Quote from: aminorex on June 10, 2014, 01:48:26 AM

One approach not yet considered, which is motivated by the lost coins issue, would be to emit in the following year however many coins have remained unmoved for the past 10 years. Actually, it would make more sense to make it a full human lifespan, but the emission curve is too aggressive to do that gracefully, and lifespans may change a lot, so whatevs. If you don't want to get diluted, just be sure to move your coins every 10 years. Sadly I can't control whether someone else moves their coins, so it's a bit unsatisfactory in that regard.

Regardless of such minor issues, recycling lost coins as rewards does make a lot of sense. It would be a deeply compelling idea if the emission curve tapered over 80 years or so. But even on a shorter frame, if the shelf period is uncorrelated with the demographic cycle, this would also have an adaptive effect, with respect to velocity fluctuations resulting from that cycle -- assuming that newly issued coins have higher than average velocity (which seems certain).

Adjusting emission depending on velocity read directly from the blockchain is such a bad idea that I won't go there again. But counting idle coins provides a somewhat similar benefit, without all the pitfalls. It is certainly less dependent on committing to a particular economic theory.

I'm not sure, if a miner holds many coins they are inclined to leave as many as possible sitting idle so that their mining rewards in 10 years are higher. Anything like this can be gamed too easily.