Hi all, I can agree on the bad distribution of PNK token, taking into account that a “fair distribution” is really important for a justice protocol when using token-weighted mechanisms. And because of that, I agree with the arbitrator change.
However, I am not completely sold about using UBI as arbitrator token. I think PNK fixed supply is an important property for the Kleros mechanism to work correctly. It makes 51% attacks really hard and expensive to perform, as buying half of tokens having fixed supply will have a huge price impact (see Kleros whitepaper for more details).
I understand generating “artificial demand” over UBI token is needed, but I don’t think it will be done at cost of Proof of Humanity court security. And I think it is important to think UBI as a project that uses PoH, not as the same one.
If UBI is used as arbitrator token, because of its constantly increasing supply, I would say that it will be easy for a whale to attack the court in a cheap way. Even if we argue that the increasing supply is constantly diluting whales, that’s true in the perspective of UBI balance (looking at it just as ERC-20), but not sure that will be translated to the court/juror context. All jurors should be constantly buying UBI and staking it in the court to fight potential whales that appear, it does not look as a desired behavior (aside of the fact that will increase UBI demand).
But I didn’t came here to criticize, I have a proposal: HPNK.
You can create a new ERC-20 token, let’s call it HPNK (Human PNK or Human Pinakion). So every human already registered in Proof of Humanity will be able to claim a certain fixed and equal amount of HPNK (note that HPNK have fixed supply).
Then you can change the arbitrator to an identical one but just using HPNK instead of PNK. All properties designed by Kleros team will still be there, as the HPNK follow same properties as HPNK, but with the difference that is “fairly distributed” to all humans in the registry. This means the DAO will be achieving its goal of self-maintainance of the registry without putting it on security risk.
Implementation is also much more easier, which is also important to reduce the chance of any kind of exploit.
Only drawback I think you can mention is that this does not give utility to UBI token. But as I said, for me is important to see UBI and PoH as different things, the former depending on the latter, but not vice-versa.