You might not have heard of most of the people on the top of the Ethos leaderboard. And while they are reputable, they are probably not as reputable as some of the top names in our industry. EIP-12, "Credibility-influence delta" will change that..

We will measure the delta between two things: - Score - Influence factor If there's a mismatch between them, there will be a penalty. Users who have very high Ethos scores but low influence will have a tougher time reaching the upper echelon of scores

As an example, take someone today who has a score of 2229. Their score is in the top bracket, but they are in the bottom 25% of "Influence factor" - aka, their existing social graph is very limited The level difference here is 4, which would suggest a harsh penalty to score

Users who are 0 or 1 level away from their bracket will receive no change to their score. As of today, we won't be boosting anyone's score for a positive mismatch (lower ethos score, higher influence factor)

For most Ethos users, this won't matter at all. Their score will not change. But this offers us an ability to balance out the top of the current "leaderboard" in a way that ensures 2000+ is reserved for people with not just reputation, but influence too.

We haven't quite worked out the exact penalty yet, but expect it to be in the hundreds of Ethos score penalty for severely misaligned profiles. We have played around with the math a bit, and are very happy with the initial findings and makeup of the top ~100 users

These changes will optimize the Ethos score to be even more accurate than it's ever been, ensuring that partners and products who leverage the Ethos score get optimal signal from determining reputable users.

The EIP process will be used for gauging customer sentiment & feedback It will allow us to discuss and understand more about the implementation, including identifying potential weaknesses. Ethos score makeup is not decentralized; we will implement this regardless of EIP voting