I'm sorry, but still being somewhat new here, just a little over one year. The 50/50 split on rewards in my opinion would be the death nail, (final nail), in the steemit coffin lid. Unless one of you can explain to me why I should vote for a person reputation level 40 or less or heck even 40 to 68 rep, when their rewards are only pennies, nickels, dimes, loose change, or not likely to break the Five dollar mark, when just about every post done by someone with a reputation of 70 or more get rewards in the hundred dollar range, (like this post).
So explain how this math is going to help content creation? How it is going to create Quality post? You think it is funny that @berniesanders can make a post with no content and earn hundreds of dollars in rewards, with a 50/50 split he will be in the 10,000 dollar reward range. @haejin will be in the 10,000 dollar reward range. There would be zero reason at all to vote on any name that is not easily recognized because that is the name that will get the votes because people know they rake in extremely large rewards. Who in their right mind looking to make a buck would not vote on their content.
Some serious rethinking needs to be done on this 50/50 reward split. What would you want a chance at a part of a 10,000 dollar prize or a part of a two cent prize.
A separate pool for downvotes is also not a good idea, but I will stay my fingers on that one.
You won't get more curation rewards just for voting for the post that gets the most rewards. You maximize curation rewards by finding posts that haven't been upvoted but will be upvoted afterwards. For example, if I want to maximize my curation rewards, I'll typically get a better return on my vote by voting for lesser known posters of quality material.
Today that may be true, with a 50/50 split a reward of $100.00 vs a reward 0f $10.00, that is not going to likely be the case if a person is looking for curation rewards. People think the 50/50 will kill bid bots out right, that is also not likely to happen. There are people out there that are all about the "look at me look at me", bernie and that youtube serial quitter guy, damn can't think of his name, the not so swift self serving video guy.
The other thing I do not see it preventing is the self vote, add on the desire to cut the curation time down
So how is this going to cause a large account to think twice about self voting at 3 minutes.
When it comes to curation, unless the curators with big pockets are willing to upvote a small account at a very large percentage, then no one else will be voting on it. So a vote on a two cent post is not going to get a real curation reward verses voting on a one hundred dollar post. Curation trails and votes are not about content, it is about money and how much a person can make, from the content of another person. If no one else votes on a piece of content you vote, then you are not going to get a very good curation reward no matter how good of quality the post is. It boils down to "Name Recognition", just like in any other aspect of voting in real life. If you have never heard of Joe Shmoe you are not going to vote for him.
I don't doubt that there will people who continue to use upvote bots as you say for reasons like you mention. But at least they will have to pay more out of their own pocket to do it (there will be higher economic costs for such behavior as they won't get most of their payment back as author rewards).
The 15 minute rule to 3 minute rule change isn't about changing self-voting. I think it's completely neutral on that point. It's mostly about saving people from losing their curation rewards because they don't know about the 15 minute rule.
New users generally won't know this rule and even a decent number of active users probably don't understand it. Even if they do, they probably don't want to wait 15 minutes to optimize their curation rewards. This kind of PITA makes you not want to bother with getting curation rewards. With it taking so much effort to get them, a lot of people stop worrying about trying to get them and just "pile-on" vote on already popular posts.
Self-voting is potentially decreased by enabling you to profit more by voting for someone else's post versus self-voting. I say "potentially" because it will only "work" if people understand the new rules enough to act in their own economic interest. And we can't be sure that these new rules will be a sufficient economic incentive to reduce self-voting even if people do understand the new rules, so it is really an experiment.
The last portion of what you said isn't correct. You don't get more curation rewards just for voting for a post that has more reward on it already (unless you're talking about compared to voting for a post that falls under the special "dust vote" case). This is one of those misunderstandings of the rules that I'm talking about that could render many rule changes moot.
Curation is one of those hard things to understand. So if I am the first person to vote on a post at the fifteen minute mark and make the payout of the post $0.30 and 19 more people vote on the post raising it to $2.00, I will get the same payout curation reward wise if I am the 10th voter on a post that has 20 voters and the payout is $20.00?
Any way, I am sure you have a lot more important things to work on, understanding and a place to find the rules and how things work on steemit is not an easy thing to do, most responses from witnesses are "it's on git-hub" "go look at git-hub", The FAQ for steemit is not really a very good tool, and does a pretty poor job of explaining things. There is no centralized help system because steemit is a de-centralized blockchain. Go to get hub or to steemit-chat or discord chat. That is almost everyone's answer to any question from a person with no programing or crypto experience. Thus a lot of confusion for the "simple" users, such as myself, and the seemingly stupid questions or misunderstood answers from those such as myself.
Reason why you should always look at the big picture. Your arguments have their merits and no one said bidbots are gonna die or peeps will stop self-voting. No.
The original point of this argument was to have a system where there will be very little difference between self-voting / delegating to bidbots and voting on other people's works.
50/50 does that. If the profit resides with bidbots, stakeholders will delegate to them. But if the profit resides with both bidbots and normal posts, stakeholders will at least think twice about what brings more value to Steem in the long term!
50/50 doesn't get to the point where there is little difference between self voting and curating. Self voting still gets 100% while curating only gets 50%. But it narrows the gap a lot from the current 100% vs. 25%. The idea is that along with other changes (especially cheaper downvotes) this is enough to achieve a balance.
Sorry for the delay answering, I missed your reply until now somehow.
I totally agree with you on the lack of documentation and I think it's a serious issue. I can't even fully answer your original question, because I don't know the exact algorithm and I haven't had the time to go delving through the code for it either.
And this isn't just an "academic" question: with users unable to know how the system rewards them, they generally come up with some guesses and react accordingly. If those guesses are wrong, even a system with finely tuned incentives probably won't work very well.
It is like playing a used computer game you bought with no rules or rule book, you just start plugging away and hope for a good outcome. There are no "cheat" codes available for the players. You try something it works you think that is the answer, then someone comes and tells you "your wrong", but not what or how you are wrong, just your wrong. It helps when people explain why someone is wrong in their thinking or observations, with out that they will just continue to be wrong with out knowing why.
Hey @blocktrades and @beshadow I was reading through these comments and just randomly thought I would drop in and give a few thoughts.
I also had a ton of questions as far as curation rewards are concerned, especially post HF20.. Right after the fork, I spent a lot of time testing and scratching my head on exactly how curation rewards work on the blockchain and how you can predictably measure your earnings and potential earnings given a set of different variables and circumstances.
I happened upon a discovery that made me so happy... It's a post made by @miniature-tiger. Although it was made 1 year ago, most of it still applies (except for the window being 15 minutes now instead of 30). I recommend you go check it out, it gets pretty technical, but I have yet to see anyone explain and distill it in such an easy to understand way as miniature-tiger does! (https://steemit.com/curation/@miniature-tiger/an-illustrated-guide-to-curation-from-the-simple-to-the-complex-with-real-examples-from-past-posts-part-2)
Here's an excerpt that I think may help answer your original question:
I hope that helps! This obviously gets a lot more complicated as other voters/curation window get added to the mix, but I've run some field tests and have found the calculations to be both useful and on target.
Hi, I'm deeply concerned with the idea of a 50/50 reward scheme...
The numbers just don't add up. At least this not where we are supposed to be driving the platform. Please read this post, I would love to have your feedback on the subject.
Thanks for wasting 4 minutes of my life :-)
No, but seriously that's a lot of math for a simple conclusion: higher rewards from investors to non-investors will spread the stake faster.
If the goal is just to spread stake faster then it's easy: just increase the payouts/inflation rate to hyperinflation levels, and it'll spread really fast.
And everyone holding Steem will sell it as soon as possible to find an investment that makes sense. Steem value will drop precipitously and ultimately the payouts will become worthless in real value.
I wasnt here yet when curation changed from 50/50 to 75/25, what is different today that negates the reasons to change then?
I've never seen a real explanation for the change from 50/50 to 75/25, but maybe I missed it. IIRC, this is also when the rule change was made that switched early curation rewards to the author, another change that seemed completely unreasonable to me (most voters would have no idea that by voting early they gave away their curation rewards).
The only thing that ever made sense to me was that the guy who decided to make the change was posting a lot at the time and those posts had very high rewards. AFAICT, mostly what the change caused was a lot more crap posts and self-voting of such posts.
Its looking more and more like the farther we get from the original design the worse things get.
Have we jumped the shark?
Mistakes were made, but I think they are fixable. My belief is that most of the changes required aren't that difficult.
The auction itself makes sense (if you can autovote immediately then the value of such 'curation' is low) but giving it to the author makes no sense and it is pretty reasonable in hindsight to view it as a money grab. The new method of returning it to the pool makes a lot more sense.
Someone looking to vote for profit (or bots doing the same) will vote at 3 minutes or 15 minutes or whatever is required. They will figure out it and play whatever games are required to maximize profits.
The idea of shortening the timer is to get it out of the way of new or casual users who aren't going to play these games. If you see something you like and it isn't heavily voted, you should just vote for it, not set a timer to come back 15 minutes later and vote for it. Nobody does that.
The timer is a game for bots, which does have some useful secondary benefits for the platform as a whole, so I don't mind having one. But let it run its course on "bot timescales" not get in the way on human timescales. (As noted in the post I think it should be even shorter, but 3 min is better than 15 min.)
I'm one of those that just vote pretty much when I see it sometimes i will wait a few minutes, I rarely see anything when it is fist out because I rarely look at the new tab. The only vote I don't like is the one I give on an old post 7 days old or more, have done that several times, and I do not understand how people can resteem things that are 2 months old, but glitches happen I guess. People voting for maximum profit are going to like you said, vote for maximum profit, and there really is nothing wrong with that.
3 minutes works, it is long enough to see if you want to vote or finish reading the post.