RE: Steemit Update: HF21 Testnet, SPS, EIP, Rewards API, SMTs!
I don't need to better understand it, I understand it very well!
You should better understand what the permissionless property of blockchains means. No one needs to be 'allowed' to do anything, they can just do it.
I addressed this above when I said stinc and high stake people could stop vote buying by excersising their right to be allowed to downvote both the people who run these 'so called' services, and the people who use them! That would be them exercising power in a permissionless environment. It would exercise a measure of control, and considering stinc have had massive stake lying around for 2 years doing nothing it boggles my mind why they haven't at least attempted to stop something that is so obviously of detriment to both the platform and the price of the steem. Obvs that would start a whole new discussion about if we're now centralized... but I feel that the stake distribution on steem (and the ninja mining by more than just stinc) means that ship has sailed.
The only way in which it can be stopped is by changing to system rules and incentives to make it a smaller or non-existent niche.
This is true to an extent, but what I said about responsible use of downvote to stop negative 'niches' from proliferating is valid.
The network effect is what can truly make steem a wold class platform. An ever expanding network of people sharing content, and the advertising that would bring and other investment streams is what can catapult steem to overtake Facebook. This isn't rocket science. It also follows that as steem's value proposition is to actually pay people for their content, we need to fulfill that value proposition to succeed.
I know that there are issues to do with balance between stake flowing out and stake being converted into SP, also the relative lack of trading of steem as a token.
But, as I stated in a comment above, if the issue of steem being a fundamentally unappealing place for content creators doesn't change soon, we will lose any first mover advantage to facebook coin or voice.
I don't disagree that the best way to change all of this is to make the system not reward vote selling/delegating behavior... and better reward manual curation or delegating to a guild that does the work for you, I'm just unsure this HF21 will achieve that, and if it doesn't... well the clock is ticking.
Steemit always intended their stake to be non-voting, in fact they implemented a feature into the blockchain code to enforce that. For various reasons the feature was never used, but the tradition of not voting with their stake remains. That tradition exists for a reason, at least historically. With the dominant stake they once had, if they did vote, no one else would have any meaningful say at all in anything. So for that reason, and because the stake was mined for the purpose of funding the company, compensating their founders, and giving away to new users, they have never voted (apart from a few specific exceptions, mostly emergencies).
Even despite that tradition, they do delegate a large amount of stake to steemcleaners and some other downvoting efforts. It simply numerically isn't enough. I haven't done an exact calculation but my intuition is that bidbots alone have more stake than Steemit (recall that Steemit has sold a lot and continues to sell a lot; the meme about Steemit having an absurdly dominant stake like 80% as they once did is complete wrong), and bidbots are not the only problem.
Trying to solve this the way you suggest would not work and has not worked. Working against the underlying incentives by trying to tell people what you think they should do with their stake despite incentives encouraging them to do something else is inefficient to the point of ineffective, even if you are ultimately right in a way. It is like trying to push a car uphill several miles to get to a repair shop but collapsing from physical exhaustion before you get there, when you could instead roll it downhill to another shop at the bottom of the hill.
I share some of your lack of certainty about whether HF21/EIP will produce the intended and desired results but I think they are worth a try. We should have tried making such improvements long ago, but since we have not, there is no time like the present.