Feedback: Proposal: reduce Hive inflation by reducing curation rewards

in #hive3 years ago

In a comment to @blocktrades post, I said I would provide feedback later that day. That day turned into a month so now it's the time to put my thought on this to paper.

https://peakd.com/hive-102930/@blocktrades/proposal-reduce-hive-inflation-by-reducing-curation-rewards

At first, I didn't like the change, I immediately thought people would still use an auto vote, because a Nickle is still better than nothing, and then I thought about the imbalance of disrupting the 50/50 which I belive helped make bid bots obsolete. Higher author rewards mean selling votes to become more profitable by a very large margin. We played this game before, and it wasn't a fun game, trending was trash. However we do have free downvotes now, and I do feel with lower curation, stakeholders will want to downvote more to protect their stake.

So in a way, stakeholders will still need to monitor the platform to stop abuse, and sure we can rely on a group or groups of people such as Marky or HiveWatchers to fight spam, but there is still some friction there. By lowering curation rewards, this brings back all the incentives to abuse the platform, which people will, and the downvotes will need to rain in strong and consistent to stomp out any bad habits.

"After thinking about this for a while, I believe this is probably the simplest near-term solution to the economic problem that is currently pushing curators to use auto-vote bots.
Most stakeholders know that auto-vote bots aren’t great for Hive as a whole. But it’s a form of tragedy-of-the-commons, where each stakeholder who doesn’t join in the auto-vote game loses out against those who do.
To break this negative economic incentive, I propose we lower curation rewards to the point where the potential gain from curation is too small to make it worth the negatives associated with using an auto-vote bot. This will bring the situation back to one where stakeholders who don’t want to spend time curating but don’t want to lose out on curation rewards, don’t need to worry about this issue anymore, because the relative loss between the two choices is small. This could even incentivize new stakeholders who previously have decided to pass on Hive or only maintain a small investment because they didn’t want to play the curation game."

Things lowering curation does accomplish, and if lowered enough, a stakeholder will no longer need to vote to compete vs inflation. However, if going this route for the short term, as a stakeholder I would like it if when I didnt vote, my inflation wasn't given to everyone else. I would like to think by not voting I am lowering the overall inflation, combine this with lower curation rewards, it almost becomes a wash, as if I'm not inflating hive from my stake it means I am further protecting my stake from dilution.

To take this a step further, I rather downvote BURN Hive, this will give stakeholders even more reason to use downvotes to fight abuse, and spam can be seen almost as a good thing as it burns Hive and lowers the circulating supply.

If we are to game theory this, in a perfect world you would have no curation rewards and people would simply upvote to distribute the token. They would be wise with their votes as every vote further dilutes their own. Then this could be seen as a cool way to onboard new users, reward good efforts, and build communities without the feeling of needing to vote. IT would be the opposite, people would feel the need NOT to vote, to vote only when important. When you add curation rewards, now people feel the need to vote because of the opprutitny cost of not voting. However, curation rewards in practice also encourage people to seek out good content to vote to avoid downvotes (auto or not.)

So in practice if we removed curation rewards we would see a lot of abuse, downvotes can certainly take care of the abuse but it will be constant work. 50/50 curation if anything stomped out the need to abuse (self vote/sell votes) but it also forces you to play "the game" not every investor wants to play.

So that's the pickle, removing curation rewards means more abuse control, and if the point of removing curation was to ease the burden of the investor, while on paper it makes total sense, in practice I think we will have lots of raising 72o from UTG, aka a lot of people going crazy with abuse.

If I had to make lower curation work it would be as follows. Expand the curation window from 5mins to 1 day, make all curation on the 1st day equal. This means every time a stakeholder logs on, whatever the day is, they can find fresh content to upvote without the need to use Autobots. With lower curation, they won't feel the need to keep up with the Joneses, and curating becomes a much more relaxed effort. I would also remove the curve to bring back micro tipping (a house sell for hive) - The drawback with this approach, as I said before, is bid bots and vote-selling will try to come back and we will need a strong downvote force to make sure this doesn't happen.

In the end, my personal goal would be as follows, keep the "tipping" part of layer 1 for distribution, but find a way to not have to play the game if you don't want without fear of being diluted. This would be getting the cake and eating it too as far as I'm concerned.

Sort:  

After reading thru the many responses on my original post, I also concluded that an expansion of the curation window, with all curation during the window equal, is the next thing we should try. It's one of the things I want to see in the next hardfork.

Finally. :)
I am looking forward to these changes.
(Look how late again I discovered, read, upvoted and commented for example this post ... Maybe in future there will be some new hope for extra slow persons like me ...) :-)

I don't really understand the reasoning as to why there is just a 7-day window to vote. I might be missing something here. As I have stated elsewhere, I am not nuanced in HIVE economics, but have a sophomoric understanding of how all of it works.

Then again there are the changes every hard-fork when I think I have gotten a decent understanding.

I suppose this is something similar to what they are doing on Leofinance, by flattening the curve. I, as a fully organic voter would like to see this.

Even though I love flat curation and was shouting about it for a long time, I really think we should leave those things for times where there is less work in front of us.

We keep tweaking those algorithms, and we keep saying it's the last time. Let's stop for real. It's getting annoying.

yes plz

@leofinance did an interesting move recently to leveling the playing field with auto-voting bots, they removed the time window (made votes time independent) and switched back to a linear reward curve. Maybe this is something Hive can also adopt if it turns out to be successful. Only reducing curation rewards will remove the interest in voting and curating I am afraid. https://leofinance.io/hive-167922/@leofinance/leveling-the-playing-field-leo-is-switching-to-a-linear-curation-curve

Fully agree...ditch that stupid time window!

Leave the other stuff alone

I like what they are doing over there at LEO, I would do the same for Hive.

I understand the need to make changes for large stake holders but it's challenging because it often negatively effects those who spend lots of time doing things like manual curation. Just when folks like myself are getting a decent stake (who have made investments in Hive, not freeloading) I see ways to shake it up in a manner that would hurt progress I'm making.

To BLocktrades comment, I would love to see an even playing field on the curation window but that can certainly also be gamed but I would like to see how that proposal shakes out.

Dialogue is important though, we can all agree to that so thanks!

Expanding the curation window will help a ton with autovotes, as the current setup pretty much forces you to use autovote or stare at the screen all day to find post within 5mins. If a lot of votes hit before you vote in that 5mins you give up a nice chunk of rewards, if you do that for a while you miss out on a ton, I know for experience ;)

Expand the curation window from 5mins to 1 day, make all curation on the 1st day equal.

@leofinance has gone to a pure linear curation model at 50/50. It does not matter when a person votes and one receives a reward commensurate with their voting power. Makes it real simple. All this 5-minute, 15-minute stuff, to me, is just confusing.

The drawback with this approach, as I said before, is bid bots and vote-selling will try to come back and we will need a strong down-vote force to make sure this doesn't happen.

Scammers gonna scam. It is a shame but they are here and are not going anywhere. In the three years I have been on this platform I have seen it. Shoot, Jerry even made posts on how to scam the system and he was a witness (top 20 for a while). Ethics can be taught, but being ethical is a choice. A number of witnesses had upvote bots which were totally anti-theoretical to the philosophy of the original STEEM white paper, so there you go.

I personally believe that the community tokens will end up being primary and the HIVE Blockchain Coin will be secondary. @leofinance is an example of this happening. I am going to quit chasing community tags and just focus on the content I like. Should I wish to author something, I am going to write and tag to everyone, not just HIVE. Sure, might miss out on some votes but we want to onboard more people... right? Of course I will probably make a HIVE-centric post now and again. Old Sarge needs to make a nickle.

I appreciate your take on things. I don't know enough about the politics (hey everything is politics) and all the economics but I know enough to throw in my two shekels as a low-level dolphin.

Blessings and groovy vibes my friend. Stay safe!

Fully agree with the fact second layer tokens (maybe 1st layer native tokens as well, or hybrid?) will take over Hives role as the main value for content creators. The reward pool was never for the social aspect IMO as I said in my previous video. One reward pool will never scale, once we have millions of users, a dozen hive whales could not keep up with the curation to make it scale. Native tokens for niche communities creates endless curators incentives by their own token to find great content. Have front ends that tie in all communities as if they where channels on youtube, and you have something very special. Communities with the look and feel of their own sites (well they are) like leofinance.io but tied into a giant network like a hive.blog or peakd, that is where the magic is.

Why sure! Now we need to get folks like UplandTM that post on MediumTM to post here instead! Though HIVE can do bunches of stuff it primarily hosts blogging as you know. Time to get inclusive and collaborate with other projects.

I really like the idea of using actual staked hive power to cast your vote. That enables more manual voting and doesn't make it necessary to cast your 10 auto votes a day, rather than picking out deserving content.

I myself am totally against any changes to the current reward pool system. I have read many different comments and post on the subject. For the 3 years I was on Steem every hard fork involved trying to fix the reward pool.

The reward pool system has never had a chance to stabilize. No changes should be done to the reward pool system until we have the foundation of Hive solidly built. Hard Fork 24 was a great start, we are now done with steem and now the next steps need to be done.

The witness selection/retention system needs to be worked on first. Hard Fork 25 should be about completing the Smart Contracts, (SMT/HMT), and the witness system.

If we as a group are going to do as steem block chain did and continue to chase the reward system then there really is nothing to differentiate Hive from Steem block chain.

First of all, trending still sucks, and any change thought to improve is better than doing nothing, I support you.

I disagree here. Trending, while not perfect, is exceedingly better then trending pre EIP.

After reading thru the many responses on my original post, I also concluded that an expansion of the curation window, with all curation during the window equal, is the next thing we should try. It's one of the things I want to see in the next hardfork.

I tried to delete this and reply at a more appropriate spot, but you'd already upvoted :-)

haha they call me ninja Dan

Check out Trending on Steemit :)

I can't. It is all in Asian language scripts and, like most ugly Americans, my international linguistic skills are lacking!

I agree with you there, but maybe I didn't explain myself well, I know you vote for what you like, and because you support the author because you know it's good content, but who hasn't seen a post with 0 reward go to 10 or more with just one vote? There's always a way to get around the system and still get rewarded for it @themarkymark

That's why I think your idea is good, trending is better, yes, but it still needs to be improved, that's where your idea comes in.

to be fair this isn't my idea, I borrowed most of the ideas to form this plan. things like expanded curation windows have been around for awhile and I credit blocktrades for the lower curation idea as I have not heard that one brought up a lot.

I propose we lower curation rewards to the point where the potential gain from curation is too small to make it worth the negatives associated with using an auto-vote bot.

Lowering curation rewards makes author rewards (and likely bot abuse) relatively higher, so back to the same problem as before.

The recent change to LEO’s curation curve can serve as a test for what might happen if Hive adopted something similar.

That said, the underlying idea that Hive’s inflation rate is “too high” seems a bit suspect to me. If/when the user base is 100X what it is today, inflation that’s already less that 8% won’t seem high at all.

I agree, the inflation being too high is not the issue, the crux of this is about investors losing out to inflation. If passive stakers got the inflation then it wouldnt matter, but the rub is forcing investors to play the game of curation in order not to get diluted by up to 20% per year. Eth has had a relativly high inflation since inception, if you add the premine selling ontop the inflation would be double digits, its def not the 8% inflation of Hive that is a big issue.

If investors are the people we want to attract it would be an Idea to distribute part of the inflation to LiquidityProviders for wHIVE and to let them invest there. Rewards on-chain would be in better hands of active users as these people fill the chain with life. Pushing them into autovoting for economic reasons is indeed problematic. Strengthen wHIVE with revenues would couple the HIVE price to ETH and may be a fast way up during the coming bullrun. Otherwise we may miss out this big chance.

I thought I read somewhere 3% of inflation went to passive investors? I could be wrong. I get mixed-up sometimes.

Not quite. The current ROI on HP is approximately 3% and 15% of the inflation goes to hodlers of HP.

About every three months, I crunch the numbers and do a post explaining how much hodlers of Hive Power are currently earning. Here’s the most recent one from 3+ weeks ago:
https://peakd.com/hive-167922/@preparedwombat/the-hive-power-you-earn-for-staking-hive

Thank you for that clarification.

I think this is a very good proposal and hope it's choosen

I somewhat agree with you

The curation reward is what keeps this platform going.
If you want to take this platform up you need 2 things -

  1. More decentralization - top 20 witnesses is a small centralized club.
  2. Create accounts via fb and google accounts to increase usage.

20 block-producers/witnesses with one in reserve is why our transactions are so fast. Ethereum is playing with the idea of doing something similar with their protocol which will be a hybrid of Proof of Stake (PoS) and Proof of Work (PoW).

EOS also uses the Delegated Proof of Stake (DPoS) protocol. Too many block-producers/witnesses would slow things down.

I notice that you are currently only voting 6 witnesses. Have you checked the witness list lately?

This is not accurate... The algorithm that is choosing the leading witness for a block gives crucial weight to the top 20.
Distributing the weight won't slow the process, in fact it might even speed it up.

That is incorrect...

Ok, Can you please explain to me why distributing the major mining to 30 witnesses (instead of 20) will affect performance?

I can not explain that to you as I am not that blockchain tech savvy. That would be a question for someone like @themarkymark or @inertia. They are smart in these things. Hope they don't mind me tagging them, but they are very knowledgeable in these topics, especially @inertia.

By increasing it to 30 witnesses you add a few areas of concern. But I first want to be clear, we do not have mining.

Confirmation times go up dramatically, right now we have roughly 51 seconds to confirm a transaction, now most people use the head block and only wait 1 block (3 seconds) but any transaction in theory can be rolled back before the super majority confirm it. This is called an irreversible block, but not every system waits this long, most don't. With 30 witnesses, this will be longer.

You also need to have more people physically available when doing a hardfork or emergency patch, this is can be a bottleneck in critical times.

Most things won't change at all, as most things are using small transfers, they will not wait for an irreversible block.

I told you he was smart @amirl. That is why he is one of the witnesses I vote for. You might want to vote his witness too! If you wish to make sure that our blockchain stays decentralized then your vote counts.

Many people don't vote for witnesses or are voting for inactive witnesses. I realize this is not the topic of this post, but we must all do our little bit to improve things, I believe.

I guess I need to open a node to fully understand ;)

Our protocol is Delegated Proof of Stake (DPoS) slightly different from (PoS) in that top 20+1 witnesses are there based upon stake-based voting.

The code or algorithm as you describe it does give the top 20 priority in the opportunity to produce blocks. That is correct. But that is how DPoS works. This is the closest thing we have to democracy on our blockchain. Oligarchical? Yes. But that is how things work. Actually that is how things work everywhere and have for ages.

There are many more than 20 witnesses, just 20 are elected from 100's to run the chain. The more nodes you have the slower your chain and without on chain voting you end up with messy hardforks leaving the protocol static.

We have dapps that created "guest accounts" for users to login with things like FB/Google. Moving comments/post offchain until they get a blockchain acct can remedy the situation until we get lite accounts, then we can scale much easier with native hive tokens and instant account creation.

Voting and number of confirmations is ok, but the 20 miners is a monopol that reduces the trust in the platform. Decentralization is what this network is all about and the top 20 should show trust building steps such as distributing the election even to 25-30 in the next release.

Define usage, in fb many memes and simple things are used to pass the time and to have fun, here all that is badly seen and the vote is concentrated in software, literature, and gaming. 🤔

This is resolved by the rc. Currently after 4 years most people never even heard of the platform, the access must be friendlier.

I mean, in an audio visual world I would like content like fb's not to be penalized here, that would attract users and potential investors for a long time.

If we want the price to go up, really we should cut inflation in half. Maybe when smts are done we can only have inflation for paying witnesses and the dao and abolish hive power, simply holding hive in your account should count.

The 5mins time window should be extended... But don't touch the curation reward whatsoever...

Why 1 day curation window? I would remove completely the curation window and, for sure, I would remove the unfair rewarding Curve urgently.
I disagree with the idea of changing the 50/50 curation rewards for the moment.

'Ban the bot' that is what I would like but I know too well that this would not be possible.

Would there be a way to reduce the voting power for each subsequent vote for the same person. Like an increasing tax sent to null with each vote they make. This is Just a thought and not fleshed out too well.

I do like the way Leofinance has done it by flattening the curve and making it more fair for people who actually vote manually and trying to promote better content. Or, maybe we could look at the tipping model like on Whaleshares.

I am worried that by trying to lower inflation at the cost of curation it will seem that the bigger stake holders are just bringing in a change to suit themselves at the expense of smaller/newer users.

@theycallmedan @blocktrades solution, and the ability to use "" HIVE DOLLARS and HIVE DOLLARS POWER
hive power and hive"", we make hive an investment and HIVE DOLLARS POWER les récompense
we have to add HIVE DOLLARS POWER and be a real stablecoin. to explain my point of view and make it easier to calculate suppose hive its value is 1 hive = 1 dollar and 1 HIVE DOLLARS POWER = 1 dollar .
I start with the publication awards :
it's the same thing 50% author 50% conservative
or can change it to 75% author
because the Conservative would have other way to earn more.
more explanation in investor.
the rewards will only be with stablecoin HIVE DOLLARS POWER.
investor hive :
the more you buy "hive" the more you earn "HIVE DOLLARS POWER" which will be used only for the vote, example : you have 1000 hive power it pays you 2 HIVE DOLLARS POWER every day ""the calculation will be on the value of hive"".
2 HIVE DOLLARS POWER you can only use them to recompose the authors.
and when a curator votes 50% conservative in HIVE DOLLARS "HIVE DOLLARS liquid" who can use as he wants,
so we can encourage investors to stay in hive power, and encourage investors to invest in new business or application.
abstract :
hive power makes you win HIVE DOLLARS POWER everyday.
hive power you can lend is win hive and HIVE DOLLARS .
HIVE DOLLARS POWER for voting only
HIVE DOLLARS and hive for trade
the authors earn HIVE DOLLARS who can sell it or invest it.

Expand the curation window from 5mins to 1 day, make all curation on the 1st day equal.

I would love to support this change and see how it works!