4 ms·
Did anyone ever think of creating a proof-of-work system that would allocate some CPU time to do something useful for humanity? Maybe for medical research or so
by olalonde 16y ago
Did anyone ever think of creating a proof-of-work system that would allocate some CPU time to do something useful for humanity? Maybe for medical research or something... Could it even work technically?
- JoachimSchipper 16y agoYes, see "bread pudding protocol". Or reCAPTCHA, for a less-cryptographic take on the same problem.
- swah 16y agoI suppose its not worth distributing this task across the network if each node will only have .5 second computing time each time.
- JonnieCache 16y agoIt is if that work is only a byproduct of a system such as described in the article. It's essentially "free" computation, even if it is only in tiny chunks. If the system became widespread, it would quickly add up. And why not?
- JoachimSchipper 16y agoFor most applications of a proof-of-work system, you need to be able to verify the results quickly,. In addition, while the aggregate CPU power is pretty hefty, communication overhead is huge. There are not many useful problems that fit these constraints.
- rmc 16y agoThere are a few systems like this already, e.g. SETI@home (try to analyze signals to find aliens) or Folding@home (model protein folding to help with cancer (?) research)
- pavel_lishin 16y agoBut that's not really a proof of work system, that just uses your computer's idle time to work on something.