Thank you for expressing your thoughts and questions on the matter. We have looked at the expected impact on our sustainability metrics. Since HF22, author rewards contribute <20% to our sustainability, and a significant portion of that comes from our top 10 upvoters. Since those are in part motivated by upvoting bonuses, we expect to lose at least half of our author rewards from this change.
However... those ten accounts are typically upvoted at 100% across most or all of our voting accounts, and we earn below average curation on those upvotes. We expect to make back most (if not all) of our sustainability 'losses' from the change in improved curation rewards. In addition, many of our upvoters are likely to continue, especially if we only cut to 50% instead of 0%. The ratio of votes received to bonuses rewarded will improve and hopefully offset any remaining sustainability impact.
There are some unknown elements, but we hope that our sustainability metrics will actually improve from this change, even though our total rewards may decline some.
With regards to downvoting... we have done a few manual downvotes here and there and have put some thought into a more broadly automated solution. However... we think that automated downvotes can be extremely problematic (worse for Steem than bid-bots and other behaviors that they target). We would like to find some quality downvote trails to follow, but most have a little different perspective on what constitutes abuse.