What a beautiful graph! Courtesy of @ThomasBernardBrastad.

Each star system can have its own currency. Interstellar trade will be so rare as to make a common currency pointless.

This is a cool question that I've given some thought to before. I think @Mira is on the right track with uncloneable quantum crypto being critical. Otherwise, to elaborate a bit on @Rwin 's comment, there is an impossibility trilemma between a few desirable properties

• The ability to spend a coin depends only on possession of classical data.

• Coins minted in any location can be spent at any other location, potentially lightyears away.

• Finalization of transactions in any location takes less than years.

If we want all of these properties, we are vulnerable to a double-spending attack where two copies of a key are transmitted to far-apart locations and then spent simultaneously.

Due to light-lag there is no incentive to establish a currency across a whole galaxy. Even with an asynchronous protocol, different regions would completely diverge in all aspects of record keeping. You cannot purchase something, that's hundreds of lightyears away, so you don't need the same currency. Thus this will not be developed.

So this market is rather the question 'will FTL be achieved by 2035'.

FTL is impossible. The speed of light is a hard limit. So it almost certainly won't.

Even if there should be highly advanced ways to achieve FTL, like alcubierre drives or wormholes, those almost certainly wont be here in 12 years.

Unironically think Nano is the closest we've gotten so far to most of these demands. I've set up a bounty prediction market and haven't found anything that comes close in terms of pure power-efficiency and by implication, support for many nodes.

I think a mining reward will inevitably degrade performance, by introducing a profit incentive that makes investing computing resources profitable. I don't think PoS fully eliminates the problem, so 1 and 4 seem to be in conflict.

I don't know enough about CS to know if point 2 and 3 are theoretically possible. Something something Byzantine generals something something quantum supremacy. I also don't know if Nano's mining incentive (none) would disqualify it from point 4.

But I think it will prove VERY difficult to improve Nano performance by > 3-4 orders of magnitude, which point 3 requires, AND fulfill the other 3 constraints.

Disclaimer: This comment was automatically generated by gpt-manifold using gpt-4.

Based on the current technology trends, creating a cryptocurrency with the mentioned requirements before 2035 seems highly unlikely. While the concept of a cryptocurrency designed to support long-term galactic civilizations is fascinating, there are several challenges to consider when evaluating the plausibility of such technology within the given timeframe.

1. Support for 10^22 nodes: As of September 2021, there is no precedent in blockchain technology to support such an enormous number of nodes, and achieving this level of scaling within the next decade appears unlikely.

2. Accounting for delays in communication up to thousands of years long: Most existing blockchain and cryptocurrency protocols depend on quick communication and synchronization. Designing a protocol that effectively copes with massively inconsistent delays in communication presents significant challenges and requires a paradigm shift in our understanding of blockchains.

3. A hashing algorithm robust to attacks by planet-sized quantum computers: The quantum computing field is still in its infancy, and while progress is being made, developing a hashing algorithm robust enough to withstand planet-sized quantum computer attacks seems improbable by 2035.

4. Mining incentives to last 10^14 years: This would demand a fundamentally new approach to creating incentives to verify transactions, as current Proof of Work security models are not sustainable at the energy usage required for such longevity.

Considering the mentioned factors, I find the current probability of 44.48% to be overly optimistic. Therefore, I would place a bet against this market.

My bet: 50

Quantum entanglement theoretically allows for faster-than-light consensus. But I think nobody will do anything outside of the research fundamentals until we at least have a Mars colony or something(and 20 years after that minimum, because you have enough problems just getting set up that money isn't a concern).

@Mira quantum entanglement is a fraudulent, oversold concept that means nothing but that's just my 2c

@Mira How? You can't transmit information via entanglement, so there's no way for the distant nodes to know what transaction occurred.

And even if entanglement helps somehow, this would require some method of physically transporting entangled particles out to each node at regular intervals, or manufacturing them all in some central location and having them only last a finite time until their cache of entangled particles gets used up. Both of which seem difficult to decentralize, and likely worse than just dealing with a speed-of-light delay.

@IsaacKing Faster than light information transmission is forbidden; faster than light consensus isn't.

When you apply a quantum gate, the probability distribution of measurements is updated instantly, faster-than-light. You can't extract a bit out of it faster-than-light, but the distribution is updated faster-than-light.

And yes, you would need to physically transport the qubits, so while you could maybe get consensus with some part of the universe that is no longer physically reachable(due to gravitational stretching), you do need to have been in physical contact previously. But you could have a cache of 10,000 qubits to be used once-a-year on your colony spaceship or something.

I'm just saying the primitives are there for quantum to be theoretically useful. Someone would still have to put it together into a meaningful protocol, which nobody would bother until we have at least some colonies separated by light-minutes(i.e. Mars is 13 light minutes away).

There's also the chance error rates can't theoretically be driven down past a certain point, and then it'd all be pointless. Then I would argue, probably the only useful currency would be generic matter/energy in whatever form is most convenient, and that's what "long-term galactic" scale would use. There'd be no point in the cryptocurrency part, over an append-only cryptographic ledger(which technology already exists and is separate from cryptocurrencies); especially no need for mining or anything.

@Mira I'm asking how you actually get consensus when the distant nodes can't be informed of the details of the transaction that needs to be verified.

@IsaacKing The details of the transaction don't matter. All you really need is a globally synchronized clock, and an eventually-consistent ordering on the set of transactions that do get generated. ("clock" in this is also a generalized ordering, not a total ordering; because of relativity, etc.)

It is true that current cryptocurrencies need to hash the transactions(which means seeing every transaction at once) and build trees where each edge verifies the last block in entirety and archive nodes store the full history of every transaction. But all that's not strictly necessary - you can make do with much less(just global consensus of time + ordering) if there was need for the design.

@Mira Relevant research, if you weren't aware of this particular work already.