@jonald_fyookball:
Yes, that sounds right. With Sign to Mine pools, there has to be someone to distribute the rewards. At the beginning at least, probably most mining on ziftrCOIN will be solo-mining, although there are ways to do pools (as a convenience) for groups where all members have mutual trust. Distributing block rewards more frequently (1 minute block time) and allowing for mini-pools where all members have established trust should help to limit the variance of rewards for miners.
@othe:
Neither Spreadcoin nor ziftr were the first to invent the idea of non-outsourceable cryptopuzzles. I believe Andrew Miller was the first one with the idea (
https://bt.irlbtc.com/view/309073.0, someone correct me if I'm wrong), but proposals like Two-phase-proof-of-work from Ittay Eyal, and Emin Gün Sirer (
http://hackingdistributed.com/2014/06/18/how-to-disincentivize-large-bitcoin-mining-pools/) are along the same lines. This person seems to have a similar idea too:
https://bt.irlbtc.com/view/652443.msg7314429#msg7314429.
It seems like ziftrCOIN and Spreadcoin both independently implemented their own versions of non-outsourceable cryptopuzzles. However, one of the main differences between ziftrCOIN and Spreadcoin seems to be that Spreadcoin is entirely opposed to pools (someone correct me if I'm wrong). We think pools are necessary to minimize the variance of rewards, but that they should be limited in how big they can get (i.e. you have to trust everyone in the pool, and that will limit how big pools can get).
ZiftrCOIN has a few other features as well. For example, in ziftrCOIN, the mature coins spent in each block are also counted and used as a tiebreaker when two blocks on the tip of the chain are discovered with the same work. This is the rough idea, there are some other conditions that have to be met to engage the tiebreaker. Note: we used to do this tie-breaker in terms of coin age destroyed in each block, after talking this over with Andrew Miller, we decided just counting the number of mature coins spent would limit fungibility problems. This tiebreaker encourages miners to actually include transactions, rather than how the default tie breaking algorithm (whichever one was received first) actually incentivizes miners to not include any transactions at all so that their blocks propagate faster.
We also limit the size of the block chain using a growth-dependent algorithm. We know Gavin Andresen thinks this will result in mining pool cartels, but together with Sign to Mine, we expect this to not be a problem.
