Quote from: smooth on July 25, 2014, 04:28:28 AM

Quote from: AnonyMint on July 25, 2014, 04:05:32 AM

The block chain scaling can't be fixed.

"Can't" be fixed is a bold statement. Care to offer proof of that? Because there are ideas being developed for doing just that (that might not work).

I will wait to see what you guys come up with. Good luck on that effort.

Quote from: Queeq on July 25, 2014, 11:53:00 AM

Or do you have any coin(s) that you particularly like? If not, would you mind writing a short summary post about those you examined and things you like and dislike about them?

Thanks.

I like Monero's (Cryptonote's) one-time ring signature. That is definitely an innovative feature. It is a form of a Zero knowledge proof. But using it every time has a cost to block scaling. Whether that trade-off can be mitigated remains to be seen.

I don't like DRK's CoinJoin. It has a simultaneity requirement and DRK employs master nodes to solve the jamming problem inherent in the two-step CoinJoin protocol, and this opens it up to Sybil attacks, etc.. But rather than argue the intricacies of how I think it is impossible for them to solve the issue fully, the simultaneity requirement is enough to make it a non-starter in my view. It is not totally useless, but I don't think the simultaneity requirement can scale well for one.

I don't like anonymity that relies exclusively on Tor or I2P, because I know that low-latency Chaum mix-nets are subject to timing analysis, even when the adversary can't decrypt the packets. Supporting it as another layer of anonymity is not totally useless however, but I am wary of masses getting complacent about it and think they've done enough to secure their anonymity and privacy.

I like the technological innovation in Zerocoin and Zerocash, and these technologies could potentially be useful in side-chains, but as for a universe where someone was trusted to delete the master key and in the case of Zerocash we will never know if they didn't and are secretly generating unlimited coinsis unacceptable in my opinion (and seems many others share this view?). Also the crypto is too complex and too new (unvetted) to trust the world's money supply and anonymity with. I much prefer where each user generates their own keys for their anonymity and ownership, e.g. Cryptonote's one-time ring signature. Geometrically more difficult for an adversary to attack all millions or billions of users' keys. In short, I like decentralization over centralization and that applies to the cryptography keys as well.

I like the mini-block chain design in Cryptonite (not Cryptonote). I helped analyze and comment on it, so I've been following it for some months. Their effort lacks anonymity and other things I like. One-time ring signatures appear to be fundamentally incompatible with it.

I don't like any of the proof-of-work algorithms over Bitcoin's thus far (at least given what I think we know about Cuckoo hash thus far, i.e. seems to be highly parallelizable even if slightly sublinear thus I don't think it will keep GPUs at parity? It might have some role if the number of lightweight cores on mobile increases to some huge number). I have expended a lot of effort on this and have some work in progress in this area. As far as I can see as of now (subject to additional insight or information), Cryptonote's hash is somewhat ASIC resistant (will be very complex to implement) but it is slow (maybe they can mitigate the ramifications of that, MemoryCoin 2.0 was even slower), but I am concerned that if ever there is an ASIC later then it could be proprietary.

I like the concept of programmable features on top of the blockchain, i.e. Ethereum. This appears to be a holy grail of decentralized economy if it can be made to work without failing into centralized outcomes, e.g. a centralized app store to combat viruses. I also like some of the technology discussion over at Ethereum's blog and wiki (probably also the forum but haven't had time to dig in there). Ethereum is off in many directions because they are considering such a huge space of concepts. However, I don't see yet that they've worked out some of the fundamental issues.

I like Lamport signatures with Winternitz optimization.

I am interested in the GHOST refinement over Satoshi's longest chain rule. There is a blog post at Ethereum about it. I have also proposed an additional idea that goes further on a dubious case if rented mining hardware becomes ubiquitous.

I hate POS. I call it piece-of-shit. There is no entropy there. Caveat: I need to digest Ethereum's blog post on POS, to see if I have missed some insight. Also I must admit that I haven't done enough formal analysis especially of hybrid PoW+PoS. But I do expect my fundamental insight of the lack of entropy to remain fundamentally true and thus my conclusion to remain valid.

I hate transaction fees. I like (smallish but not too small) percentage perpetual debasement. My rationale is contained in my thread (and follow links off to discussions in other threads, such as Monero economy thread).

I don't like coins that have a rapidly declining debasement schedule, e.g. Monero is much faster than Bitcoin, because this means early adopters (which could potentially by botnets early on for any coin) get a disproportionate amount of the coins and I have argued upthread that I think this diminishes network effects by the N squared in Metcalf's Law.

I've probably failed to mention quite a few things I like and dislike in the decentralized crypto-economy space.