or it requires a crew of humans to do those functions.


Doesn't have to function. Most of it will be turned off and stored in a hard vacume.


And remember, please that 5000 years is 50 ly. Our chances of finding a place that can be colonized within 50 ly appears to be small.

Only if you insist on planets. For a spacefareing civilisation planets even with space elevators may not be ideal (so hard to control everything).

[BenBurch]

I don't think you can send a cold ship and have that work.

Look up "cold welding" some day. No moving parts will still move after that much time.

And forget about any springs, gaskets, or anything that has to be flexible...

Nor will integrated circuits function after all that time, as these degrade due to cosmic rays and due to materials issues like implanted doping diffusing to the point of uselessness and etc.

And *something* has to be alive in there for it to know when to wake up.

Show me a machine that has run for 100 years without a rebuilding anywhere.

Show me a 100 year old machine disused and unmaintained for most of that which can be made to work today as well as when it was made?

Watchmakers routinely have to repair presentation watches that have not run in decades because NOT running is not a benign state for them even in dry, clean storage.

[PixyMisa]

Again, assumes technologies not in evidence. Just because we can imagine a technology does not make it possible.
Oh, it's possible. Difficult, sure. But there's no new physics involved.

And I assert that a machine cannot be built that will function in 5000 years. So any such machine has to either be 100% self-maintaining, including the ability to make its own engineering changes to itself, or it requires a crew of humans to do those functions.
That's a good point. Pioneer 10 and 11 are still working fine after 35 years, but 5000 is a lot more than 35. On the other hand, lightweight unmanned probes can go faster than 1% of the speed of light.

And remember, please that 5000 years is 50 ly. Our chances of finding a place that can be colonized within 50 ly appears to be small.
Uh, no, not really. There are 249 known star systems (http://www.chara.gsu.edu/RECONS/census.posted.htm) within 10 parsecs of Earth (and 2029 within 25 parsecs). That 249 includes 6 type F, 21 type G, and 44 type K stars, which are the ones you'd probably be looking at as prime colony destinations. And, so far, 10 planets in those systems.

Many of those stars are in binary or multiple systems and may not be suitable, and others might be flare stars or variables. But there's still a lot of candidates.

Tau Ceti is one. It has low metallicity, so it might not have rocky planets, but it seems to have a good sized cometary cloud (http://adsabs.harvard.edu/abs/2004MNRAS.351L..54G), which could provide all the resources needed for a colony.

Epsilon Eridani has at least two planets, but it's a variable star, which would make things harder. You could still have a colony there, but probably not a habitable Earth-like planet.

Epsilon Indi has a binary pair of brown dwarfs orbiting it at a distance of about 1500AU. That could well indicate the presence of other objects.

[Roboramma]

You then need an AI that is an expert in everything that it needs to be to complete the mission.
Of course, but I don't see why that should be a problem.

[BenBurch]

Of course, but I don't see why that should be a problem.

Except we have zero idea how to make that work...

I've been in computing since the early 70s, and AI has been "right around the corner" that whole time.

Sort of like fusion energy.

[Roboramma]

We can do hampsters indefinetly. We will work out how to freeze humans at some point. Or genetically engineer them to be freezable.

But anyway, fertilized eggs are freezable. Bring along enough of them and you don't have to worry about genetic diversity at all.

[PixyMisa]

Expert system type AI works and is in use right now. It's natural language AI that is proving difficult.

[Roboramma]

Except we have zero idea how to make that work...

I've been in computing since the early 70s, and AI has been "right around the corner" that whole time.

Sort of like fusion energy.
But there's no reason to believe it's not possible. In fact, we know that it is possible. We have an example of something that's capable of it - the human brain.

Given that there is nothing magical about the human brain, it's entirely possible for us to design and build a system capable of doing anything that it can do. The fact that it's taking us longer than we thought at first to figure out how doesn't mean that it can't be done. It necessarily can. But I agree that we don't know now how long that will take.

If you're arguing that interstellar travel won't be possible within the next 50 years, okay, maybe that's a good point. But 500? 1000? 10,000?

[BenBurch]

Expert system type AI works and is in use right now. It's natural language AI that is proving difficult.

No, we have some TRICKS we can call AI. I've done expert systems in OPS-5 back in the 80s. What we were doing is not AI. And it still is not.

Neural networks are a shade closer, but we have zero idea how to extend that into a mind.

And a mind is what you need. Not just a self aware machine as smart as a dog - which is a lofty and very useful goal we may never get to - but one actually smarter than we are and more able to consider novel ideas than we are.

And we would also have to assume that this superior machine mind is going to WANT to go on this little journey of ours - and I can see no ethical way to ensure this.

[Roboramma]

And a mind is what you need. Not just a self aware machine as smart as a dog - which is a lofty and very useful goal we may never get to - but one actually smarter than we are and more able to consider novel ideas than we are. Why does it have to be smarter than we are? Why not a bunch of "machine minds" that are no more intelligent than us? If the job could be done by a bunch of human experts it should be able to be done by the same number of machine experts.

Mind you, it might be more efficent with fewer, smarter (or at least more expert) minds, but that's not necessary.

And we would also have to assume that this superior machine mind is going to WANT to go on this little journey of ours - and I can see no ethical way to ensure this. By programming it to want to. The same we we're programmed to want to eat.

[Loss Leader]

Do you agree or disagree with the Desert Island Hypothesis?


Disagree. As any economist can tell you, we are almost always increasing our production. We have more money because we have more things and that has been the case virtually uninterrupted throughout history.

Whatever limitations you see on our resources today will be viewed as silly and quaint by our children's children.

[PixyMisa]

No, we have some TRICKS we can call AI. I've done expert systems in OPS-5 back in the 80s. What we were doing is not AI. And it still is not.
By the definition "AI is whatever we haven't done yet".

Neural networks are a shade closer, but we have zero idea how to extend that into a mind.

And a mind is what you need. Not just a self aware machine as smart as a dog - which is a lofty and very useful goal we may never get to - but one actually smarter than we are and more able to consider novel ideas than we are.
Why? All it needs to do is fix stuff.

[BenBurch]

Why? All it needs to do is fix stuff.

It needs to be able to re-design itself, because there is no way we will know in advance anything about how such a machine will function with age, and deal intelligently with whatever unknowns are there to find. (Do not forget, this is exploration; If we knew what we were going to find there would be no reason to go.)

[Roboramma]

It needs to be able to re-design itself, because there is no way we will know in advance anything about how such a machine will function with age, and deal intelligently with whatever unknowns are there to find. (Do not forget, this is exploration; If we knew what we were going to find there would be no reason to go.)
No it's not - if it were exploration you wouldn't need a million ton ship. You'd send a probe with a camera, or a bunch of probes with different observation equipment, and the means to send signals back to earth.
In which case it can travel at much more than 1% the speed of light.

This is trying to make use of the resources of another solar system. In order to do so we should know as much as possible about said system - which is why we'll observe it as best we can from here with telescopes, and then send small, fast, probes to learn more about it before actually sending anything that can take advantage of those resources in a major way.

[BenBurch]

I don't think that even a small probe can encounter interstellar dust grains at 5% of the speed of light and not find this to be a problem. So we reduce the time to 1000 years.

That is a long time to wait for a probe to report home.

Even if you could somehow deal with 10%, that is 500 years.

And I have to wonder if, in 500 years, we couldn't build an immense synthetic aperture optical telescope and see anything a small probe's instruments can? And in fact look for things not yet thought of when the probe was designed?

That is why you need a mind (or minds) onboard. To adapt. To build instruments not thought of yet. To conduct research.

[Dogdoctor]

Given enough space in a spaceship you could create an ecosystem that would survive a long time. Ok maybe we lack the knowledge to do this but fusion or fission engines and solar power may help.

[Roboramma]

And I have to wonder if, in 500 years, we couldn't build an immense synthetic aperture optical telescope and see anything a small probe's instruments can? And in fact look for things not yet thought of when the probe was designed? My point was that we don't need to send a huge interstellar ship to do exploration. This only adds to that point.

That is why you need a mind (or minds) onboard. To adapt. To build instruments not thought of yet. To conduct research. No, that is why it would be nice to have a mind. Again, I don't see any reason why we couldn't build one. But regardless, this isn't a reason why you need one.

[PixyMisa]

It needs to be able to re-design itself, because there is no way we will know in advance anything about how such a machine will function with age, and deal intelligently with whatever unknowns are there to find. (Do not forget, this is exploration; If we knew what we were going to find there would be no reason to go.)
No, it's not exploration, it's colonisation. The exploration program can begin with lightweight high-speed probes. (Where high-speed might be 3% or even 5% of light speed.)

And no, it doesn't need to be able to re-design itself, because it will be in contact with the Earth at all times. Not that it would need to do so in any case.

[PixyMisa]

That is why you need a mind (or minds) onboard. To adapt. To build instruments not thought of yet. To conduct research.
You send out bunches of little probes. If the destination is 10 light years away and we're travelling at 5% of the speed of light, they're going to take 200 years to get there. So just pop another one off on its way every decade or so.

If probe #1 sees something interesting and doesn't know what to make of it, we have time to analyse the telemetry and send an update to probe #3. Probe #2 unfortunately is just going to zoom blithely past.

And once you're happy with what your probes have found, you can launch the million-ton colony ship, already knowing what you're going to see at the other end.

[BenBurch]

No, it's not exploration, it's colonisation. The exploration program can begin with lightweight high-speed probes. (Where high-speed might be 3% or even 5% of light speed.)

And no, it doesn't need to be able to re-design itself, because it will be in contact with the Earth at all times. Not that it would need to do so in any case.

You cannot make a machine to run that long. No machine to date has, and the ones that have run the longest, Pioneer 6 and its ilk, have things that have failed onboard. You need intelligent maintenance if you want any hope of any of them making it.

And I thought we WERE talking about the probes; The colony will still have to be the size of Chicago.

[BenBurch]

In fairness, this is 106 years old and still going;

http://www.centennialbulb.org/images/bulbcam.jpg

But its not really a machine.

[geni]

I don't think you can send a cold ship and have that work.

Look up "cold welding" some day. No moving parts will still move after that much time.

Effect is greatly reduced at 3K


And forget about any springs, gaskets, or anything that has to be flexible...


most of these do not work too well at those kind of temps anyway. electro magnets are a better option.


Nor will integrated circuits function after all that time, as these degrade due to cosmic rays and due to materials issues like implanted doping diffusing to the point of uselessness and etc.


A mixture of shilding and overdesign should take care of that.


And *something* has to be alive in there for it to know when to wake up.


Not quite. your alive object could simply be the target of a ionising radition source.


Show me a machine that has run for 100 years without a rebuilding anywhere.

Show me a 100 year old machine disused and unmaintained for most of that which can be made to work today as well as when it was made?


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

The problem is that at present humans have little insentive to build something that will outlast the expected lifetime of the nuclear reactor. For anything else it is cheaper to build with the expectation of overhalls or replacement with a better machine.

[PixyMisa]

You cannot make a machine to run that long. No machine to date has, and the ones that have run the longest, Pioneer 6 and its ilk, have things that have failed onboard.
Yes, some components have failed. Materials science is substantially better now than in the early 70s, and so is computer technology.

You need intelligent maintenance if you want any hope of any of them making it.
Yes, for long voyages, but you do not need a natural-language AI to do this.

And I thought we WERE talking about the probes; The colony will still have to be the size of Chicago.
Why?

[PixyMisa]

Why?
Sorry, that was unfair.

BenBurch, you do raise salient points. I've already said why I find most of your arguments unconvincing, though, and I'll add that this last statement that "The colony will still have to be the size of Chicago." is absurd. Even granted everything you've said, you still can't support that claim.

[BenBurch]

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

Not the case I was talking about as it was not disused for that long; It moves constantly. Still there is evident wear and we can be sure that failure will come.

And I am quite sure that there are clocks that old too, but they also have never been still. Ask any clockmaker if a 100 year old clock that has been unwound for that whole period would run if wound up? Most will tell you "no" and for good reason. The spring steel will have changed and become brittle and the lubrication will have evaporated or turned to varnish and gears will have chemically merged at the bearing surfaces, especially if there were any residual tension in the mechanism. Such a clock needs to be disassembled, cleaned, re-sprung, and lubricated to function after not moving for so long a time.

But please do not think that nuclear power stations are not in a constant state of refurbishment. They are OBSESSIVELY maintained by large crews of technicians, operators and engineers.

And no, we do not design machines to last forever. I don't think we CAN.

[geni]

Not the case I was talking about as it was not disused for that long; It moves constantly. Still there is evident wear and we can be sure that failure will come.

You want a fairly solidly built item built before 1907 that has survived two world wars and has not been switched on in that period? Not going to happen. Scap value and human curiosity alone would see to that. The closest would be the John Bull steam locomotive. I supose there might be a few other simular odds and ends in museams but a high oxygen high water vapour enviroment is not a good place to store things.


And I am quite sure that there are clocks that old too, but they also have never been still. Ask any clockmaker if a 100 year old clock that has been unwound for that whole period would run if wound up? Most will tell you "no" and for good reason. The spring steel will have changed and become brittle and the lubrication will have evaporated or turned to varnish and gears will have chemically merged at the bearing surfaces, especially if there were any residual tension in the mechanism. Such a clock needs to be disassembled, cleaned, re-sprung, and lubricated to function after not moving for so long a time.

All those problems can be delt with by useing a different design. Just no reason to do so.


But please do not think that nuclear power stations are not in a constant state of refurbishment. They are OBSESSIVELY maintained by large crews of technicians, operators and engineers.

Bits of them yes. Other parts you have to rely on continueing to function for the lifetime of the plant because there is no practical way to fix them.


And no, we do not design machines to last forever. I don't think we CAN.

You didn't ask for forever you asked for 5K years. A very differnt objective.

[BenBurch]

You didn't ask for forever you asked for 5K years. A very differnt objective.

Nope. 5K years == forever for any imaginable machine that is not intelligent enough to fix itself.

BTW, steam locomotives are not a good example; There is hardly a steam engine that exists that is not mostly made of parts from rebuilds. In fact that is a problem for museums who want both the authentic article and a working machine; The two trade off for each other.

[Tumbleweed]

Your space travel fantasies don't seem to include the juggernaut of relativitity. Since a mile and a second are one thing to one observer and quite another to another how can you possibly agree on how far away something is or how long it will take to get there The old one twin goes into space and comes back still young while his twin has aged bit. Or consider the Star Gate hypothesis of 2001 a space odessy, also based on relativity
Consider two points in space, not matter, moving towards each other at the speed of light, no physics involved, merely by mathematical definition. The distance between those points would be zero. Yet if the points became stationary or slowed in some point in the equation, distance would be created and would be proportional to the velocity of the moving points. An outside observer would simply see two points moving towards each other at great speed, a great distance between them.
You cannot accelerate away from a fixed point at the speed of light, but if two galaxies are already moving at great relative velocity to one another, isn't most of the relative velocity needed to make space time "warp" in situ?

[PixyMisa]

Um, what?

[drkitten]

Your space travel fantasies don't seem to include the juggernaut of relativitity. Since a mile and a second are one thing to one observer and quite another to another how can you possibly agree on how far away something is or how long it will take to get there The old one twin goes into space and comes back still young while his twin has aged bit. Or consider the Star Gate hypothesis of 2001 a space odessy, also based on relativity
Consider two points in space, not matter, moving towards each other at the speed of light, no physics involved, merely by mathematical definition. The distance between those points would be zero. Yet if the points became stationary or slowed in some point in the equation, distance would be created and would be proportional to the velocity of the moving points. An outside observer would simply see two points moving towards each other at great speed, a great distance between them.
You cannot accelerate away from a fixed point at the speed of light, but if two galaxies are already moving at great relative velocity to one another, isn't most of the relative velocity needed to make space time "warp" in situ?

The quickest answer is that relativity doesn't work that way.

[geni]

Nope. 5K years == forever for any imaginable machine that is not intelligent enough to fix itself.

Not at 3K in a hard vaccume.


BTW, steam locomotives are not a good example; There is hardly a steam engine that exists that is not mostly made of parts from rebuilds. In fact that is a problem for museums who want both the authentic article and a working machine; The two trade off for each other.

Which is why I selected a very specific engine. Steam engins won't be ideal but they are about the only pre 1907 object that just might have been a stoped position for that length of time for reasons other than a breakdown.

[BeAChooser]

Our chances of finding a place that can be colonized within 50 ly appears to be small.

According to this,

http://www.atlasoftheuniverse.com/50lys.html

there are about 1400 stars systems within this volume of which 133 are visible to the naked eye from earth. Most of the fainter ones are red dwarfs. And according to this, most of those 133 are very similar to the sun. Based on what astronomers are finding, a large portion of the 133 likely have planets.

And this

http://findarticles.com/p/articles/mi_m1511/is_3_21/ai_59535400

states "Finding Another Earth As early 2011, NASA hopes to launch what may be the most ambitious telescope ever conceived: the Terrestrial Planet Finder. ... snip ... Planet Finder will consist of a football-field-sized array of four massive telescopes and a mother ship. Each telescope will train its powerful infrared eyes on a star within 50 light-years of Earth, filter out glare, and scan for pinpoint images of individual planets. Light from each of the telescopes will be beamed to the mother ship and combined into a single high-resolution image. "We'll be able to take a snapshot of the system and see individual planets orbiting around a star," says Beichman, who works at the Jet Propulsion Laboratory. The telescope won't be able to spot continents and certainly not any little green men. But its spectrometers will be able to sniff out the presence of atmospheric gases like ozone that, on Earth at least, are linked to life."

So maybe we'll find out before long where to head. :)

[tracer]

Couldn't I just vote myself off the island?

[BenBurch]

Well, John Bull, though she is operable (last time in 1981) is a restored museum piece.

Were you to want to operate her again, you would need to perform a hydrostatic test because, amazingly, boiler plates and tubes can crack all by themselves even under ideal dry storage, and seals and gaskets, usually made of an organic material, rot even in dry air. And the law would require a valid boiler certificate in any case.

And I just do not buy the concept that you can store a huge and complex machine near absolute zero for thousands of years and get it to operate when turned on. We simply do not know how the materials we are using will age. Without knowing that, you absolutely need active inspections and repairs.

[PixyMisa]

And I just do not buy the concept that you can store a huge and complex machine near absolute zero for thousands of years and get it to operate when turned on. We simply do not know how the materials we are using will age. Without knowing that, you absolutely need active inspections and repairs.
I only partly agree with this. You can make predictions based on several separate bodies of evidence - archaeology (what has happened to 5000 year old samples of metal, glass and ceramics), projection (what happens to these materials over 10, 20, 50, 100 years), accelerated wear (what happens in 1, 2, 5, 10 years if we add thermal, mechanical or other stress), and computer simulations.

You will still need active maintenance. Even if you can design every component to have a 10,000 year MTBF, some components are going to fail just 100 years into the voyage. You have to be able to find them, replace them, recycle them if at all possible, make new parts, repair the machines that make new parts, and so on.

I definitely don't buy that you need a fully human-level, much less superhuman, AI to manage that.

[BenBurch]

I only partly agree with this. You can make predictions based on several separate bodies of evidence - archaeology (what has happened to 5000 year old samples of metal, glass and ceramics), projection (what happens to these materials over 10, 20, 50, 100 years), accelerated wear (what happens in 1, 2, 5, 10 years if we add thermal, mechanical or other stress), and computer simulations.

You will still need active maintenance. Even if you can design every component to have a 10,000 year MTBF, some components are going to fail just 100 years into the voyage. You have to be able to find them, replace them, recycle them if at all possible, make new parts, repair the machines that make new parts, and so on.

I definitely don't buy that you need a fully human-level, much less superhuman, AI to manage that.

Well, yeah, you might; What if the reason a part is failing is a design flaw? It will need to be changed. And since none of our machines are this old, we cannot know what flaws will be revealed after so much time.

After a lifetime of being an engineer, I know that engineering is only ever an approximate art. We can be as exact as we want to be about what we do know, but sometimes what we do not know makes that all moot.

And if the flaw is in the AI software?

I'm not saying that a less-than-sentient self-repairing spacecraft could NEVER work, I just don't think I would EXPECT it to.

[geni]

Well, John Bull, though she is operable (last time in 1981) is a restored museum piece.

Were you to want to operate her again, you would need to perform a hydrostatic test because, amazingly, boiler plates and tubes can crack all by themselves even under ideal dry storage, and seals and gaskets, usually made of an organic material, rot even in dry air. And the law would require a valid boiler certificate in any case.

No one is going to let you carry out a hydrostatic test on that locomotive. Less invasive tests are more likely. The train has not been significantly restored.


And I just do not buy the concept that you can store a huge and complex machine near absolute zero for thousands of years and get it to operate when turned on. We simply do not know how the materials we are using will age. Without knowing that, you absolutely need active inspections and repairs.

Only if our preditions as to how they will age turn out to be wrong. The relivant envitorment is a fairly simple one so it should be posible to make reasonable projections as to what will happen to the various materials over that time peroid.