First: Any process in the natural world can be replicated with human technology. This assumption makes sense in light of the fact that humans have generally been successful at copying or co-opting nature.
Human technology obviously cannot replicate neutron stars or black holes in the near or even medium term future. So this assumption seems very exaggerated.
A limited subset of the natural world can be plausibly replicated with human technology.
They assume:
Solar captors with an efficiency of ⅓.
Only 1⁄10 of the energy will be used to propel material into space. The rest will be used for mining, reprocessing material, or simply lost.
It takes five years to process the material into solar captors and place them into the correct orbit.
Only half of Mercury’s material will be used to construct the captors.
Under these assumptions, the power available will increase exponentially every 5 year cycle. Mercury will be disassembled in 31 years, with most of the mass harvested in the last four years.
They are also assuming coordination and communication costs between each unit are non-exponential. What’s more likely is that coordination and communication problems will increase exponentially rendering vast swarms unfeasible.
Coordination is only exponential if most units have to coordinate with most other units rather than following more localized coordination. Huge flocks of birds and swarms of drones coordinate just fine. They only need to be aware of local variation and broad trends.
Which has obvious parallels to possible failure states of robotic swarms.
Large quadcopter type drones also don’t ‘coordinate just fine’, you can look into it yourself what the requirements are for hosting those huge shows. The requirements are not simple or easily reducible to a few paragraphs of instructions.
Intel even released a promo video touting, in 2016, how a challenge of 500 drones in simultaneous operation was groundbreaking and incredible even with their state of the art technology.
“The difference between 100 and 500 is mind-blowing.”—Natalie Cheung, Light Show Business Lead, Intel
You need to do a bit more research if you were unaware of the complexities.
Thanks for the links! My intuition was that space is big enough that global coordination isn’t always needed to avoid basic failures like collisions but I definitely do need to do more reading/thinking/modeling to figure out how valid that intuition is.
Does anyone have a link handy related to complexity of coordinating Earth’s satellites?
Earth has fewer than 10^4 satellites. Taking apart Mercury over year-like timescales would mean enough equipment to remove a thousand tonnes of rock and metal per hour in every square metre.
I’m sure a superintelligence can find a way to do that, but I’m pretty sure that the optimal solution won’t be a flock of duck-sized probes nibbling it to death, no matter how numerous.
Human technology obviously cannot replicate neutron stars or black holes in the near or even medium term future. So this assumption seems very exaggerated.
A limited subset of the natural world can be plausibly replicated with human technology.
They are also assuming coordination and communication costs between each unit are non-exponential. What’s more likely is that coordination and communication problems will increase exponentially rendering vast swarms unfeasible.
Coordination is only exponential if most units have to coordinate with most other units rather than following more localized coordination. Huge flocks of birds and swarms of drones coordinate just fine. They only need to be aware of local variation and broad trends.
Huge flocks of birds also occasionally dive straight into the ground, injuring many of them. Such as https://www.boston.com/news/world-news/2022/02/17/mexico-viral-video-explained-100s-of-birds-diving-to-ground/
Which has obvious parallels to possible failure states of robotic swarms.
Large quadcopter type drones also don’t ‘coordinate just fine’, you can look into it yourself what the requirements are for hosting those huge shows. The requirements are not simple or easily reducible to a few paragraphs of instructions.
Intel even released a promo video touting, in 2016, how a challenge of 500 drones in simultaneous operation was groundbreaking and incredible even with their state of the art technology.
“The difference between 100 and 500 is mind-blowing.”—Natalie Cheung, Light Show Business Lead, Intel
You need to do a bit more research if you were unaware of the complexities.
Thanks for the links! My intuition was that space is big enough that global coordination isn’t always needed to avoid basic failures like collisions but I definitely do need to do more reading/thinking/modeling to figure out how valid that intuition is.
Does anyone have a link handy related to complexity of coordinating Earth’s satellites?
Earth has fewer than 10^4 satellites. Taking apart Mercury over year-like timescales would mean enough equipment to remove a thousand tonnes of rock and metal per hour in every square metre.
I’m sure a superintelligence can find a way to do that, but I’m pretty sure that the optimal solution won’t be a flock of duck-sized probes nibbling it to death, no matter how numerous.