RE: Although subjectively downvoting posts to zero is anathema to me, the Layer 1 ability to do so must remain (for now at least)
Thanks for your support Dan, it is appreciated here. I just read and reblogged your post on downvotes and you make some interesting points. I know that @r0nd0n is moving to amplify the actions of @freezepeach in a way that is not dis-similar to your ideas, but having an inbuilt system function within a DAO along the lines you suggest would maybe be the best possible option. I think that maybe your free upvote suggestion might be made simpler by just reducing the amount of free downvotes that people have to 1 per day, though the comlexity of the various proposed options will no doubt yield outcomes that I am not able to predict in advance, so maybe your model would work better.
It's definitely sub optimal to expect large stakeholders to police the network and usually it isn't needed - but then it usually isn't the case that other large stakeholders attempt to use negative reinforcement to increase their curation payouts, rather than offering positive, creative additions to the community. While it is unpleasant to see this in action and particularly when it is done under the guise of being a 'service' to the community, at the same time it is inevitable that such exploits and barriers to sustainable growth will get exposed in systems over time. The key is to focus on the solutions rather than get caught up in constant battle.
As a long time system engineer and now digital marketer, I am a fan of iterative testing of ideas and A/B testing. These allow us to see without any doubt which ideas/designs work best to achieve performance goals. Layer 2 solutions offer a definite way to test out a wide variety of ideas and I have advocated for this since the days of the mythical SMTs on Steem. We have never quite gotten there though and the high cost to entry for Layer 2 sites on Hive currently is significantly holding this process back.
If layer 2 solutions become cheap enough to enable rapid development and testing of new ideas then I think we will see these wider systemic problems resolved quickly too. In the absence of that, I would suggest being open to testing variations on downvoting via future Hard Forks, such as reducing the amount of free ones available and/or the DAO idea you mentioned.