Thanks Tony, On Aug 22, 2014, at 4:46 PM, Tony Arcieri <[email protected]> wrote:
> On Fri, Aug 22, 2014 at 3:33 PM, Tao Effect <[email protected]> wrote: > Speaking of unsupported assertions, that "the merits of proof-of-work [are] > debatable" needs to be substantiated with something, especially if you are > comparing it to pre-PoW concepts. > > Systems that use a Bitcoin-like proof-of-work function are both: > > 1) Monumentally inefficient I'm guessing you are referring to either having to wait for transactions to confirm, and/or "wasted CPU". I addressed the latter in previous email with a reference to Vitalik's discussion of PoW algos that serve to better society (think folding @ home, SETI, etc.). Re the former (txn delay), yes that's an issue with PoW consensus algos, and why some look to others. > 2) Vulnerable to an attacker who wins the proof-of-work lottery most of the > time, like has recently happened to Bitcoin Yes, this is a very interesting problem that many are working on. Solution to pooled mining is known: http://hackingdistributed.com/2014/06/18/how-to-disincentivize-large-bitcoin-mining-pools/ There are many others who are suggesting interesting ways of combatting 51% issues: - https://blog.ethereum.org/2014/07/05/stake/ - https://bitcointalk.org/index.php?topic=309073.msg7385002#msg7385002 - https://eprint.iacr.org/2014/452.pdf > Compounding this problem is the desire to prevent key compromise in > Bitcoin-like systems via the use of multisignature trust and trusted third > parties which sign-off on certain operations. This approach centralizes > authority, in which case a consensus-based system like Ripple with trusted > gateways could be used instead. If it were, it'd be much more efficient, and > arguably have better security properties. I'd rather chance DPOS to address efficiency concerns [1] than rely on "trusted gateways", that reminds me too much of the CAs we're trying to get away from. [1] http://wiki.bitshares.org/index.php/DPOS Thanks for bringing up the 51% issue btw, this is something that I need to do more reading on. At the moment I'm just collecting links... :P Kind regards, Greg -- Please do not email me anything that you are not comfortable also sharing with the NSA.
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ Messaging mailing list [email protected] https://moderncrypto.org/mailman/listinfo/messaging
