Thanks @ChainflowPOS for these considerations!
This discussion may be helpful for us (Luminara 
), since we also want to do some delegations very soon. As an aside, we intend to be sensitive to operators who may have fell through the cracks of the recent Anoma Foundation delegation selection.
Transparency vs bikeshedding
Personally, I think that too much transparency can create practical problems, like a) bikeshedding over evaluations (arguments leading to misaligned focus and breakdown in cohesion ie. it will make our community toxic), b) gaming the evaluations, which can be hard to codify in a meaningful way and c) expanding the work involved in evaluating every period. If the majority of our community are like “yeah this seems about right,” imo the delegations have been done well. But there should be room for incremental improvements so the process yields better outcomes.
Bottom-up feedback
I think it would be helpful / informative for the entity offering delegations to highlight ideal examples of recipients, because it signals what it is that the entity values and helps the future set of candidates update their efforts accordingly. It could also be helpful for the candidate to make their case for why they should have been selected, since this could help the AF improve their process and ideally not have as many misses next round. It would also help Luminara to see the missed candidates, so we can help retain high-value operators.
Perhaps the AF could make a lightweight process for this, but it may be simpler to have a forum post that a passed-over operator could publicly reply to and eg. give the top 3 reasons they expected to be offered a delegation. This could be helpful feedback for the AF, and also a way that others (like us) could use to consider how they delegate.
An advisory could be ++interesting! It would be great to get a few different perspectives and some insight to help broaden impact and reduce misses. It would be great to the lightest-weight form of this that integrates into our process without completely exposing our process. We want to be judged on the overall results/outcomes of our decisions, not invite a public discordant discussion for each of the micro-decisions within the process.
Impact vs cost of process
I love that there’s this kind of considered feedback, criticism is ++important. Also, perfect is the enemy of good, and imo it’s better to have a “good enough” process than to miss bigger opportunities or have burn-out perfecting it.
Luminara will strive to keep our process as lean as possible, relative to how impactful we assess delegations to be. Impact vs cost of process, we want to be as economical as possible, minimize unrest and/or trigger good operators to exit, while maintaining focus on our primary mission.