Just a side note: all GETs are “unrewarded” in the current model. I’m not sure if the plan is to return to rewarding GETs or not, but the current scheme seems to be good enough on the face of it.
Currently all nodes in a section are rewarded when the section splits, receiving an amount from the section wallet in proportion to their age.
So nodes are rewarded for good behaviour over an extended period of time, using the accumulated income from PUTs (plus a premium of newly minted SNT up to the section threshold).
So the level of a farmer’s resources (speed of upload primarily) is no longer a factor in rewards, just the node’s length of service? Doesn’t this detract from the network’s goal of speed and efficiency?
Several things are factored in to you being a good enough Node. I expect playing with this to try and improve or break it will become a significant focus of test networks once the core functionality is in place and obvious bugs squashed.
Everyone is free to suggest improvements and point out weaknesses, the more thought out into either the more useful that could be.
What are your thoughts about how the current rewards algorithm affects speed and efficiency of the network, versus the old scheme of rewarding the most responsive nodes?
I haven’t thought about the comparison, it would be a good thing for someone to work through and explore in a new topic but I’m not planning to do so.
I like the simplicity and what I expect is much greater efficiency of the ‘Section split’ approach but accept that it might not be adequate without some tweaks. I also like the clever way it seems to make nodes more sticky (hanging around for unpaid rewards), although this could also cause more nodes to leave a Section at the same time (just after receiving their reward), depleting a newly formed section.
That might not be a problem, but if it is it might be possible to spread the payments out enough to minimise that effect.
On your comparison question though I don’t have any other thoughts at the moment.