Hacker News Reader

Comment by kbody | Hacker News Reader

I appreciate the research of zcash, but trusted-setup is still just a very sophisticated security theater. The least they should have done is have constructed an open participation.

View on HN

Indeed. Secure multiparty computation with large numbers assumed malicious participants to do the initial setup would've been enough. As it was, I just can't trust it - even if all the parties are honest and completely trustworthy. I may be paranoid but, in my opinion, if you have to do the same rigmarole that the CA system does then you're still fundamentally broken.

The problem was the existing MPC protocol for generating the paramaters didn't scale that well. 1) It required participants to stay secure during the entire computation. With 6 people we still had problems because it took long enough people had to sleep.

2) If anyone aborted, the protocol had to restart

The next version of the protocol will resolve both of these issues.

At the time, the multi-party computation protocol could not scale to a large number of participants.

Define large. In any case, I would rather have a protocol in which only one of the participants need successfully discard their local state for total security than n participants.

You still would need just 1 participant, it's the computation part that would be open to any number participants in order to reduce the possibility of collusion.

That was the case. If one of the 6 people completed the computation and discarded the results, the paramaters are secure.

Ideally it would have been more than 6 people, but that protocol really didn't scale to more than a handful of people.

Reply on news.ycombinator.com