1st Talk

NP-Hard Proof of Work for Blockchain Construction

Speaker

Faye

Synopsis

Bitcoin proof of work uses a Hashcash proof of work. Revolutionary and brilliant, but there is some potential to make some “good” utilisation of the computing power (and ultimately electrical power) expended to form the proof of work. Also the Hashcash proof of work is cornered by a bunch of airplane hangers in China using huge ASIC based mining rigs. The game is over for hobby miners.

This talk is about:

  • creating a proof of work that still uses the existing Hashcash Proof of Work for the first round, and then taking the lowest hash values belonging to the n miners who mined those values and then creating an instance of the Travelling Salesman Problem in order to solve a problem that is: useful to solve, ASIC resistant, and actually requires a competition in Algorithm design to solve.
  • Furthermore, Hashcash Proof of Work is based on a heuristic thought that finding low hash values is “Hard”, where as the Travelling Salesman Problem is known to be NP-Complete if considering the decision problem and NP-Hard if considering the Optimisation problem - which I propose we utilise.

Links:

2nd Talk

Efficient Privacy Preserving Set Intersection for a Decentralised Web of Trust

Speaker

Alvaro Garcia-Recuero

Synopsis

Today mass-surveillance and personal data collection about individuals in the WWW has become a serious concern to liberal societies and citizens. We already know about the existence of some large organisations and governments that contribute to such ends, but a number of less prominent actors are also profiting from the massive amount of content generated by individuals. We have designed an efficient Private Set Intersection protocol that is compatible with a (de)centralised Web of Trust, which together with a future decentralised Internet can help discourages abuse and surveillance.

Links: