Hacker News new | ask | show | jobs
Help needed to build the colony on Moon, Mars and the Dyson Sphere (asgardia.space)
48 points by grandrew 1525 days ago
18 comments

Sadly, from a lot of experience in tech, one man projects that have "build self-aware AI" as a step in their roadmap have a tendency to underdeliver.
It s a crypto company, this increases the scam factor a tad.
We are not a crypto company. Moreover, being as clean and open as possible is the only way to win the battle of the good guys vs. the bad guys in supply chain.
Can you say more / link to more info about these "bad guys in supply chains"?
First, you can just google for "logistics is a shady business"

If you want to be more specific, you can google for "carrier haulage tax evasion" - this will be just the tip of the iceberg.

Supply chain is mixed with shady practices for tax evasion, bribery, laundering and all sorts of shady things at all levels that "compete" with actual process optimization. Combined with overall opposition to automation this incapacitates any effort to make the process efficient cuz it's hard, it "steals jobs" and not easy money.

Thanks.

For the first query I only found this:

https://www.reddit.com/r/logistics/comments/6mvptr/why_are_s...

For the second query I only found this, which discusses something that to my understanding isn't happening yet?

https://fiata.org/news/press-releases/2021/8-july-2021-2.htm...

Gotta say, I expected to find a lot more, compared to industries I know are shady (compare to googling "vpn shady"). I don't think I have a specific model of what shady practice are common in the logistics industry.

It's a non-crypto company (HyperC) advertising on a crypto organization (Asgardia)'s forum.
Instructions unclear. I’ve converted the entire Solar System to paper clips instead.
In the early days of Long Covid.. I played that game through 100 times.
From this moment on your goal is to play this game as many times as possible.
A tad ambitious, missing a few intermediary steps. Try solving logistics problems for say… a sandwich shop… before going on about “all of e-commerce” much less a Dyson sphere.

It’s not really obvious that this is at all necessary or the bottleneck. Or that running a planned economy ruled by AI is a goal to celebrate.

We're starting with small Amazon sellers. And it's just what I believe is right, the technology must take over the logistics and stop the supply chain gangs from thrashing the planet with waste. We aren't getting anywhere with the status quo.
Who knew an infinitely recursive self learning true AI was just a few python files away that haven't been updated in 5 months?
We're a tiny team that so far only got the confirmation that it works. The focus right now is to make the first batch of customers happy, and continue to spend effort on the metaplanner.

Btw, I advise you to take a look at the metaplanner's code. It's the definition of the planning task for the planner. So that the planner can learn to solve the planner's (it's own) task. It is leveraging the HyperC core that sits in a separate repository in the same github org.

I heard the approach was circulating in the AI planning academic circles but no one had put it together as many academically-uninteresting details needed to be cared for first. That's exactly what we did.

Hm, actually that's exactly how I'd expect the first infinitely recursive self learning true AI to look like. Until it rewrites itself in Rust for better performance.
Imagine Asgardia building a Dyson swarm, owning it, and then becoming a greedy landlord (swarmlord?). When you can’t afford the rent, the yeet you into space with a trebuchet.
Everyone who cannot contribute becomes polycarbonate material for the mirrors!
Stack ranking escalated quickly lol
HyperC is not affiliated with Asgardia aside from that one of the HyperC's founders being a long-term supporter
Starring Monty Python and the Flying Circus
How about something smaller scale like food waste. We have maybe 50-ish productive farming seasons left before soil is over taxed. If you can get locally grown food to people’s tables with minimum waste for a decent price that ensures everyone is fed that would be far more impressive than failing at Dyson spheres.

Maybe we can find a way to regenerate the soil while we’re at it.

That’s a hard supply chain issue and would have a much greater impact on our survivability.

Or good luck with building a self aware AI that can solve all of our problems or whatever.

That is not true about the soil. Areas in Europe have been farmed for hundreds or more years and still productive.
I am not sure where the GP gets the "50 harvests" from, but Europe didn't feed 8 billion people for hundreds of years. The Green Revolution aka widespread use of artificial fertilizer didn't start until the 1950s and 60s.
I believe OP is referring to the impacts of climate change rendering farming unsustainable.
I'm thinking he's talking about running out of phosphate [0]. Phosphate is a critical plant nutrient and mining it has been essential in increasing yields in the last century - it's literally sustained the population growth. The known supplies are running low, and it washes out of the soil on an annual basis so needs to be replenished to keep up yields — or we need to find other sustainable sources. The 50 years correlates with other (very) rough estimates of remaining supplies available to use for the current agricultural paradigm.

[0] https://www.discovermagazine.com/planet-earth/phosphorus-is-...

The overarching problem with the statement is soil is so different everywhere. And managed differently everywhere. You can't just make a blanket statement. If I was going to make a blanket statement I would be more concerned with erosion than nutrition because soil is an ecosystem that can be repaired. But if it is gone, then that's a different story.
True, and good point — erosion rates also fit the "50 years and it's gone" prediction
We're also depleting geologic/fossil water like the Ogallala Aquifer far faster than recharge.
"before the soil is over taxed"
> How about something smaller scale like food waste.

How bout smaller still… like maybe author starts with using ai to organizing their monthly budget, or whatever.

Nice intro video on what may be realistically required to build a Dyson sphere:

https://m.youtube.com/watch?v=pP44EPBMb8A

By Kurzgesagt, all their videos are excellent and informed by the literature.

There will of course never be a Dyson Sphere, or anything even vaguely reminiscent of it.

Instead, long before anything of such a scale would be conceivable, we will have hydrogen-boron fusion for unlimited energy, and moved industrial operations out to the Kuiper Belt, where the truly valuable commodity -- cold -- is abundant.

Suns and planets are for extreme rock-banging primitives.

Alternatively we could merge with the machines and all go virtual. But I am to believe that part of the humanity will remain explorers and will want to go beyond our star. That's where the physics kicks in. There simply isn't enough energy that can be mined and extracted from minerals.
Where do you hope to get your energy, beyond our star?

This is, presumably, after you have fused all of Neptune and Uranus. I guess you could start in on Saturn, or Planet X.

There are several options how to make harvested energy work for interstellar travel

1. Just beam it with a huge laser to the spacecraft (google for several proposed projects already in development and plans for launch) 2. Store it onboard the spacecraft and then use, with multiple options and different levels of today's technology feasibility like creating a small spinning black hole or creation of exotic matter for warp drive

Super cool and insane ambition, but I don't get the part about the Dyson sphere. I don't see that as a guaranteed requirement to accomplish the goals set out, and it's going to be insanely controversial. That part seems like a solution looking for a problem, to be built mostly because it's 'cool'.

But I don't want to rain on the parade of long-term, crazy-ambition in space, so I think this is pretty awesome overall. But planning unilaterally to build a dyson sphere will definitely make you a lot of enemies on Earth.

Otherwise, godspeed.

Thank you very much for support!
This seems to be some kind of planning tool that uses Python and a database to do something logistics related.
Yes, behind the scenes it lowers the high-level logic of Python to predicate logic, uses RETE on that predicate logic to do the fastest possible JOIN operations on the tables, and uses machine learning to extract heuristics from the task and data to enable feasible monte-carlo search for valid plans.

All that means you input the rules for the "game", and it finds the way to reach the goal, no matter how complex the interactions or data are inside.

Andrew here, the founder of HyperC. Thank you very much for your feedback, and especially for your attention and time to look at the note.

Our current and today's focus is, of course, not to send the sattelites to the sun to build the Dyson sphere but rather to automate sending average commodities from China to Amazon FBA. Everything else comes after.

And you're right, the supply chain mess isn't caused by the absense of tech (although this does contribute) but rather by the "dark" forces in the logistics world that would prefer to keep things messy and hard to trace. The entire logistics thing has historically always been a shady business. But big data is coming for the dark guys.

So the immediate plan is to build the model of the "black box" of logistics by analyzing the data we can actually collect, then use the combination of technology and HyperC's reputation-leveraged optimal financing offers to beat the shady schemes out of the market. That itself could be overly ambitious, but this is what I believe is right.

Appears to be a scammer scamming a scam.
Well, nobody can fault them for not being ambitious enough.
Yet, I do.
Why? Don't we have enough unambitious things to go around? I mean - as long as the project owner realizes the probability of failure is high.....
Not ambitious enough.
First enhance on-earth supply chain efficiency. Then worry about space. Purchasing and sales departments really should be eviscerated and replaced with semi-autonomous systems in most cases. Get that done and you'll have more than enough money and cred to tackle space. My ideas @ https://raw.githubusercontent.com/globalcitizen/ifex-protoco...
> First enhance on-earth supply chain efficiency. Then worry about space.

These are not mutually exclusive. Moving heavy industry to automated facilities in orbit is not unrealistic in the coming decades. It does require massive investment in space tech.

My thoughts :) Get the mess solved here first. As a comment on your ideas - there are extremely powerful forces that would oppose any openness in logistics as it's filled with shady gangs. We need to act together to beat them off the market.
What are some supply chain inefficiencies you've identified?
A few... lack of standard transaction identifiers, fixed hours of business with bespoke holiday schedules, use of humans, use of telephones, physical travel to trade shows, printed literature, cold calling/emailing, manual price checks, incapacity to provide actual volume or delivery dates prior to purchase, manual pricing, non-transparent pricing, manual provisioning and re-entry of client and supplier information such as corporate identity, tax, banking and addresses; bait and switch, tedious inefficient reputation systems, lack of machine-readable digital specifications, lack of 3D models, lack of (translated) documentation, material/regulatory/supplier certificates and associated documentation that is impossible or tedious to verify... one could go on.
Among these, "use of humans" stands out.
Supply chain is a black box with people who are aggressively not interested in transparency. So the approach is to model the black box as good as we can, and start cutting away the pieces with automation.
“ Build the MVP prototype of commercially usable, no-compromise, pure A.I. engine that learns to solve any formally defined task”

So, PROLOG from 30 years ago then.

Yes, PROLOG with modern Python syntax and mostly imperative flow, commercial database experience instead of "build everything yourself", and ML-based heuristics that are learned on the fly instead of whatever hardcoded logic the particular interpreter had.

This is sometimes referred to as "Automated planning and scheduling" or simply "AI Planning" in academia.

I feel the future is in the stars, but it needs to be a future for everyone. This doesn't feel like the right group to get behind.
I can spare $6 for this project. What do my children get in return?
Only by reading the plan, he reminds me of an early Elon Musk. It has the first Paypal then Space X vibe. I wonder how many other people have similar dreams, but can't deliver cause their parents aren't rich.

Edit: added middle sentence

Just because "reusable rockets" got laughed at by the existing space industry doesn't mean every laughable idea is secretly feasible.

"My new database will make building a Dyson sphere easy" is a slightly... loftier claim than "we can probably land rockets and reuse them".

I see what you mean, but I wasn't referring to Space X accomplishments but its ambitions.

The "we can probably land rockets and reuse them" is what it delivered so far, the claimed ambitions (laughable or not) are about making humanity interplanetary.

The number of steps between "I wrote a quirky database" and "I built a Dyson sphere" are substantially more numerous than there are between "I got reusable rocketry working cost-effectively" and "I used those rockets to put people on Mars".
While I agree with you on the guesstimate, surely you can't exactly know the number of steps in question.
Ambitions are free. Capability is expensive. Achievements are valuable. Ambitions do not predict achievement.
>just because "reusable rockets" got laughed at by the existing space industry …

Were they laughed at? After all, the Space Shuttle was around for decades. It turned out that reusable rockets were not cost effective nor particularly safe. Space X has yet to show reusability is viable.

> Space X has yet to show reusability is viable.

This is a bizarre assertion. It's clearly viable; no other American launcher can currently carry crew, they gobbled up 60% of the global commercial launch market by 2017, and they're undercutting everyone on price very successfully.

I'd say what this all proves is that a free market and private sector is much better at those things than a communist'y government-run project.
I want to send something to a carbonaceous chondrite asteroid that can extract the volatiles and turn the coal-like substances and stony substances into large plastic (Kapton!) and metal films. These could be used for solar collectors, solar sails, sunshades.

It pays for itself by manufacturing space sunshades to the Earth-Sun L1 point that get their own their own power.

The "something" is probably a factory that builds a factory that builds the factory, at least twice you face a situation like building a ship in a bottle except you are in the bottle. It probably imports microcontrollers by the ton from the Earth. It's an interesting question: do you send people who can interact with the system but need a place to live or do you run it all by remote control and face 60 minute or more round trip delay requiring that the thing make mistakes and recover autonomously.

Kapton requires a great deal of hydrogen to make.

L1 sunshades would cement global catastrophe, as unrestricted accumulation of CO2 would further acidify ocean shallows eliminate the base of the ocean food chain.

My guess is that some CC asteroids may have generous amounts of hydrogen. Any reasonable plastic has a lot of hydrogen in it, kapton is harder to make than PE and PET and requires nitrogen which is more of an unknown in terms of a availability.

I think some asteroids might be rather gassy and might require devolatilization before doing anything else and that is awkward because you are going to be at your bottom in terms of equipment inventory, particularly storage tanks.

As for ocean acidification it is a real problem but the severity of what we’re up against means we shouldn’t leave solar geoengineering off the table. The fact that one party could do something very dangerous in fact might break the ‘collective action’ problem.

I presented on this idea at a seminar on geoengineering and I was accused of trying to build a Dyson sphere, I said "no, there is not enough mass in the asteroid belt to cover the Sun but you can build something pretty big, possibly a human habitat much "bigger" than the Earth in land area."

By then we will have pB fusion, and have no need of solar arrays. The Kuiper belt will then be more attractive than the asteroids. There is plenty of frozen nitrogen there, and abundant precious cold.

Heat is easy to get, fundamentally hard to shed.

D + D probably wins in conditions out there. The T and He3 that it breeds will be handy.

My guess is that anybody who gets out that far loses interest in dry inner solar system planets completely and isn't going to be interested in exporting products back.

By then they might be most interested in getting as far as possible from any nearby magnetars and other existential threats.

Although other people might be the most immediate threat right up until the moment a magnetar sterilizes the vicinity.

I don't know whether we are certain where the nearest one is.