RE: Wholesale witness voting changes.
As a witness I consider the value of STEEM Power to be the primary focus of Steem's economy because without valuing STEEM Power, Steem's overall value will fall to pieces. Without valuing STEEM and SP, post rewards and thus user engagement would continue to fall until nobody was left.
How is this not the description of a pyramid scheme?
I like Steem because I think on some level the voting system can work, with a large and diverse range of interests each holding an effective amount of power, relating to each other in communities, and valuing the work that each other produce. If that works, it works no matter what the Steem price is.
If it doesn't work, I don't think I care what the Steem price is.
In some ways I see what we're doing here as true price discovery for content, and my experience in content industries suggests that it's currently highly over-valued because of the influence of faddish prices in the overall crypto market. So if we're worried a lot about Steem working at $0.80 and trying to drive new users by driving the price up, that's not something that makes me optimistic about the future.
If we can drive the price up by bringing in new users and establishing more use cases, then Steem makes sense. Otherwise I see it as just another coin.
It isn't a pyramid scheme precisely because, in your words, its value can go up by bringing in new users and establishing new use cases. That's what the whole system is designed to do.
Cutting curation in half back in 2016 was a huge change made on very flimsy support (as @pfunk points out, sort of an accident since the original idea was to pay 25% to comments but when that got shelved somehow the 25% ended up with the author), and it hasn't helped at all in actually directing the rewards to where they do the most good in attracting new users and use cases. Instead it dramatically shifted incentives away from curation and toward reward extraction.
Does anyone really believe that a poor allocation of rewards suddenly works well if you pay out $1.50 in author rewards to a user instead of $1? (A lot of which ends up being hit-and-run reward cash out schemes?) I don't. The 50% increase in author rewards was almost entirely wasted.
I'm not sure I'd say "works well," but it absolutely makes programs like mine, which are dedicated to distributing rewards to create a class of users with effective stake, possible, along with linear rewards. Without those two things building for those goals becomes somewhere between much, much harder and essentially useless.
I think you are so caught up in a point of view you are fooling yourself. A change to between $1.50 and $1 or vice versa does not make anything "much, much" harder (or easier). It could make it a little harder (or easier).
I do agree on linear rewards. Potentially smaller stakeholders can be completely or nearly completely shut out from any meaningful allocation decisions, even under what is being called "moderate" superlinear. That is indeed where things could indeed be "much, much" harder. II'm generally opposed to it.