Orbiting, Relativity and Mind-effects

[some bloke]

It would be wholly exaggerative for me to say that I have this story "in the works", so "I have an idea for a story" would be far more fitting!

This story involves a method of terraforming which requires manual control of robotic drones from a high-speed orbital ring, made from unobtainium and handwavium so that it's strong enough to survive the forces involved, and uses a "plot drive" to prevent people from being turned to jam by the acceleration. I have a few soft-scifi answers t othe issues of putting people in a ring around a planet and spinning it up to relativistic speeds, so that's not an issue.

Now, the concept is that there are people in this ring who are controlling the robots on the ground, which are doing the terraforming work. My theory is that, as data is sent to and fro mthe ring, the people on the ring will perceive the surface of the planet as traveling faster through time - in the same way that an astronaut in the ISS experiences a slight bit of time dilation, this would make the effect much more significant.

So a person watching a "live feed" of the planet would see everything moving a lot faster. For the sake of TV, the signals might be sent up from the planet every hour, and arrive every second, forming a sped-up report of what is going on. I hope that I have got this the right way around - I know that the effects of gravity are such that clocks closer to a gravity source tick slower than those further away (which is accounted for by GPS Satellites), and that the opposite is true for moving fast.The next thing to consider is that, if the planet is sending a pulse every second, and these pulses are arriving at a rate of 3600 per second on the station (keeping the 1 hour = 1 second ratio for the theory), then if a computer were sped up to function 3,600x faster than normal, that computer could monitor the events on the surface as if it were in the same timeframe as them.

Now replace that computer with a brain. If a person were subjected to some science and the like which causes their minds to operate faster than normal - say, 3600x faster - then their perception of time (only how it seems to them, not how it actually affects them) would be 3600x slower. As such, this mind could feasibly be connected to something on the surface of the planet - one of these robot drones - and control it in real time. The robot would send 1 signal per second, which would be received by the ring 3600 times more frequently, would be answered 3600x more quickly and return as if it were answered immediately. As such, a sped up mind could be lying in stasis in the ring and perceive 1200 full 24 hour days of toil on the surface in control of a robot drone, whilst their body only ages 8 hours.

I am wondering how badly I have butchered relativity here! Aside from the exact mechanical how's of getting a ring spun up to silly speeds without turning it into a device from separating meat from bones, if this were done, would it work like I'm imagining?

 

[Rive]

I am wondering how badly I have butchered relativity here!

Up to you. How does the required antigravity-like drive (which preventing the crew becoming thin organic film on the floor) affects time dilation? No such device exists in relativity yet.

If a person were subjected to some science and the like which causes their minds to operate faster than normal ... perceive 1200 full 24 hour days

I think this would be the more interesting point. If it's about 'complete' mind, then it's not science, but a futuristic labour camp producing mentally ill prisoners at rapid speed.

Maybe you could do something with partial conscience-simulations running on hardware, instead?
With periodic sync and review from the 'original'? That would be far less unrealistic/cruel.
Also, it would spare an awful lot of handwavium about that mind-turbo process too.

 

[some bloke]

Up to you. How does the required antigravity-like drive (which preventing the crew becoming thin organic film on the floor) affects time dilation? No such device exists in relativity yet.I think this would be the more interesting point. If it's about 'complete' mind, then it's not science, but a futuristic labour camp producing mentally ill prisoners at rapid speed.

Maybe you could do something with partial conscience-simulations running on hardware, instead?
With periodic sync and review from the 'original'? That would be far less unrealistic/cruel.
Also, it would spare an awful lot of handwavium about that mind-turbo process too.

The counter-field for preventing the people from becoming primordial soup is a powerful magnetic field, based on the fact that we can currently levitate objects in a magnetic field. The field would be effectively directed to levitate the people to the middle of the ring, which they are stopped from doing by the floor of the ring, and thus it acts like gravity. I honestly don't know if Magnetic acceleration has the same effects as gravitational on relativity, that is possibly a good topic to ask about... I will probably have some explanations about the iron in the peoples bodies being replaced with non-ferrous artificial alternatives to prevent it all being ripped from peoples bodies!

I like the idea of transferring a consciousness to the ground and syncing it at regular intervals, I think I can pick that up and run with it... It also suits the general plot I was loosely forming around the concept. Thanks!

 

[PeroK]

Surely the terra-forming robots can do the job without continuous human instruction?

 

[some bloke]

Surely the terra-forming robots can do the job without continuous human instruction?

Only if I were to write it as such ;)

The reasoning that this doesn't work is that the AI systems regularly failed, performed dangerous and incorrect work, and couldn't adequately anticipate results on entirely uncharted and foreign planets. The issue then becomes this - how do you use someone who's expertise is in the correct methods for terraforming and building a world, when their lifespan is significantly shorter than how long it takes to terraform a planet?

It's also worth noting that the technology for making materials and force fields to make this technology work is a different one from creating AI that can autonomously terraform a planet. There's also the possibility that AI is a banned concept, due to skynet-like results (potentially being why new planets are needed)!

 

[PeroK]

The reasoning that this doesn't work is that the AI systems regularly failed, performed dangerous and incorrect work, and couldn't adequately anticipate results on entirely uncharted and foreign planets.

That's the humans who operate like that, surely.

 

[some bloke]

That's the humans who operate like that, surely.

I have seen humans perform a perfect round of guitar hero on maximum difficulty, because of their ability to learn and apply their skills.

I have also seen videos of AI robots destroying a door because it was too complicated for them to work out how to open it.

Whilst humans are prone to make mistakes, it's also important to note that the majority of humans are adaptive enough to know that they made one, and to try different ways of achieving the desired results without unwanted side effects. an AI that only knows that they are supposed to be getting from outside to inside has no concept that breaking down the door is a mistake, so will continue to break down every door it comes across because it knows this works and achieves the desired result.

An AI is only ever as good as it's programming, after all!

 

[PeroK]

An AI is only ever as good as it's programming, after all!

By your argument, no chess computer could ever beat the programmer who programmed it.

For example, I recently wrote a program to solve Sudoku puzzles. It can solve any puzzle in less than a second.

In principle, I can follow my own algorithm and solve any puzzle. Except that I get lost in trying to maintain ten sheets of paper with all the data and crossing off the right numbers. In other words, I wrote an algorithm that a computer can execute but I can't. I tried it! It's too complicated for me.

An AI terra-former could figure out the local variations in seconds; whereas, a human would take years to gain the experience. That's the science-fiction, surely.

 

[Rive]

we can currently levitate objects in a magnetic field

I don't think that that would do. We can do that in Earth gravity, in a limited space, with absurdly strong magnetic field, with really costly equipment. To scale that up is just less 'realistic' (in a sci-fi setup) than the good old antigravity drive.
Especially, since the setup is about a moving object. Are you planning a solar system sized generator?
Something like an artificial magnetar?

Ps.: with an antigravity drive you can spare the ring too, actually.

 

[some bloke]

By your argument, no chess computer could ever beat the programmer who programmed it.

For example, I recently wrote a program to solve Sudoku puzzles. It can solve any puzzle in less than a second.

In principle, I can follow my own algorithm and solve any puzzle. Except that I get lost in trying to maintain ten sheets of paper with all the data and crossing off the right numbers. In other words, I wrote an algorithm that a computer can execute but I can't. I tried it! It's too complicated for me.

An AI terra-former could figure out the local variations in seconds; whereas, a human would take years to gain the experience. That's the science-fiction, surely.

Okay, so let's take your soduku-bot as an example. Firstly, kudos for that, programming is awesome and I want to be clear I'm not belittling it!

The Soduku-bot can solve soduku puzzles very quickly, but I wonder what it would do if there was the number "10" in one of those boxes as supplied? Or if one of the boxes were missing, or if there was an extra box? I am also curious as to the summary of how it does it - does it work it out methodically, or does it attempt every combination until one fits?

Sodoku is a very strictly regimented game in which there are a finite number of rules. Three, if I remember correctly - 1-9 in each row, 1-9 in each column, 1-9 in each box. And, as you said, it required more information to solve it than you, as a human, could keep track of. And yet, without using 10 sheets of paper and getting lost, people solve these puzzles all the time.

Now imagine how colossal a program it would take to terraform a planet, where you have all the laws of physics to anticipate and react to. A human can see that a rock might need to be moved to make space for something, and quickly use their experience and logical deduction to decide that they need a crane, and to do that they need access for the crane, and so on. An AI would see the rock, and probably become paralyzed by trying to establish every possible means by which to move it, and how to achieve it, working out everything to try and find the most efficient path to the desired result. What humans take for granted as "the obvious choice" is not a given for AI.

I don't think that that would do. We can do that in Earth gravity, in a limited space, with absurdly strong magnetic field, with really costly equipment. To scale that up is just less 'realistic' (in a sci-fi setup) than the good old antigravity drive.
Especially, since the setup is about a moving object. Are you planning a solar system sized generator?
Something like an artificial magnetar?

Antigravity drives are notoriously poorly explained - who's to say that an antigravity drive isn't a magnetic field accelerating everything diamagnetic in a downwards direction, and that there isn't a process to make traditionally magnetic things diamagnetic? "Gravity" is just an acceleration downwards, and it is widely accepted that if you put a scientist in a box and accelerated them through space at a constant acceleration of 9.81ms-2, then that scientist would be quite cross that scientist would not be able to distinguish between that acceleration and gravity. So if you're accelerated downwards by Gravity, Magnetism, Tachyons or anything, it would function as antigravity!

 

[PeroK]

It's an interesting for a sci-fi writer not to be able to imagine AI development!

 

[some bloke]

It's an interesting for a sci-fi writer not to be able to imagine AI development!

Oh I'm able to imagine it, but I am also able to imagine a lot of scenarios - this being one of them - where AI is not the best solution!

Here's an off-the-top-of-my-head example:

The AI is given a specific set of conditions for the world to conform to - Oxygen levels of the air, amount of food growing, UV light coming in through the atmosphere, etc. - and they let it terraform the planet.

People turn up and every condition has been met, but the ground is thick swampland. Now they add the requirement for solid ground to stand on.

Next time they arrive, and the ground is solid, but now the air is full of dust, so they add the requirement for the air to be dust-free.

Now they wait for a long time before finding the AI is in a logic loop, where they are trying to eliminate all dust from the air but at the same time have pollinating plants for food, so they have to add a tolerance for dust in the air.

Then they get back to find large extractors everywhere noisily sucking in the dust, so they add a requirement for noise levels, and so forth.

The chances of an AI missing a specific instruction and carrying on blindly because it is lacking the knowledge of its task as a whole is significantly higher than that of a human brain doing so. Most humans would only do so to prove a point or because it's "more than their jobs worth".

An AI on wheels might get permanently stuck at a crossing because there's no little green robot to tell it to cross, only a little green man. A human will easily negotiate this through sound reasoning and what is obvious to us!

 

[PeroK]

You're not going to be buying a self-driving car, I assume?

The I in AI stands for "intelligence". Something that generally computers do not yet have. There are, however, self-learning computers now. How far and quickly these will develop is anyone's guess. Personally, I think that development in this area is significantly more achievable than the heavy engineering of relativistic space travel.

In other words, serious AI that completely outperforms human thought will predate the heavy space travel and terraforming engineering technology.

That said, most sci-fi I've seen is very much the other way round.

 

[some bloke]

You're not going to be buying a self-driving car, I assume?

The I in AI stands for "intelligence". Something that generally computers do not yet have. There are, however, self-learning computers now. How far and quickly these will develop is anyone's guess. Personally, I think that development in this area is significantly more achievable than the heavy engineering of relativistic space travel.

In other words, serious AI that completely outperforms human thought will predate the heavy space travel and terraforming engineering technology.

That said, most sci-fi I've seen is very much the other way round.

You assume correctly! Though that's more because I'm a firm believer that people should do things for themselves. I'm actually against the self-cancelling indicator because it makes people lazy, meaning when it doesn't work, people spend minutes or even hours driving with their indicator going because they've learned to rely too heavily on it turning off for them.

Intelligence is one thing, but intelligence divorced from empathy breeds psychopathy. Perhaps AI would be better directed into first teaching computers to feel, and then to think. That is, after all, how life started on the planet! Senses came first, and then the intelligence.

It's also worth considering that if the Terraforming AI were implemented, would people be happy to trust it to tick along for hundreds of years unsupervised? Chances are the end result wouldn't be exactly what people had in mind...

Either way, for the purposes of my story, I might have AI having been outlawed for any number of reasons. Or scrapped as requiring too much upkeep to stop it from planting exclusively brussell sprouts because someone on the internet said "you can never have too many sprouts". Depends on how silly a direction I take this in... XD