Re: Ripple

> Everyone using the same list obviously works just fine. However,
> surprisingly little overlap is needed and the network is still robust. For
> example, we simulated with 1,000 validators, each picking 32 of the other
> nodes randomly to use as a trust list. There were no problems whatsoever. We
> then tried with each picking 12 other nodes randomly. Again, no problems
> whatsoever. All these topologies result in everyone trying to come to an
> agreement with everyone else.

Can you elaborate a little on this? How many of the 1,000 validators
to choose from were Sybils/malcious or otherwise adversarially
controlled? Or what was the attack model?

For comparison, in Bitcoin, there's a consensus style proof-sketch in
Satoshi's whitepaper for when you assume half the hashpower is honest
- the other half can be arbitrarily malicious and collusive. Since
it's based on power rather than identities, sybil attacks aren't
relevant. Rational modeling (i.e., under what conditions Bitcoin is
incentive-compatible) is a more complex matter, but validators in
Ripple aren't directly rewarded anyway so I'm happy deferring this
till later.


-- 
Andrew Miller

Received on Monday, 18 November 2013 06:03:25 UTC