yes!
That is basically the definition of compound interest.
It is an emergent property of how KMD 5% APR works, it is not specially coded to compound. It compounds if you compound by actively using KMD.
Some coins have used demurrage to try to motivate economic activity. While losing funds is a spur to action, I think getting a bit of compounding by being active is a more effective approach.
The difference between 5% simple interest and 5% compounded wont be that much, but it will be more than zero. And people like money.
Thanks!!!!
"And people like money" muahahaha. ( btw: I´m also interested in "how things work" so money wasn´t my intention )
The big issue was how to get universal consensus to the satoshi, not to mention the wallet tx construction needed to know what actual value a utxo had. And this needed to be done to the existing bitcoin codebase.
The way I achieved consensus was to use the locktime field, which is typically unused and set to 0. I added a restriction that you cant specify a timestamp locktime that is far in the past. And locktime cannot be in the future as if it was it wouldnt even be accepted.
So, now we have a consensus safe timestamp for each utxo.
But we also need the "now" timestamp and we clearly cant use any clock as there is no way all nodes will have the exact same time all the time (if ever). What I used was the blocktime of the chaintip. So this means that when you make a new tx, the now is the time of the previous block.
Given these two timestamps, we get the fraction of the year and onto the number of satoshis for the prorated interest.
Now, astute readers will notice that between the time a transaction is submitted to the network and the time it is actually confirmed, might be hours away (ie bitcoin unconfirmed backlog). So wouldnt that mean that there wont be consensus on the interest calculation?
It turns out that there is a consensus, as by the time a transaction is confirmed, it will always have the prior block to the one it confirmed in. And thus there will always be a "now" timestamp that gets consensus. The discrepancy is that there will be a bit of interest that the wallet did not account for and this excess would end up as a higher than expected txfee.
maybe it is a bit wrong that the mining nodes that might be responsible for delayed blocks and the unaccounted for interests getting this as txfee, it is a very small amount and I dont intend to ever let KMD end up having hours and hours of backlogged transactions.