Von Neumann probes

[Bill Phil]

I'd like to bring one smart man named Fermi into this discussion. So, if is so easy to colonize space via these probes, where are they? Given the age of the Universe, there should be hundreds or even thousands of these probes already on Earth.

Maybe they had self destruction programs???

Or maybe they're so small ( a few meters) that they're very difficult to locate...?

[PB666]

We're not self replicating... We create anatomically similar copies, but they are genetically different. And two people of the opposite gender are required.

Neither can replicator ships. Heisenberg uncertainty principle. And we could using parthenogenesis achieve the same affect with humans. The limitation is a function of telomerase activity, the only problem is that in higher animals parthenogenesis is limited to females.

Artificially engineer a human female germ-line cell that creates its own new lineage, it will become a psuedoovary.

That expresses

1. A polymerases have 100-fold less erroneous proofreading function (there are bacteria that do this)

2. Make sure enough of this polymerase is made so that enough last until the tissue forms in the replicon.

3. Give the telomerases that same restorative capacity as cells in female germ line.

4. Make sure the cells in the organ spend a fraction of their time in diploid condensed phase and never go through meiosis.

When new humans are needed take 100 cells from the pseudo ovary, expand the cells to eight fold stage, sequence the cells looking for errors. If the cells have no errors remove

an ova, remove its nucleus and replace with confirmed nucleus. Then insert back into the original female and allow to develope.

[Whirligig Girl]

A downside is that you might replace the entire universe with probes.

An entire universe of Soviet Luna 3s. How wonderful.

[PB666]

Biology is the main reason why I don't believe that self-replicating machines could work. Organisms do reproduce, but only with the support of an entire ecosystem. The more complex an organism is, the more support it requires for reproduction, and the more likely it becomes that it can't continue reproducing after changes to the ecosystem.

Mechanical replicates don't need to grow, and you can simply do a digital data transfer to learn them. What is missing is diversity, we assume that the lessons of our little corner of the universe would suffice as a template for potential experience and problem solving elsewhere. Black swan theory applies. Humans under recombination evolution, its why we are more complex than bacteria. Recombination offers unanticipated solutions for future problems.

Lets give an example, sociobiologist try to understand how humans learn, we then feed the paradigm into a computer algorithm and give it roughly our senses (so for instance we give the computer the ability to see colors, semi-panoramic vision, stereoscopic sound, etc), sense hot and cold, sense smells. And then we give it roughly our facilities. We could also give other abilities, to see infrared, more smells, ultrasound, ultraviolet, gamma, cosmic, neutrinos, etc. Then we set it off on its little way into the great void and it disappears. All the while we find the standard model is complete, and we find the dark energy and dark matter wave and other things. We cannot see these things, but if they are important to see we can build machines that can sense them. The old machine traveling in the void in space is not going to get the richness of experiences, it may have other experiences. So if the ship were human and resources, you could send the humans the instruction to build the new sensors and they could be added. In addition humans will evolve, and our senses will evolve to suit whatever new worlds we make it to (not likely but . . . ). So we have both the biological capacity to sense and the appended mechanical capacity to sense.

[Technical Ben]

Biology is the main reason why I don't believe that self-replicating machines could work. Organisms do reproduce, but only with the support of an entire ecosystem. The more complex an organism is, the more support it requires for reproduction, and the more likely it becomes that it can't continue reproducing after changes to the ecosystem.

True.

So we'd need to send everything. Or hitch hike on a nearby passing rouge planet.

[AndrewBCrisp]

There is also this article that had some good thoughts on Von Neumann probes and the Fermi Paradox. The last paragraph makes a great tl-dr summation:

And perhaps there lies the answer to Fermi’s Paradox. Maybe intelligent extraterrestrials are more interested in making a good first impression than the incessant consumption of resources. Perhaps that is why the Solar System wasn’t scoured by a wave of Von Neumann probes long ago. The folly of our assumption is that we see all before us as resources to be utilised, but why should intelligent extraterrestrial life share that outlook? Maybe they are more interested in contact than consumption – a criticism that can be levelled at other ideas in SETI, such as Kardashev civilisations and Dyson spheres that have been discussed recently on Centauri Dreams. Perhaps instead there is a Bracewell probe already here, lurking in in a Lagrange point, or in the shadow of an asteroid, watching and waiting to be discovered. If that’s the case, it may be one our own Von Neumann probes that first encounters it – and we want to make sure that we make the right impression with our own probe the day that happens.

[Jouni]

Mechanical replicates don't need to grow, and you can simply do a digital data transfer to learn them.

You missed the point. All the evidence, from biology and elsewhere, suggests that complex machines require more external support for replication than simple machines. A self-replicating spaceship sounds less feasible than a self-replicating shovel. If you add the machinery for manufactoring spaceship components to a spaceship, you need five times more machinery for manufactoring the components for that machinery. Every system you add to help in self-replication just pushes the goal further away.

[Camacha]

Biology is the main reason why I don't believe that self-replicating machines could work. Organisms do reproduce, but only with the support of an entire ecosystem. The more complex an organism is, the more support it requires for reproduction, and the more likely it becomes that it can't continue reproducing after changes to the ecosystem.

Are you confusing support with resources?

[Jouni]

Are you confusing support with resources?

Not resources. More like other machines for helping in the replication, or refined resources that the self-replicating machine can use.

[Camacha]

Not resources. More like other machines for helping in the replication, or refined resources that the self-replicating machine can use.

I do not really see how that applies to the organisms in your quote. Large organisms need more resource, but not really more support.

[Jouni]

I do not really see how that applies to the organisms in your quote. Large organisms need more resource, but not really more support.

A simple ecosystem can support simple organisms such as bacteria. Complex organisms such as wolves require more complex ecosystems.

[Camacha]

A simple ecosystem can support simple organisms such as bacteria. Complex organisms such as wolves require more complex ecosystems.

Still, ecosystems are merely providers of resources, mainly food and often water.

[Jouni]

Still, ecosystems are merely providers of resources, mainly food and often water.

Just like a factory is merely a provider of resources.

Complex organisms, especially active ones, require their nutrients in concentrated, highly refined forms. They can't sustain themselves by slowly extracting resources from the environment.

[Camacha]

Just like a factory is merely a provider of resources.

I feel this comparison has deteriorated to the point of not functioning at all any more.

Complex organisms, especially active ones, require their nutrients in concentrated, highly refined forms. They can't sustain themselves by slowly extracting resources from the environment.

That makes more sense

[magnemoe]

You missed the point. All the evidence, from biology and elsewhere, suggests that complex machines require more external support for replication than simple machines. A self-replicating spaceship sounds less feasible than a self-replicating shovel. If you add the machinery for manufactoring spaceship components to a spaceship, you need five times more machinery for manufactoring the components for that machinery. Every system you add to help in self-replication just pushes the goal further away.

This is the stark opposite of the machines will replace us idea who is pretty common.

Yes you will get increasing complexity up to an point, you have to be able to make any part of the ship and its support structure.

You have one benefit in that you can scale up, speed or klutziness in the start is not very relevant. You can also cannibalize part of your ship until you get mining running.

Still you might end up with an mobile factory who is much like an generation ship with the benefit that it don't need life support.

No we have no idea how to do this now, in some hundred years it would be more practical, its an drive for more automation and self repair capability especially in space.

[andrewas]

Complex organisms, especially active ones, require their nutrients in concentrated, highly refined forms. They can't sustain themselves by slowly extracting resources from the environment.

True for biological organisms that evolved in a world where less complex organisms were already there to prey on. I don't think you can generalise from that to self replicating machinery

[Jouni]

Yes you will get increasing complexity up to an point, you have to be able to make any part of the ship and its support structure.

This is a huge assumption. So far all evidence suggests that the complexity of manufacturing equipment grows faster than the complexity of manufactured items.

True for biological organisms that evolved in a world where less complex organisms were already there to prey on. I don't think you can generalise from that to self replicating machinery

It's the reasonable assumption, based on all we know about producing things. Von Neumann machines are essentially wishful thinking, because we have no evidence that they could ever work.

[magnemoe]

This is a huge assumption. So far all evidence suggests that the complexity of manufacturing equipment grows faster than the complexity of manufactured items.

It's the reasonable assumption, based on all we know about producing things. Von Neumann machines are essentially wishful thinking, because we have no evidence that they could ever work.

I agree that manufacturing is getting more and more complex, this however is because it makes most sense, if you want to increase performance a lot while keeping price down the best way to do this has been to increase the complexity, you also use more and more complex manufacturing plants as you want to produce in large volumes at an low cost, you will also often want the factory to be able to produce something else without rebuilding it.

An self replicated probe would work in an totally different way. Its rarely an need for automated repair an mechanic is cheaper and way smarter and more flexible.

Self diagnostic on the other hand is an rapid growing field, it let the mechanic identify problems and fix them fast, often before they causes shutdown.

Yes an Von-Neumann system will need nanotechnology to some degree, to produce chips and sensors at least and to produce the printer who makes them.

I say its less likely we get an sapient AI than Von-Neumann machines, as we have do not even know the system requirements for the first while the second is mostly engineering.

[KSK]

There is also this article that had some good thoughts on Von Neumann probes and the Fermi Paradox. The last paragraph makes a great tl-dr summation:

This. Even if we could, why on earth would we want to indulge in celestial vandalism on that sort of scale.

[Camacha]

This is a huge assumption. So far all evidence suggests that the complexity of manufacturing equipment grows faster than the complexity of manufactured items.

This has only been true for a while. With the advent of CNC manufacturing and, more recently, 3D printing, product complexity has sky-rocketed, while machine complexity is at the same level for each part.

[Jouni]

I say its less likely we get an sapient AI than Von-Neumann machines, as we have do not even know the system requirements for the first while the second is mostly engineering.

I disagree. Von Neumann machines are are based on handwaving, wishful thinking, and science fiction stories. Nobody in the world has any idea how they could be built, or whether they are possible at all. Sapient AI, on the other hand, could feasibly result from incremental development. We already know that sapient machines are possible. We should also have enough computing power to build non-biological ones. All that remains is "mostly engineering".

The first sign of real AI will probably be years of confused debate. Experts start to suspect that some systems they have built could be sentient. There will be debate about what that really means, how it could be tested, and how one could actually communicate with an AI. People will debate about ethics, whether AIs should be covered by animal rights laws, whether AI research should be regulated, and whether AIs are really dangerous. The first AIs would not be the omnipotent entities from science fiction. More likely, they would be heaps of research-quality code, full of bugs and beginner mistakes. The AIs would crash often, and the code wouldn't even compile half the time.

This could be reality in a few years, or it could take decades or even centuries. In any case, it will happen, unless we manage to destroy ourselves first.

This has only been true for a while. With the advent of CNC manufacturing and, more recently, 3D printing, product complexity has sky-rocketed, while machine complexity is at the same level for each part.

3d printing just increases the complexity of everything. Instead of producing things on-site with simple tools, or mass-producing them at a central facility, we will start using a huge bunch of complex tools for on-site production. While 3d printing can make the on-site production of simple things cost-effective, it won't change the fact that different tools are good at different tasks, while being completely hopeless at others.

[Camacha]

3d printing just increases the complexity of everything.

You will need to explain that, because I would argue the opposite is true. You seem to forget that modern tools can make a much broader spectrum of products with the same machine, and, maybe even more importantly, that a relatively small number of these machines together could make another machine that can make different things.

[Jouni]

You will need to explain that, because I would argue the opposite is true. You seem to forget that modern tools can make a much broader spectrum of products with the same machine, and, maybe even more importantly, that a relatively small number of these machines together could make another machine that can make different things.

Modern tools are more versatile than old tools, but you still need more modern tools to make a modern tool than old tools to make an old tool. A 3d printer is an extremely complex piece of equipment, combining the latest achievements from many branches of technology.

[spudman2]

An entire universe of Soviet Luna 3s. How wonderful.

You know you would like it. :hailprobe:

[Camacha]

Modern tools are more versatile than old tools, but you still need more modern tools to make a modern tool than old tools to make an old tool.

That's my point: it is not true. To woodwork a spoon or cabinet on traditional machines, you need a properly filled workshop. With modern CNC equipment, you could do with one machine if you wanted. Yes, it is a more complex machine, but like I said, also much more versatile.

There is a reason digital manufacturing and printing has taken the industrial world by storm: you can produce much, much more products at a somewhat reduced (but not unusable slow) production speed. Only true mass products are still manufactured in 'dumb' production lines, everything else (including books, bridges and bikeframes) is done in more flexible digital production. Assuming the probes have some sort of self improving evolutionary development mechanism, digital manufacturing is the way to go.