This is my thinking to. If we use the rank stabilisation mechanism I discussed earlier where when nodes are located in the network they are accepted if they have rank above the average (current) and the current close group is below average. They are rejected if the close group is currently above average.
Bit techie, but the close group knows their own close groups average rank. So kinda like this network average is based on all the groups surrounding your group.
This flattens the imbalance of network data storage capability (an important factor for the new network design with safecoin). Importantly here it also allows a more accurate network average metric. It will vary across the network, but that’s OK. There are arguments about moving vaults etc. We do not need to go that deep here. Folk will not go to massive extra hassle (some will, but negligible).
Then we use a metric very close to the suggestion here. Under average nodes get safecoin via mining at a pretty slow (but increasing) rate, At average and above they are on max rate (average → average + 20%) it then drops off again. So no vault will earn safecoin if it’s way oversupplying. We may need to look at that though as splitting a vault is possible. I hope people do split vaults as it adds to the number of good nodes on the network (safety in numbers does work in decentralised systems).
I hope this reads OK. We can document this, but be good to argue the algorithm, coding this part is pretty quick, getting it right will be where the time is spent. It’s not that there is a load of options, but the shape of this curve is interesting. I prefer a sigmoid curve like approach, but I have no tie to that mechanism for sure. It will work, but it may work better if it is refined more and that means more thought.
All this input on these matters is superb BTW.