Topic: Proof of Stake (Read 357 times)

I'm not sure why Blackcoin made this change, but Neucoin took the same decision and abandoned the effect of coin age from PeerCoin (see p. 21 of their whitepaper at http://www.neucoin.org/en/whitepaper/download)

I have recently published an article https://hackernoon.com/decentralized-objective-consensus-without-proof-of-work-a983a0489f0a#.3xro8wbcp where I discuss the Nothing-at-Stake problem and propose a punishment scheme as a defence.

calculation by coinage is important 

I have been studying proof of stake, PoS3 in particular as it is the adaptation i am using in my project. The centralized check-pointing issues and dev centralization were resolved by creating a hard limit to reorgs, essentially equal to the number of blocks it takes to confirm a coinbase/coinstake transaction.  One strange thing i found while look at the Blackcoin implementation was the removal of coinage from the calculation of stake and instead using a static 3/2 * COIN for rewards. I did not notice any code limiting the input threshold so i am guessing people in the "know" quickly made a ton of small transactions which would all stake equally with large transactions. I can see the two sides of the coin here ie, Monopolization of the network by those who have larger amounts becomes harder, but on the other hand, multiple small inputs bearing stake may result in hyper inflation from gaming the system and also it discourages large holders from staking as the concept of proportional rewards that you put at stake vs reward is totally obliterated. If anyone has a link to a paper that discusses why these changes were made and their pros/cons as well as a case study of existing implementation i'd be most thankful for it.

For those who are aware of existing issues and the proposed (heavily) contested "nothing at stake" argument, i would like to propose the following modifications to PoS3 :- reinstate calculation by coinage, set a max coinage and penalize any over due coins. Also instate a minimum input amount ie, users/attackers cannot simply spam the chain with micro-stakes, there by reducing consensus via competing forks. And lastly add a signalling detection system that penalizes those who signal on multiple forks.

I'm hoping for a constructive conversation, i'd rather this thread not turn into another PoW/PoS swamp.