Educational proof of work

[asse.sauga]

Hello guys, wanted to share and discuss with you one idea. We have really lot of very poor people around the world. But they have lot of free time and they are quite unwise too generally. Most of the people from Africa and Asia. What if we could figure out a way, how to use their time and resource for making work to verify proofs. And so that the work what they are doing are actually educational one. For example mathematical tasks which are logic and easy to verify. And we can motivate them with small amount of cryptocurrency for every successful task. Problem is, how to put those small tasks meant for educate people through cryptographical solutions to actual proof of work, what we could use to verify different transactions.

With this kind of solution we could solve many problems same time - educational program for poor people and countries, earning money opportunity for poor people and much more green proof of work. All they need just smartphone, internet and time.

What do you think? Do-able?

[tyz]

...But they have lot of free time and they are quite unwise too generally...

I stopped reading after this sentence. Have you ever met poor people? Have you seen that they have a lot of time? Most of the poor people work very hard to survive. Probably not always for money but for food and clothes. They usually work harder than the most of the people in the first world.

[fluffypony]

Asse my friend! How is LHV? I'm going to be in Europe again next month, if you can you must definitely join us in Germany for the meetup on the 24th: http://forum.getmonero.org/14/events/237/monero-meetup-berlin-germany-may-24th-2015

On to your post. I think the idea of a human-driven PoW is interesting, although I suspect you'll end up with something more like Mechanical Turk to hit some lowest common denominator (although the the "work" would be much harder to independently verify, if not impossible for most jobs).

An educational PoW is interesting, but I'd imagine you'd have to have a fixed curriculum and a fixed set of multiple choice questions, in which case there's an easy Sybil attack vector (it being impossible to uniquely identify a child vs. a computer answering the questions from a list). I think the line of thinking is commendable, but I can't see a way to do it without specialist hardware (eg. vein readers) to identify individuals and reduce the Sybil attack vector.