@muf18: I agree that there should be a continuous improvement of the algorithm. So I would principally not opposed to hard forks, only that they must have ample support in the community. For me, for example, the 10% staking reward is a bit high - I would reduce it to 3% or 4% but distribute it only to nodes that are online for a longer time, like @dzarmush proposed.
The only problem is to find a suitable algorithm (see below), because the solution to that problem is not trivial (how do you check collectively which nodes are online without cluttering the blockchain?). Until now, I only knew about Timekoin (one of the first altcoins, but it used a completely different algo that was not easily portable to Slim) and the theoretic concept of "
Proof of Activity" that I think has never been implemented in a real coin.
Particl and Lisk, I think, have yearly inflation that goes to nodes that stay online. I thought Nxt is also the same, but now I remember it's not.
Thanks. Lisk, as far as I know, is a DPOS coin where only "delegates" (that must be voted) get Proof of Stake rewards. They must be online, but the DPOS algorithm type is flawed for me (it is open to social engineering attacks because the validator list is limited) and it should be very, very difficult to port it to a bitcoin-based currency like Slim.
But Particl could be interesting as a model to follow because it seems to be built on the Bitcoin code, so it could be implemented more easily. I didn't know it until now, so I'll investigate it.