Thanks for the continued discussion around this proposal, it is good to see the community engaged in an active discussion about an important topic.
A lot of feedback relates to doubts about the specific criteria being robust enough to select for top candidates and/or that it may unfairly advantage incumbents. This is admittedly a limitation of this first iteration as we currently have a limited set of attestations or objective data to leverage and we need to continue to collect data about which attestations/data serve as the strongest indicators of performance. We will continue to build towards a more robust graph of attestations and upcoming elections will further build our understanding of which attestations are useful when evaluating candidates. Having more robust data about this will also allow us to expand selection criteria to rely less on previous participation and move towards more open-merit-based criteria (which was an acknowledged tradeoff in this first iteration.) We would love to see continued discussion about the type of selection criteria delegates would be supportive of and/or find more useful.
We would like to clarify some of the feedback about the use random selection, or “sortition,” as a small component within this proposal. Importantly, this is not a sortition experiment. In fact, we acknowledge that the use of sortition here is a sub-optimal and temporary mechanism to be replaced in future iterations.
Sampling is a temporary mechanism to be used only until we collect enough information on attestations to be able to incorporate a rank order into the selection mechanism itself.
We have experimented with sortition before (see here and here) - in instances where we used a sample to approximate the full population in a way that is robust to manipulation and has individual fairness properties.
The goal within this proposal is simply to (temporarily) use random selection to determine a set of final members from the list of eligible (and opt-ed) in members without relying on the Foundation or a process that mimics an election (application process.) This experiment is meant to test the use of an attestation-based selection mechanism (similar to what we did with the Collective Feedback Commission members), which ultimately will not use random sampling.