A datacenter in space might be the silliest idea ever proposed.

[Phishfry]

Look at the old publications from Microsoft (and others?), building data centers that can be submerged. I think that was in the late 90s or early 2000s.

Post #24 covered that. Microsoft's Project Natick

Probably the closest thing to space we have actually experimented with:
https://en.wikipedia.org/wiki/Project_Natick

 

[lanin]

Look at the old publications from Microsoft (and others?), building data centers that can be submerged. I think that was in the late 90s or early 2000s.

Yes, I read about the project linked in post #24. I think the time just hasn't come yet. But it will. I'm more thinking about a mobile data center that can navigate in three dimensions and dynamically connect to global networks in the depths of the ocean...

 

[Maturin]

bearing balls and some explosive.

Not even need this. Some piece of junk on a collision course is all it needs (a Tesla for example.)
Don't underestimate the velocity orbiting objects need to stay there, the kinetic energy they possess and how fragile sattelites or space stations are.

For many people space seems to be some modern kind of "settlement in the west without indians" - our place is trashed up, so now we need more space to dump our junk.
When I see all those fancy ideas what all should be launched up into space it's proof to me, many people too strong believe in fairy tales science fiction movies tell, but have no real idea how space reality actually is.
One example that proves this to me is, like until the 1980s there have been only discussions, what all needs to be build where, but only some not to be taken seriously fancy eco-freaks anybody is allowed to make fun of tried to think of how to avoid trash and cleaning up.

Before one thinks of sending up even more junk, one better answers the question first,
how to clean up to get enough space for all that new junk?

Wkipedia: Space debris

We needed 300 years to trash our planet, 50 years to even start any cleaning up at all lagging crucially behind, and still the trashing up still continues - accelerating.
We already trashed up our space. Needed 70 years for that. Not yet even started to clean up there, but only thinking of what more junk we can launch up there and trashing up space even more - accelerating.

Currently, where the limits of what our planet can bear are apparently exceeded, we elect governments telling us, that all this green environment protection bogus is just a complete load, climate change ain't nothing we need to care about at all, lack of resources is all just fairy tales made up by fancy morons lost contact to reality better not listen to.

Elon ain't no better do-gooder than all the rest of those capitalists. He also decoys and disguises his ideas with making the world a better place. That's called white- or green- or brain-washing.
In fact it's only even more trashing up, even more waste of resources and even more destruction of the environment for to make a few ones shovel even more money in their own pockets. And nothing else.

It's like in the bad old days again we greybeards already saw in the 1970s...1990s.
Cleaning up does not pay off. Only producing even more junk does. So:
Who wants to clean up? Nobody.
Who wants to produce even more junk? Everybody.
What about resources and environment? Let others care about that later.
Where shall all the power come from we need for that stunt? Nuclear power.
How shall this all be payed? Subventions by tax payers money.
Groundhog Day.
The only new thing so far by the last 300 years is, we don't trash up Earth but space instead.
But that's only because Earth cannot be trashed up anymore.

If we don't learn how to save resources, stop wasting space, stop destroying the environment, stop to trash up our own place, and how to clean up, it's completely pointless to expand anywhere, because that will not solve anything but expand our problems only.
And if we learn that, we can also stay here.

 

[Tomolok]

Not even need this. Some piece of junk on a collision course is all it needs (a Tesla for example.)
Don't underestimate the velocity orbiting objects need to stay there, the kinetic energy they possess and how fragile sattelites or space stations are.

For many people space seems to be some modern kind of "settlement in the west without indians" - our place is trashed up, so now we need more space to dump our junk.
When I see all those fancy ideas what all should be launched up into space it's proof to me, many people too strong believe in fairy tales science fiction movies tell, but have no real idea how space reality actually is.
One example that proves this to me is, like until the 1980s there have been only discussions, what all needs to be build where, but only some not to be taken seriously fancy eco-freaks anybody is allowed to make fun of tried to think of how to avoid trash and cleaning up.

Before one thinks of sending up even more junk, one better answers the question first,
how to clean up to get enough space for all that new junk?

Wkipedia: Space debris

View attachment 26625View attachment 26626

We needed 300 years to trash our planet, 50 years to even start any cleaning up at all lagging crucially behind, and still the trashing up still continues - accelerating.
We already trashed up our space. Needed 70 years for that. Not yet even started to clean up there, but only thinking of what more junk we can launch up there and trashing up space even more - accelerating.

Currently, where the limits of what our planet can bear are apparently exceeded, we elect governments telling us, that all this green environment protection bogus is just a complete load, climate change ain't nothing we need to care about at all, lack of resources is all just fairy tales made up by fancy morons lost contact to reality better not listen to.

Elon ain't no better do-gooder than all the rest of those capitalist. He also decoys and disguises his ideas with making the world a better place. That's called white- or green- or brain-washing.
In fact it's only even more trashing up, even more waste of resources and even more destruction of the environment for to make a few ones shovel even more money in their own pockets. And nothing else.

It's like in the bad old days again we greybeards already saw in the 1970s...1990s.
Cleaning up does not pay off. Only producing even more junk does. So:
Who wants to clean up? Nobody.
Who wants to produce even more junk? Everybody.
What about resources and environment? Let others care about that later.
Where shall all the power come from we need for that stunt? Nuclear power.
How shall this all be payed? Subventions by tax payers money.
Groundhog Day.
The only new thing so far by the last 300 years is, we don't trash up Earth but space instead.
But that's only because Earth cannot be trashed up anymore.

If we don't learn how to save resources, stop wasting space, stop destroying the environment, stop to trash up our place, and how to clean up, it's completely pointless to expand anywhere, because that will not solve anything but expand our problems only.
And if we learn that, we can also stay here.

A good news for you is that I and many other people I know completely(or at least largely ) agree with you,and we are not greybeards.

 

[Crivens]

Maturin I can not avoid seeing the parallel between the greybeards of old warning about environmental pollution and the greybeards of today warning about AI. I fear this will take a bad end as well.

 

[kent_dorfman766]

It is actually an issue for some specialized applications. Those usually involve satellite-to-satellite communication. For example from a listening or photo satellite to a processing satellite (the "data center") to a disposable weapon satellite (with kinetic or energy weapon). If you want to do these things at sub-ms latency, you're not going to go down to earth and back up.

I think your response takes things out of context. Of course there will always be "low latency requirement" apps, but those apps are not on the internet anyways, and the "bring it to the masses" component means TCP/IP which allows for variable latency in the seconds range. Where space packet switching breaks down is that TCP/IP is an end to end channel, whereas space comms must deal with variable "next hop" conditions without regard to whether a complete path exists at time of transmission. The bundle protocol attempts to address this issue. For LEO stations TCP/IP is achievable. For deep space stations...it is not.

 

[Maturin]

A good news for you is that I and many other people I know completely(or at least largely ) agree with you,and we are not greybeards.

Thanks.
Such words always help. And we shall not forget that, and keep supporting each other, cause we are still not a rock solid majority.

I used the term geybeards, 'cause it's a BS term suggesting it's about old and young, while like anything else it's actually about right or wrong.

The question is also for whom is that good news. You see, I am 54. My wife and I have no children. So we don't really need to bother anymore. But we do. While at the same time we see so many having kids and don't.

And when I look at the current worldwide economical and political situation we are by far neither enough, nor heading the right course, but in contrary steering more the opposite course again.
So there is more than enough left to do, before we can lean back.

I fear this will take a bad end as well.

Me too. But it doesn't necessarily has to.
I don't want to stress all the details about that topic again and again - difference between some LLM some programmer uses locally for their own work vs. students cheating on tests in school with CrapGPT.
Just last week my wife (teacher) had their students write another class test. Even she told them explicitely any use of a smartphone is a attempt to deceive and additionally hang up a foolproof sign "smartphones strictly forbidden" she collected 26 [!] tests during one single class test: attempt to deceive, failed, F -
Summarized it's just a technical tool that left the laboratories of science too early, put in the hands of greedy capitalists, and confronts a society by far not even remotely mature enough to deal with it.


I like to recommend the movie WALL-E, USA 2008 at this point.

If this movie had been published when I was in school (1980s) perhaps a teacher had shown it to us on last day before summer vacation.
After the movie we were asked, 'What is the framework plot of it?'
One pupil, 'A lonely robot cleans up Earth, until there comes another one visiting...'
'No. That's the plot. I mean the framework, the superior plot, the overall idea.'
Quiet agonizing.
The nerd, 'It anticipates a possible future by the current situation where we are heading at if nothing is changed.'
'Very good! That's right.'
Everybody looking with grim at this grade grubber for not only knowing the answer on a difficult question, but also for putting it directly into the right words. "anticipating" What a geek!
Teacher, 'What question does this movie subtle ask the audience?'
Quiet agonizing.
Teacher trying to help, '...a possible future, where we are heading at if nothing is changed. What question does this raise?'
Quiet agonizing.
Teacher again, 'How is life for the humans in that future?'
Class clown, 'Boring, because there is no sex.'
Everybody laughing.
Teacher, 'Boring. That's right. The people are very bored.'
Other pupil, 'And they are all fat and immobile.'
'Decadent!' comes the nerd.
Teacher, 'Right. And how all this ends up?'
'They return to Earth.'
Teacher, 'Exactly. And now again: What question is raised by that subtle?'
Quiet agonizing.
The bell rings, everybody starts packing and storming out.
Teacher loud, 'If it is worth all the trouble at all, and if to head for this in the first place. Nice summer vacation!'

 

[Crivens]

steering more the opposite

MADE to steer, please. The propaganda and gaslighting is huge in that area. People get lied to that it's OK not to care.

 

[Maturin]

MADE to steer, please. The propaganda and gaslighting is huge in that area. People get lied to that it's OK not to care.

True. But they are still free to learn and think for themselves, and to decide what they buy and who they vote for.

 

[ralphbsz]

I think your response takes things out of context. Of course there will always be "low latency requirement" apps, but those apps are not on the internet anyways, and the "bring it to the masses" component means TCP/IP which allows for variable latency in the seconds range. Where space packet switching breaks down is that TCP/IP is an end to end channel, whereas space comms must deal with variable "next hop" conditions without regard to whether a complete path exists at time of transmission. The bundle protocol attempts to address this issue. For LEO stations TCP/IP is achievable. For deep space stations...it is not.

What I meant is much simpler. Say you have one satellite which sense, like takes a picture. And another satellite that shoots things based on the pictures taken by the first one. In the trade, that is called a fractionated autonomous kill satellite (look up fractionated satellites and the F6 project on Wikipedia). Say in the above example, we need an enormous amount of compute power, like an AI data center's worth of compute power, between the two satellites, to make the kill decision. But the decision has to be made very fast, in milliseconds. In this example, the AI data center can't be on a fixed location on earth, since the latency is too long. Therefore the AI data center has to be right up there in space next to the sensor and the shooter satellite. This is one (and in my mind the only) justification for building an AI data center in space.

And obviously, the whole setup is deeply troubling. Militarizing space is highly dangerous, since any armed conflict there gets us much closer to the Kessler syndrome, perhaps really fast. Matter-of-fact, the Kessler syndrome becomes a tool of mutually assured destruction and is like "salting the earth" (see Carthage). And making kill decisions autonomously, in particular based on AI, seems also quite insane. But it is being considered.

 

[blackbird9]

You cannot violently attack an AI datacenter in orbit.

Yoda says: "so certain, are you..."

Crystal clear: China sharpening its anti-satellite laser edge - Asia Times

China is using crystals to enhance the power of satellite-destroying lasers in a daring attempt to blind US satellites and gain strategic superiority in

asiatimes.com

How China is expanding its anti-satellite arsenal

The Space Force chief doesn’t like the “curves” of how China is progressing vs the U.S.—and says he doesn’t have the funding to reverse it.

www.defenseone.com

 

[blackbird9]

Datacomm latency isn't much of an issue for LEO stations (low earth orbit) but becomes a huge issue for deep space or lunar datacenters. Traditional TCP/IP is unsuitable for high latency, variable connectivity topologies, and the current work-around is something called DTN (delay tolerant networking) via the bundle protocol.

I was very impressed by the IR laser link that was demonstrated on the Artemis II. They had near-real time video feed from the spacecraft, and claimed data transfer rates of up to 260 megabytes per second. An impressive technology demonstration. Of course that's a long way from making the kind of always-up reliable physical link supporting (presumably) TCP/IP that you would need for something like an AI datacentre on the moon.

NASA's Orion Artemis II Optical Communications System (O2O) - NASA

Throughout Orion’s 10-day journey around the Moon, the Artemis II mission will rely on NASA’s Near Space Network and Deep Space Network as its primary

www.nasa.gov

View: https://www.youtube.com/watch?v=QJJfxxbqhf0

 

[kent_dorfman766]

Yoda says: "so certain, are you..."

Crystal clear: China sharpening its anti-satellite laser edge - Asia Times

China is using crystals to enhance the power of satellite-destroying lasers in a daring attempt to blind US satellites and gain strategic superiority in

asiatimes.com

How China is expanding its anti-satellite arsenal

The Space Force chief doesn’t like the “curves” of how China is progressing vs the U.S.—and says he doesn’t have the funding to reverse it.

www.defenseone.com

to be fair, I was speaking of non-state actors, but I didn't make that clear. It's like saying joe-sixpack is gonna become a nuclear power...probly not...but a rogue nation, that's a different story.

 

[Phishfry]

Russia is jamming GPS from space

Russia is jamming GPS from space

spacenews.com

 

[resora]

The Pirate Bay plans low-orbit server drones to escape legal jurisdiction

Torrent site The Pirate Bay wants to "build something extraordinary" to keep its servers -- "the ones that re-direct your traffic to a secret location" -- safe from prying eyes

www.wired.com

The Pirate Bay plans low-orbit server drones to escape legal jurisdiction ​

 

[Crivens]

to be fair, I was speaking of non-state actors, but I didn't make that clear. It's like saying joe-sixpack is gonna become a nuclear power...probly not...but a rogue nation, that's a different story.

How about a satellite dish and a magnetron from a microwave, would that not produce (if aligned/tuned exactly) enough interference on the satellite to make problems? Having some of these operate is within reach for non-state actors. And as current events in the middle east show us, you don't need to be a nuclear power to royally screw up the world. It's like "I have the biggest GUNS!" "Yeah, but I make your heart medicine, so..."

Russia is jamming GPS from space

Russia is jamming GPS from space

spacenews.com

There is a veritasium episode about this on YT, which I highly recommed.

The Pirate Bay plans low-orbit server drones to escape legal jurisdiction

Torrent site The Pirate Bay wants to "build something extraordinary" to keep its servers -- "the ones that re-direct your traffic to a secret location" -- safe from prying eyes

www.wired.com

The Pirate Bay plans low-orbit server drones to escape legal jurisdiction ​

Not the onion?

When the RIAA took umbridge to the ISS not providing DVD players which only had one country code but were "hacked", my position was to go ahead and if they were so keen on the bad press - have the feds knock down the door.

 

[Phishfry]

Nothing to see here just move along.

https://www.livescience.com/space/space-exploration/chinas-top-secret-dragon-space-plane-just-released-another-unidentified-object-over-earth

 

[Phishfry]

Contrary to popular belief, space isn't cold.

Space also has it's own weather phenomenons to worry about.

Homepage | NOAA / NWS Space Weather Prediction Center

www.spaceweather.gov

 

[6502]

Pro: No risk of thiefs - save expenses for security guards, locks, etc.
Con: What happens if SSD or other hardware fails? Is there regular visits from personnel e.g. every 6-12 months?

 

[Crivens]

Is is an absurd idea, and it was most likely floated to boost the IPO. Now you can flush that. You will not be able to have all the HW rad-hardened to work in that place. You will need an awful lot of shielding, the DLR ran tests and research along that path back when I was in university. You would basically double the weight of the data center, the cooling problem is already mentioned. Does it have resources for a de-orbit? Will it impact? That are all things not taken into account when launching those things as they are hurting profits. When this gets done, someone will sell the future down the river.

 

[kent_dorfman766]

Is is an absurd idea, and it was most likely floated to boost the IPO.

Most space commercialization efforts are little more than corporate welfare that plays on the public's fascination with space and thus govt is willing the give out "stimulus" to companies that shouldn't get it, but it's good politics for the districts where the money is spent.

NASA has a habit of giving "small business awards" at a funding level that guarantees failure and multiplies that across several award recipient companies, instead of giving the total amount to a single tier-1 vendor and letting them choose the best subcontractors. From the perspective of return on investment it is a total scam.

 

[Crivens]

NASA has a habit of giving "small business awards" at a funding level that guarantees failure and multiplies that across several award recipient companies, instead of giving the total amount to a single tier-1 vendor and letting them choose the best subcontractors. From the perspective of return on investment it is a total scam.

On the other hand - who would these tier-1 vendors be and what would happen with any money you give them?
But a few bright folks with a good idea and a sudden wad of cash that will not go into exec compensation and shareholder value, they might get something so far off the ground that it proves worth giving more money to them. Just imagine if the wright brothers were forced to work with a big railway company in a project to speed up transportation.

We know what happens in such cases, when a big player in nuclear and fossil fuels is tasked with the prototype of a wind turbine. It was called "growian" and it failed, just like ordered. That pushed back the wind energy here in germany for decades, I might say. I'll stop here now because I start ranting...

 

[kent_dorfman766]

On the other hand - who would these tier-1 vendors be and what would happen with any money you give them?
But a few bright folks with a good idea and a sudden wad of cash that will not go into exec compensation and shareholder value, they might get something so far off the ground that it proves worth giving more money to them. Just imagine if the wright brothers were forced to work with a big railway company in a project to speed up transportation.

That is unfortunately how it is suppose to work, but in reality the small business grants often go to politically connected businessmen who take a large share of the award off the top, before hiring "cheap labor" to to the actual implementation...and those jobs are short term (usually for the duration of the project) In essence stacking the deck for failure because the focus becomes receiving the award money payments instead of producing the product. I've not only seen it, I've worked in it...and it's probably not fair to compare european business models with those here in the US. Our system of "corporate welfare" is a bit different than elsewhere.

 

[kent_dorfman766]

We know what happens in such cases, when a big player in nuclear and fossil fuels is tasked with the prototype of a wind turbine. It was called "growian" and it failed, just like ordered. That pushed back the wind energy here in germany for decades, I might say. I'll stop here now because I start ranting...

and isn't choosing a "consumable fuels" company stacking the deck for failure when developing a resource for sustainable energy production? Here in the states the smear campaign the coal industry did to the wind farms was incredible. They posted pictures of lightning struck wind turbines claiming they are a fire hazard, and way overstated the avian mortality problem of birds and bats flying into the turbines. They actually convinced most residents of a small redneck community (Central City in Cambria country PA) that the proposed wind farm was like having a toxic waste dump in their back yard. The issue I have with PA windfarms is that the benefits is not passed on to local electric customers but the energy is shipped to places like New York.

 

[blackbird9]

That is unfortunately how it is suppose to work, but in reality the small business grants often go to politically connected businessmen who take a large share of the award off the top, before hiring "cheap labor" to to the actual implementation...and those jobs are short term (usually for the duration of the project) In essence stacking the deck for failure because the focus becomes receiving the award money payments instead of producing the product. I've not only seen it, I've worked in it...and it's probably not fair to compare european business models with those here in the US. Our system of "corporate welfare" is a bit different than elsewhere.

That's a real shame, it makes me wonder how any real progress is ever made. It's hard to think which is worse, giving it out like that or giving it to a firm like Boeing. Pretty depressing.