In our plan, the “scarce" resource is time.
The proof is on-chain historical accumulated transactions in a “harvest club”, the reward method is block transaction fees distributed back to all addresses in the club. The “time” will trace back 365 days (our current time frame selection). This shall make speculating costly and unpredictable. This pervasive, long time horizon distribution makes speculators wait one year to complete their return cycle.
I like long-term-engagement rewarding algorithms, so this looks positive for me. But I continue to have doubts about this model, see below. Can you link me to the explanation of a "harvest club"?
On the flip side, there is a positive externality from speculative transactions. This will stabilize the transaction fee, just like derivatives make futures prices logical. If speculative trx exceeds 25%, they will be punished in ROI mathematically.
So basically your plan is that if users transacted only to increase their probability to find blocks, they would rise transaction costs. I fear however that the algorithm will give incentives for a kind of boom-bust cycle: when transaction fees are low, then there will be a race for cheap transactions, filling up the blocks until a certain price is reached; then the speculative harvesters will lower their activity temporarily - they have time for a year to increase their score, if I understand well. Still we have the problem that speculative transactions will lower the space available for "real transactions" - even a 25% decrease of the available space would increase the total costs of the system. Transaction fees would be, at average, higher than with PoS or PoW coins.
Regarding egalitarian, the core assumption is: the wealthy (as a group) make less transactions than the poor. The wealthy spend more per transaction then the poor, but collectively as a group, the poor make more transactions. The reward set will tend to go to a group with more transactions. [...] Regarding exchange attacks, most POS models are vulnerable to this, however our novel on-chain transaction counting methods give less advantage to “hoarders”. Most exchanges work off-chain, so those will not be counted in our transactions count.
I understand you intention. But as I already wrote, exchanges and similar businesses tend to be the "richest" participants in a cryptocurrency. Exchanges, while "trading" occurs off-chain, make a lot of on-chain transactions too (deposits/withdrawals, transfers from cold to hot wallets). So they tend to dominate the transaction count. I don't see a way how to distinguish between "exchange addresses" and other accounts (I don't know what you mean with your "novel" way to count transactions, please give me a link).
In POS, the wealthy receive a linear reward, however, in POT, collectively poor club members are much more pervasive than rich. Most exchanges will pursue the lowest possible on-chain transaction fees to retain their own commissions.
As a conclusion, my problem with this algorithm is this:
- there will be a certain amount of speculative transactions even if it's "only" 25%;
- these speculative transactions permanently pressure transaction fees to the upside, fees will be higher than in comparable PoS or PoW currencies
- the algorithm is also not egalitarian, because a sub-group of "the rich" (crypto businesses) will get most transaction fees.
I don't know if the problems are solvable, but I fear it will be difficult. And as much I would like a more "egalitarian" consensus model, I think taking transaction count as a base is not the right way.