Hi there,<p>I had an interesting idea on how to perform useful compute with proof of work. We could wrap a problem using Homomorphic Encryption techniques. Each node gets a different problem to solve. If the solution matches a certain pattern, they can write the next block (similar to already existing methods).<p>The key difference is that the research institution interested in submitting a job would pay a fee, and that fee is split among all miners that worked in solving the puzzle - creating a second, STEADY source of revenue to GPU miners!<p>This would put a massive number of GPUs out there into service of science, while increasing the revenue of miners. Efficiency would improve overtime as FHE algorithms evolve. ASIC miners would be very welcome.<p>Please criticize.
I don't get the homomorphic encryption part... so it's still proof of work, just got science. many such projects exist, mainly the "X@Home" projects and more.
I’m not sure what the FHE part of your proposal is trying to solve but to address the main point, the “Work” in Proof of Work needs to be useless apart from securing the network.<p><a href="https://bitcoin.stackexchange.com/questions/97030/beginners-question-why-must-proof-of-work-be-useless" rel="nofollow">https://bitcoin.stackexchange.com/questions/97030/beginners-...</a>