Understand that it will not avoid pools.
Stakeholder will split their balances and created pseudo-nodes (if whatever forging limit you impose on accounts or nodes).
If you want to avoid pools the right way, find a crypto-solution.
Either I'm totally off-base here, or we are having trouble communicating.
Is not intended to penalize individual account holders. Capping effectiveBalance is certainly a very simple answer (I'm big proponent of KISS philosophy). The idea may be flawed, and you could be correct in that a more correct cryptograpically-sound solution might be preferred.
One account owning 5% of NXT is not the problem. Small number of pools forging on behalf of large majority of NXT is... and, is only a problem if/when unbound leasing of forging power is implemented.
If the only problem with the approach is that it inconveniences large stakeholders, so what? If it adequately addresses the need to avoid large pool centralization, then it could be a reasonable tradeoff.
Okay. I see.
Let us define several terms. NXT is a layered network (consisting of 4 individual networks = 2 real networks and 2 meta networks):
A2) account holders (meta layer - 1 account holder controls n accounts)
A1) accounts (crypto layer)
N2) node holders (meta layer - 1 node holder controls m nodes)
N1) nodes (physical layer)
The accounts of an account holder act as one.
The nodes of a node holder act as one.
If one forces a split of accounts/nodes, the holder is still the same. So, these accounts/nodes still act as one. So, the security level is the same. In terms of the networks, if one forces N1 and A1 to split, N2 and A2 are still the same.
I hope that clarifies my conclusion.
In the end, it would be an inconvenience for large stakeholders but for nothing.
