This has been a demonstration of optimal posting strategy under Hardfork 20.

in #steem6 years ago (edited)
This has been a demonstration of optimal posting strategy under Hardfork 20. For those coming in late, this was the original appearance of this post:

15.jpg

I edited it at the fifteen minute mark to add content.

Why did you do that?

Under Hardfork 20, when a vote is given in the first fifteen minutes of a post's life, it will accrue curation value, but that value won't go to the curator or to the author of the post. Instead it will be redistributed among all posts in the reward pool in proportion to how large their rewards are.

So when I vote early a bunch of my curation rewards go to Banfield and Haejin?

Yes! You have it exactly. But it means more than that - every time someone votes on the post after you, some of the value of their vote goes to your nominal curation reward. So not only are you supporting Banfield and Haejin by voting on the post early, you're also tricking later voters into doing it.

That seems like a bad idea.

Doesn't it? But wait, it gets worse! If nobody votes on a post before fifteen minutes, the total rewards for the post's curators will be 25%. But for every vote that comes in earlier, that percentage decreases. (This is how it works now, too, except the extra money goes to the post author.) So you're supporting Banfield and Haejin, you're tricking other users into doing the same, and you're taking money away from other people who voted on this post to do it.

OK, but at least that's as bad as it gets, right?

That also means that people who want curation rewards will be more likely to pass up voting on this post in order to vote on one that hasn't had votes in the first fifteen minutes. When you upvote a post in the first fifteen, you will be discouraging other users from upvoting it later. So everyone who posts will be trying to keep you from doing that. My method might be a tiny bit over the top.

OK but it's just posts that are the problem, right?

Wait, there's more! This also applies to comments. If you upvote a comment in the first fifteen minutes after it's made, part of your vote will go to rewarding popular posts in the pool. So you have a choice between feeding Banfield or delaying your votes and making your comment section less friendly to active, immediate discussion.

This seems like a bad plan.

Now you understand why I made this post.

Sort:  

For someone who's not technical your observation are pretty astute

Here is a brief of the new logic:

  1. Discourage Self upvoter that guarantees position 1
  2. Forces Poster (with self or alt account) to set auto self-vote at 15mins and now have to compete with 1000 other minnows also with curation autovote at 15mins
  3. Puts back all curation reward in first 15mins into reward pool

Logic of your Argument (just using round numbers for logic argument sake):

  1. Let say Pure Reward Poll Abuser make up 10% of all reward today, bots make up 30%, and everyone else makes up 60%:
  2. So 40% of these Re-Pool Curation Rewards will go back to RewardPool Abuser Curation and Bots Curation from that 25%. Simplistically said that's max extra of 16%.

Q.E.D.

BUT Wait. Until Now Big Reward Pool Abuser is getting like 70-90% due to position 1 of self-upvote mechanism of that 25%. Losing that they may now go down to 40-60%, losing a minimal of 20% of curation.

Last i checked 20% > 16%

So it's still good. BUT this is a great argument to had, Kudos. Upvoted!

Sure, as long as you only care about how much abusers lose and not about how much anyone else loses, how easy it is for new users to understand, or how readable the site is. If you're just willing to throw all of those things out for your goal, then great, you've reduced a few abusers by a few percentages! Awesome!

Sure, as long as you only care about how much abusers lose and not about how much anyone else

1stly don't personalized as investors we all care about wholelistic goals of steem growth...if I just wanted to proof anyone wrong, i would be accusing you of sensationalizing a none-issue or a non-event, instead of honoring the discussion

how easy it is for new users to understand, or how readable the site is

that's a different issue and animal to address, there will be platform running off the blockchain that will address usability, but to be fair usability will also differ from generational usage to different segment of society, eg. those younger who love snap-chat may prefer to use appics and those who read wallstreetjournal online may prefer busy

If you're just willing to throw all of those things out for your goal, then great, you've reduced to a few abusers by a few percentages

you do know that the reward pool payout is like a zero sum game, the less an abuser get, the more there is for others? no sarcasm intended, just poiting out the obvious

Awesome!
you're welcome.... pardon my conceit, can't think of an milder respond to your baseless sarcasm

Okay. So the way I understand this works now is the self-upvoter keeps all of the rewards that they give themselves within the first few minutes of doing so, including all of what would go to curation, with diminishing returns up to and then through the 15th minute, as well as getting up to all of the portion of the curation from the upvote of others depending on when they place their upvote.

If I am understanding that correctly, that means no one upvoting now should do it within the first 15 minutes if they hope to have some of the curation rewards.

With HF 20, the way I understood the change would work is that the portion of the self upvote that could be set aside for curation after a few minutes through the 15th and so on will instead go to the reward pool. Now, you're saying that anyone else who votes within that 15 minute period could have at least portions of the curation reward end up back in the reward pool and distributed to the posts with higher rewards, too, which if such is the case, is pretty self-defeating.

I guess I'm going to need to go back an reread the last few updates. Also, I believe the window is being reduced from a total of 30 to 15 anyway. Haven't seen that talked about lately but it was in an earlier update.

Regardless, it seems to me that this could be more readily solved if instead of returning to the reward pool, the allocation went to curation for that particular post. Since some of the higher SP folks are asking for a 50/50 split to encourage more curation, this change as described will be taking a step back.

you're saying that anyone else who votes within that 15 minute period could have at least portions of the curation reward end up back in the reward pool and distributed to the posts with higher rewards, too

Yes, that's how it works. All of the early-voting curation rewards that now go to the author would go to the pool instead. (If you think about it, just trashing the self-vote rewards doesn't help because you can always use a bot vote or a dummy account instead.)

Regardless, it seems to me that this could be more readily solved if instead of returning to the reward pool, the allocation went to curation for that particular post.

That would definitely be a better solution. A big self-vote at the beginning would be an incentive for more people to vote on the post, which I suppose might be an issue, but it would be an improvement over the proposed method and might be an improvement over the current method.

when is this ridiculous change going to take place?

And I'm going to tag a few of my friends and see if they are aware of this... What do you all think: @abh12345 @paulag @danielsaori @jackmiller @sircork @comedyopenmic? You are all witnesses, is this something you agree with or disagree with?

when is this ridiculous change going to take place?

We don't know.

I'm going to write up a new post with a clearer discussion of this issue based on the long discussion with eonwarped here, if anyone you tagged wants to wait for that. This one is really something of a shitpost that went viral somehow.

@tcpolymath are you backing away from your assertion that this is taking money from the people's posts and sending it to the few that can afford to be "popular"? I hope not, because I think you hit the nail on the head with this one! I am eagerly awaiting your next post about it, if you think to let me know when its out that would be awesome!

are you backing away from your assertion that this is taking money from the people's posts and sending it to the few that can afford to be "popular"?

No, it's just stated differently. Destroying your rshares - whether it's this method, voting on declined payout posts, flagging, or just sitting at 100% VP - distributes that part of your stake reward to active posts based on popularity.

I don't have plans for yet another post on the topic at this time, but nobody's come along to argue with me on the most recent one yet.

So I read your first post right?... I want to make sure because I want to make a stink at least a little if I read it right. Basically, they are taking money from the post and paying "someone other than" the curator or the author. And that someone will be the people who can afford to buy their posts to the top of the chain. Is that about it?

Maybe the goal is to encourage/force people to buy into the system by raising the barriers for those that wish to join through posting content and curation? I can see how this could be attractive to an administration.

Thanks for bringing it to my attention @davemccoy. And thanks Tim for highlighting it in this post.

I have to do some more reading to get a better understanding. Don’t recall this as a conclusion after studying the initial hf20 proposal. Or is there anything new, recently suggested?

When you get time it would be great to know what you thought. It does seem like they are taking money from people's posts and then giving it to reward the higher earning posts! The rich get richer!

On reading the initial detail of the voting window/curation rewards change 6 months or so ago, this was not how I understood it. (And thought the rewards would be put into the pool of the actual post being curated).

I'm hoping the pool is bigger for all, but it sounds like big votes (bot votes) could potentially scoop up most of the 'extra' rewards. Not ideal if so :/

As always, more reading required - tag me in if you spot something Dave, cheers!

@abh12345 I look forward to you learning more about it... If it was how you read it the first time, I would say nothing...I frankly don't care who gets the money from the post (author or curator)...But taking money away from a post just to add it back to the pool, just adds another layer of absurdity that we will have to adjust to...And the optics are horrible.

It will be interesting to see how it pans out, and I'm hopeful it's not as bad as folks suspect. STEEM at 1.7 right now, about time we had a lift!

I'm part of the @noblewitness team with @sircork, and if I'm understanding this correctly, I don't agree with this at all. As stated above, this is altogether worse than the reward distribution system we have now.

I mean it shifts the time, people adjust their bots, life overall, doesn't change much at all, and the way I read it, the spurious rewards go back into the pool, not to the high profile stuff some are reading it as. I could be wrong. Your mileage may vary. Always assume there is reason though for the "top" to spin things in ways that don't hurt them.

My thoughts exactly! I hope there is some pressure that you guys can put on to stop it... If they are going to take away from the author of a post, then they should give it to the curators of that post. It shouldn't be given to the "pool". In my opinion of course. Thanks for the reply! and nice to meet you @anarcho-andrei!

@davemccoy I am looking at some data on this, will respond with a post

Regardless, it seems to me that this could be more readily solved if instead of returning to the reward pool, the allocation went to curation for that particular post.

This is how I read one of the earlier @steemitblog updates regarding HF20 and agree with you that a better option might be to 'boost' the curation pot for the actual post being curated.

How with this pan out? Sounds like 'more for everyone', but could actually be more for the ones at the top :/

It's more for the ones at the top but it's less more for them than the hf20 proposal, if that makes sense. Moving from the current system it would take away from authors (particularly people who are exploiting the bug but also some innocents) and distribute that in a stake-weighted fashion among the voters on the post. So it still benefits the biggest voters on each post but it doesn't give value to the site's biggest voters from posts they didn't vote on like the hf20 proposal does.

Anything that increases the percentage of rewards paid to curation is always going to be more for the top in some fashion unless the core algorithm is changed.

no, proposed solution is better. Giving the curation share from a big self vote to curators of that particular post is very easily gamed, and would be gamed. All an author has to do is set a second account to upvote the post at optimal curation time to get the reward from the self vote. Returning it to the pool is much fairer as it is impossible to game by the individual self voter.

All an author has to do is set a second account to upvote the post at optimal curation time to get the reward from the self vote.

They can still do exactly this by waiting fifteen minutes, voting with the sockpuppet account, and immediately self-voting.

yeah fair enough that is exactly true :) Ultimately, proposed change as is still decreases the % of the reward pool dominated by the self voting behavior more than returning that reward to the reward pool of the individual post. it isn't a hypothetical "if" a huge account that self votes will set up a bot system to maximize return from the self vote under any new system, it is a certainty.

Think about the huge sums we are talking about at the top end - the haejin's of the world - I think it is clearly better that the large share from the self vote should be distributed to the pool vs. stay in the post where ranchorelaxo will get almost all of it. And also think about what you want to be motivating - do you really want to motivate people to upvote haejin and similar posting so that they can get the curation reward? Whatever small chunk of the extra curation returned to the Haejin post pool under your proposal that is not gobbled up by ranchorelaxo will go to users in the proportion that they upvote a haejin post... LOL

You would rather increase the curation pool on a haejin post vs. send that huge chunk out to all the other posts? Think about it as simply as you can. I am not taking the time to look up what % of the reward pool haejin dominates right now - lets call it 10%. If the chunk of curation from the haejin self vote is returned to the reward pool at large, haejin only recovers 10% of it. If that chunk of curation stays within the haejin post itself the haejin/ranchorelaxo combo recovers almost all of it and whatever they don't recover goes to people who upvote a huge self vote abuser. I don't think this is what we want to incentivize.

You are better at reading than me... lol (I was good at math in school though ;) )

If you do get into the weeds and figure this out, please let me know your thoughts Glen... I think it sounds horrible and will make things even more lopsided on the surface. But you always have a good level head, so I would like to know what you dig up!

Loading...

and they wonder why steem is weak and we have lost like 95% of our new users

Steem needs time, and I'm not sure the community gives it enough time.
Lots of proposals, not much actual work!

@enn0 I agree... unfortunately they are making the biggest mistake to any customer centric organization... They are not listening to their customers. If they would listen to the customers, then they would get time... Frankly its only due to the terrific product they have that people have stayed this long (that aren't rich I already I mean).

Maybe its the beer but not following that well. WOuld love to see this post written in an if then process.:)

If you self vote early you do not get rewarded. If you vote for anothers early it goes to a pool where those who are the whales will get the unfair share? Is this along the right lines?

If you vote for anothers early it goes to a pool where those who are the whales will get the unfair share? Is this along the right lines?

Yes. If you vote for a post or comment in the first fifteen minutes a portion of your vote and all the votes after yours will be distributed to all of the posts on the site in proportion to how many votes they have.

hi @tcpolymath, found this post from the ping by @davemccoy and his follow up post. I decided to have a look at the data and see what the impact of this would be. as it currently stands with these changes Haejin has the most to lose, but I am sure that the votes will be changed from before 15 min to after.

I put the data into a post so that people can make an informed decision based on some actuals. You might like to have a read.

https://steemit.com/analysis/@paulag/hf20-exploratory-data-analysis-on-proposed-payout-changes

So let me get this straight.

During the first 15 minutes a post or comment is live, vote value goes to the author in a declining portion from 100% at the instant something is published down to 87.5% at 15 minutes.

And curation value that goes to the reward pool increase from 0% of vote value up to 12.5% at 15 minutes.

Is that correct?

If so, it's a marginal improvement to the rewards pool (which includes jerry and haejin) in order to accomplish... what?

It won't curb self-voting. Anybody running a curation trail can just alter the timing parameter slightly. I guess it is a small deterrent against some voting rings.

I don't know that I can answer all the specifics, though this might be close enough. I don't have time to read through the code today and figure out the methodology; I might do that next week.

Right now what happens is that the early vote bonus comes in the form of assigning a higher percentage of the post rewards to the author. My guess is that they're just going to cap that at 75%, so any value beyond that goes to the pool. But I'm not sure about that.

As far as I can understand it the main goal is to stop people from doing what I'm doing: self-vote and/or buy a chunky bot vote in the early minutes of a post to increase the percentage that goes to the author, and then buy a larger, adjusted vote later to take advantage of the middleman's curation adjustment being set to 25%.

It's probably not a bad idea to stop that exploit, but this is a terrible way to go about it.

During the first 15 minutes a post or comment is live, vote value goes to the author in a declining portion from 100% at the instant something is published down to 87.5% at 15 minutes. And curation value that goes to the reward pool increase from 0% of vote value up to 12.5% at 15 minutes.

Reading this again I think it's backwards. It's the "vote value that goes to the author" that they're getting rid of. The voter will still get the curation that goes to them now - the curve is steepened so it's 25% at 15 minutes - but the portion of curation that doesn't go to them will go to the rewards pool.

In practice at the moment that can easily be 12-15% of the total post rewards. So not chump change. It will undoubtedly get smaller as people stop voting early, though.

It does lead to a weird possibility of aggressively voting early in order to reduce a post's rewards, too.

Ah, I think I understand that. So they are removing the author's share of curation during the first minutes the post is live.

Of course that's a bit circular since the rewards pool goes in large part to the author anyways.

It might do something to slightly curb bid bot usage since (aside from minnowbooster) they don't adjust the vote value to account for curation.

Smartmarket also does, in fact they're better at it than MB is. But yes, the actual bidbots require you to do any curation math yourself, and I suppose making the curation math harder might hurt their market a tiny bit. But what I see on the various support forums is that there are a ton of users who aren't that sophisticated and judge whether the bot's vote size is enough for them by completely nonsensical criteria.

I guess bot users losing money is just a second-order version of the same effect.

So if I understood it right this change will only affect authors. Curators will loose nothing and even get more since the part take from the author is distributed as curation reward. My problem with this is that is goes to the popular post, knowing that most of the curtation reward of popular post goes to bid bots, it means bid bot owner and curator get more and the authors less.

So if I understood it right this change will only affect authors. Curators will loose nothing and even get more since the part take from the author is distributed as curation reward.

Sort of? It should help curators a little bit as people being discouraged from voting in the first fifteen minutes should lead to real curation distributions closer to the nominal 25%. But as @glenalbrethsen pointed out they have the opportunity to give the full 25% to curators instead of doing the rewards pool nonsense, and chose not to.

It also hurts readers and unsophisticated voters who will have to deal with efforts by authors to not get votes in the first fifteen minutes.

Yeah this solution could also be good.

Complete and total bull shit. Thank you for making this post as it is how I was picturing it yet so many people are acting like this is a good thing. Probably because they are sheeple and didn't bother to think about who would gain from the rewards going back into the pool.

Omg Poly you are hilarious at times, and the best part is that it's just you turning up the sardonicker to 10 .

I don't like that particular change myself, but it might be a move in the right direction... possibly... maybe.... it remains to be seen.

I would upvote your comment now but I'd be giving money to Haejin. I'll wait until it's 15 minutes old.

This gif summarizes how I feel about the change

When is hardfork 20 happening?

Yeah, pretty much. We don't know yet when they're going to release. There are some things in it that promise to be really nice as well as this one and a couple others that are problematic.