I really like that comparison with the gold to raw ore ratios as a frame of reference, but isn't mining hashes in general just a completely arbitrary selection for a definition of work and wouldn't harnessing some more useful proveable activity tied to a lottery system be better ?
Even all the Miners locking cash in a single account with a constant reward function could be put to some real use and still generate the same return outcome of new minted coin to investment/work; just with fewer steps and without all the electric overhead.
> wouldn't harnessing some more useful proveable activity tied to a lottery system be better
Yes, but such a puzzle needs to be
- easy to generate at a certain difficulty
- difficult to solve
- easy to verify and
- the solution needs to be dependent on previous solution (so you can't silently pre-solve them).
Solving (stupid) hashes so far is the only activity that checks these requirements.
Another option might be to design a certain hash function such that when the hardware miners dedicated to solve them become obsolete (i.e. too expensive to be used for crypto mining), they can be easily repurposed for another problem set.
Thank you for the constructive comment but I think you are approaching it from a cryptographical computer science angle which even if solved still results in work as an empty puzzle of floating point mips and what I cannot yet rule out is a reimagined scope for real provable work that also has some positive social value like say creation of 1kwh of renewable energy placed on the grid. Suspend disbelief for a moment as to the many actual mountains you would need to move and consider that such a scheme is hypothetically possible on paper under laboratory conditions given an infinite capacity of unrestricted action. Puzzle solving in a sandbox might have been fine back in the OG whitepaper hypothetical days of Bitcoin et al but the stakes are sufficiently high now that considerable thinktank type effort could reasonably be spent on this problem. My reductio ad absurdum argument of all miners proveably locking and unlocking real money as work in to and out of a single bank account tied to a snapshot lottery mimicking the existing mint award outcomes seems at least less impossible and the mining account's interest can at least then be put to a real world good use.
> wouldn't harnessing some more useful proveable activity tied to a lottery system be better ?
Can you devise such a provable activity that has a difficulty that can be adjusted to the levels Bitcoin requires and is dependent on the last block's hash?
Even all the Miners locking cash in a single account with a constant reward function could be put to some real use and still generate the same return outcome of new minted coin to investment/work; just with fewer steps and without all the electric overhead.