HF21 is extremely bad. People who put in the effort to make blog posts get their rewards cuts. People just upvoting get more. Even whales get increased curation rewards just for clicking on an upvote button. Maybe the best strategy for everyone is power up and delegate their complete SP out to get automatic rewards and do nothing. Have they gone mad?
You are viewing a single comment's thread from:
Thanks for sharing your views.
Do efforts on blog posts equate to value? If some people work harder than others on a post, does that make the most more valuable? How are you defining value here? If a post is considered "valuable" does that value change the price of the STEEM token?
"Just upvoting" with a non-linear curve no longer means they will get a meaningful curation reward. They have to upvote stuff other people agree is valuable. That means effective curation. That means more generally-accepted-as-valuable content rising to the trending page. That, hopefully, means more investment in Steem and Steem Power by brands who want to be associated with this high quality content site.
Delegating SP to effective curation systems would actually be a good use of SP. Some might say it's a better use than posting valueless content and self-voting it to get a reward. One fills the blockchain with crap, the other improves content discovery and has the potential to make Steem look like a better investment.
Edit: btw, your "have you gone mad?" statement fits nicely into the post I linked to first in my article: https://steemit.com/relationships/@lukestokes/we-disagree-are-you-ignorant-immoral-or-stupid
Well, I see you still have a really different opinion. Nothing stops bots/people from finding the most profitable posts to upvote to get the most rewards. Got nothing to do with the value of a post. So we will see that the top popular bloggers get the most votes. People will cluster around those. Newbies will have a lot of difficulties getting popular when blogging. We will have more a "winner takes it all" thing. Self votes also become more profitable I think?
I do not really believe most people buy Steem to make Steemit better (it's a pity, but that's the case I think). Most just invest in cryptos that will give them the highest ROI. If you delegate SP out, those people/bots will in a lot of cases 'optimize' their votes to get the highest ROI.
You're highlighting many of the reasons why I was in support of a linear rewards curve. Unfortunately it didn't work out that way and it was just gamed even more. Yes, the little guys will get very little. It will go back to being more like a lottery as originally designed and as outlined in my post. I do think more of the reward going to curation combined with the ease of downvotes (and hopefully tooling to make content to downvote easier to discover) have the potential to change things up this time. We'll see.
It's not really a lottery when there is near-linearity among everything above 16 STEEM. It should be quite different from the original n^2.
Thanks for clarifying. I need to go back and re-read the deep dive post on the curve change. I'm not the best at the maths.
I have never downvoted anyone. In some cases down voting might be a good option. However I have mostly seen that downvoting is being misused. Like "oh you don't share my opinion, well I will downvote, and if you continue posting the same thing, I will continue downvoting everything you put on Steemit".
Yep, that happens, but so does people extracting thousands of dollars by self-voting and voting via sock puppets on completely useless (often spam) content. Just because most people don't see it, doesn't mean it doesn't happen. It's usually done on older posts as comments to avoid being noticed. People complain about low rewards or a low STEEM price, but often they don't understand what causes those things. People extracting value from the Steem rewards pool is a big part of that. At least vengeance based downvotes can be countered with upvotes.
I just read this, please tell me this is not true:
<p dir="auto">Under HF20 a post receiving $1 in upvotes would give the author $0.75. <p dir="auto">Under HF21: <pre><code>The reward pool will be 10% lower to fund the SPS. The CLRC will reduce the payout by a further 40% compared to linear. The payout to the author will then be 50% (rather than 75%). I make this: $1 * 90% * (1 - 40%) * 50% = 0.27 under HF21 <p dir="auto">So broadly: $0.75 under HF20 vs $0.27 under HF21, a reduction of about two-thirds for smaller payouts.If the $1 vote was a single vote from a single account, then yes, the non-linear curve would give a different result by design. This will hopefully prevent comment farming and self-voting we see which leaves more of the rewards pool for valuable content. If it was a number of votes, then I think you'll begin to approach the same amount as previously. See the deep dive for details and examples: https://steemit.com/steem/@vandeberg/reward-curve-deep-dive
The end result is, yes, I think the author rewards per post will go down. If that's too much of a problem, people may want to try other experiments like https://palnet.io or do what I suggested in my post and hide the $$$ to figure out what motivates you to be here.
There is no trivial way for a bot to find the most profitable posts to upvote because the curation algorithm is designed such that voting on anything is equally profitable at any given point in time. To tell the difference requires anticipating how others are going to vote which in turn requires evaluating some aspects of the post. People and perhaps some forms of AI can do this, but simple bots can not.
Self-votes are not any more profitable, there is actually no change there except: a) small payouts are penalized, which will include a lot of small self-votes in practice, and b) with cheaper downvotes it is hopefully more likely that undeserved self-votes will be downvoted (though it is not guaranteed this will happen)
But let's say there are a few popular bloggers, wouldn't that mean that people with automatically votes more and more for a few of the top bloggers to get the highest payout, even when a certain post is garbage?
First of all, it goes linear after about 16 STEEM, so based on current prices, once the payout goes above $6, piling on more votes does not increase the payout on one such post more than another.
Second, even generating a high payout for a blogger doesn't directly benefit you as a voter, unless you are the blogger and are self-voting, which is a different issue.
I think we will need easy-to-use tools for finding these posts and downvoting them. Without that, it's unlikely people will spend time doing this and these changes kind of need this to happen for this to actually improve things around here.
I don't think there is any real shortcut apart from people being willing to subjectively use their downvotes on payouts that just aren't worth it to Steem to be getting paid what they are. Which very well may not happen due to a variety of reasons.
When I say self-voting (and vote selling) that is a shorthand for posters opportunistically extracting more than their contributions are worth. There are many ways for that to happen other than the obvious and literal self-voting that can be easy found by tools, including multiple accounts and sock puppets, vote trading, purchased votes, games, social engineering, etc. and this is not at all a complete list.
You're right, the time and effort (as well as why bother to get involved) factor is a major consideration.
Know of any projects actively working on this we might support through SPS?
Maybe my comment wasn't clear. I do not think this can be helped much by tools.
There are probably some existing apps that identify literal self votes but I don't think that is very useful and I wouldn't support much SPS funding for that. If we did build a super-duper self-vote finder, then milkers will just get more clever and the investment will have been largely wasted. There is no substitute for voting based on value, which is almost entirely subjective, or at least requires "intelligence" to assess.