5 ms·
Useful PoW is unlikely to be successful. You need a puzzle that’s difficult to compute AND very easy to verify the result. It also needs to be robust to many di
by leishman 4y ago
Useful PoW is unlikely to be successful. You need a puzzle that’s difficult to compute AND very easy to verify the result. It also needs to be robust to many different attack vectors.
- libertine 4y agoBack in the ol'piracy days the "proof of work" were in the form of your download/upload ratios, so essentially your storage and the volume of data your shared with others. So I don't think the PoW needs to be some CPU intensive task, it can be something of very low effort and value. But the problem becomes that yield must also be of low value -> and this should be ok! Except people seem to prefer the perception of high-risk/high-reward over some that's slow but builds up over time. So it's not about PoW, but instead people are looking for gambling opportunities.
- randomhodler84 4y agoBT upload counts were a client side value, they were trivial to fake and the tracker would trust the client. PoW cannot rely on client side trust, it must be universally verifiable. We have a few ways to do this, partial hash collisions, finding large primes, finding prime constellations, calculating cuckoo cycles… The primes seem to be valuable, alas nobody cared about primecoin or riecoin.