I really like the extra diversity training and the provision of resources for best practices, the moderation tools for the teachers lounge and the two level check system for complaints. I think that this training and these resources should be made public as far as possible for all users to benefit.
Judging by the score composition there seems to be a rift in the community though and I attribute it mainly to the "no discussion about it"-policy. What surprises me most is the need for it without many such discussions. In 2020, I regularly check on meta, and do not remember anything related.
However, the biggest problem for me is missing quantitative knowledge. How much of a problem is it currently? It could be very low, high or immensely high. Compared to last year, the measures taken then could have had a big impact, a small impact or no impact at all. The time evolution could have been positive, neutral or negative. It could become worse by the day or it could be effectively not an issue anymore. Do we need to be much more strict than now or are we on a good way - there is nothing really publicly available to gauge that.
Without really knowing what's going on and without seeing anything (may be me though), I'm a bit lost and can only take notice of what is done and what is not.
After some days of discussion here the composition of the score has further trended towards an equal number of up and downvotes. Maybe the increased moderation activity also plays a role here. It certainly feels a bit reminiscent of moderation in October 2019. Not sure if there is a better way, but it feels like if only the wording of the "no discussion about it" thing and the moderation approach would have been a bit less strict, maybe the announcement would have been considerably better received.