Yeah for sure, but I think the overall picture is that bitcoin is needlessly energy expensive. A proof of stake system would work much better in terms of efficiency.
I was thinking it would be great to base a cryptocurrency onsomething like folding@home. Instead of just doing meaningless computation, you’d find something like new protein folds which is actually useful. And this kind of work can’t be automated easily, so the work to actually mine things would be a human effort. And if this led to somebody figuring out how to do this automatically that would be great too cause we’d be advancing science that way. Having a science coin would be really cool. :)
Was it not that a few years ago, a BOINC team had tried to fake results, to get more BOINC points (and reach higher in the performance charts)? And that (IIRC) was not even about earning Gridcoins as incentive to fold/crunch one such example.
Although Gridcoin pools have misused BOINC project as well: example.
So the main goals of distributed scientific computing would certainly get into some kind of danger there, if people fake WU results already without extra incentive (from the actual BOINC projects) to do so.
You could require manual verification of the results before the coin is validated. The whole point here is to limit the rate of coin creation after all. You could create a pool of coins using the known protein folds, and then new ones can trickle in slowly as they’re certified.
Perhaps, but that seems like a lot of extra resources required to make that work. And now already I have the gut feeling that some of the smaller projects are mostly there because the scientists could not get a supercomputer from their institute.
And even if the latter were not to be the case, I still think that most projects (certainly WCG) are primarily meant to be used philanthropically.
My main point here is that it would be better if the work needed to make crypto currency function actually had some purpose to it. It doesn’t need to be the specific example I gave, but I’m sure it’d be possible to come up with some meaningful computation that could power the underlying processing.
Yeah for sure, but I think the overall picture is that bitcoin is needlessly energy expensive. A proof of stake system would work much better in terms of efficiency.
I was thinking it would be great to base a cryptocurrency onsomething like folding@home. Instead of just doing meaningless computation, you’d find something like new protein folds which is actually useful. And this kind of work can’t be automated easily, so the work to actually mine things would be a human effort. And if this led to somebody figuring out how to do this automatically that would be great too cause we’d be advancing science that way. Having a science coin would be really cool. :)
Was it not that a few years ago, a BOINC team had tried to fake results, to get more BOINC points (and reach higher in the performance charts)? And that (IIRC) was not even about earning Gridcoins as incentive to fold/crunch one such example. Although Gridcoin pools have misused BOINC project as well: example.
So the main goals of distributed scientific computing would certainly get into some kind of danger there, if people fake WU results already without extra incentive (from the actual BOINC projects) to do so.
You could require manual verification of the results before the coin is validated. The whole point here is to limit the rate of coin creation after all. You could create a pool of coins using the known protein folds, and then new ones can trickle in slowly as they’re certified.
Perhaps, but that seems like a lot of extra resources required to make that work. And now already I have the gut feeling that some of the smaller projects are mostly there because the scientists could not get a supercomputer from their institute. And even if the latter were not to be the case, I still think that most projects (certainly WCG) are primarily meant to be used philanthropically.
My main point here is that it would be better if the work needed to make crypto currency function actually had some purpose to it. It doesn’t need to be the specific example I gave, but I’m sure it’d be possible to come up with some meaningful computation that could power the underlying processing.
I see. Indeed, in that way you are right.