A response to proposed changes to witness/miner rewards in the next Steem hardfork

in steem •  2 years ago

The proposal for changes to the Steemit economy would abruptly change STEEM inflation to 9.5% APR, including a reduction of witness (including miner) block creation rewards to 10% of the new yearly inflation. The proposed changes to witness and miner Steem Power reward distribution in the tentative hardfork that Dan has proposed would rebalance witness rewards by reducing the top 19 (full time) witnesses' SP rewards to 1/29th of all witness rewards.

The backup witness and miner slots will each receive 5/29ths of the daily witness rewards. Part of the rationale for this is to give backup witnesses a more proportional reward, which I support. Backup witnesses have votes from stakeholders in Steem and generally are contributing in some way to have earned them. However, this proposed change would give the miner slot 5 times the SP reward as a top 19 witness.

Daily numbers

At the current supply of STEEM (214.2 million), multiplied by the proposed annual inflation (9.5%), multiplied by the proposed witness cut of that inflation (10%), divided by 29 and divided again by 365 days, we come to an estimate of a top 19 witness being rewarded 192 Steem Power a day. Multiplied by 5, we see the miner slot receiving around 960 SP per day.

This is roughly an 86% reduction in SP rewards for top 19 witnesses and a 30% reduction in SP rewards for the backup pool and miner slot.

Mining queue monopoly?

Given the situation that the Steem mining queue is in, this would effectively reward two people, perhaps one, with outsized SP rewards compared to the top witnesses.

If you don't know, the mining queue has been dominated for months by a couple sets of accounts. One set is named rabbit-00 through rabbit-79. Another group, gxt-1080-sc0001 (to 0150) and supercomputingXX (120 accounts) are claimed by someone who goes by 'nobody' on steemit.chat. Interestingly, at almost the same time a few days ago, the rabbit and supercomputing accounts both started mining after a hiatus (11 days for supercomputing and about 14 hours for rabbit). It's not definitive evidence they're owned by the same person but it's too coincidental to ignore. Here are their first PoWs of the day:

https://steemd.com/tx/890ef0cb8de6d16b538d3c0e198a1951c68a26cc
https://steemd.com/tx/440a800cda5e472c60b0a81884fd7f554f9aeabb

16 minutes apart. You can verify these are their first PoWs after the hiatus by visiting https://steemd.com/@rabbit-00 and https://steemd.com/@supercomputing-0

The mining queue thus may be monopolized. The person who claims ownership to the 150 gxt-1080-sc-XXXX and 120 supercomputing accounts does not, as far as I know, provide anything to Steem beyond filling in the 21st block producer slot. They have monopolized the mining queue in the past using an exploit. They came back to dominate the queue again, claiming to have a private GPU miner. The supercomputing accounts' departure from the mining queue when Zcash was released corroborates that story. However, the rabbit accounts were still dominating the queue, off and on, during this period. Using the original PoW exploit, 'nobody' named their accounts "gxt-1080...", perhaps as a deception. I think using the rabbit account names could also be a deception that they have any competition for the miner queue at all.

If the miner slot is monopolized, as I believe it may be now, the proposed changes to witness rewards could give a single person nearly 5 times the SP given to the 19 witnesses who are trusted enough by Steem's stakeholders to be at the top.

I agree that the curve should be smoothed so that the backup witnesses are paid in more SP. Backups need to be able to cover server costs and be compensated for the time they put in. But I think giving the miner slot just as much is a mistake, given the situation.

Addendum: The @steemitblog proposal brings up the idea of using Zcash's Equihash mining algorithm, which would even the playing field a bit. This is because CPUs are somewhat competitive with GPUs, and there are publicly available GPU miners for it (though it's said private ones are a few times more efficient). However, Dan's proposal kicks the Equihash can down the road, while making the changes to witness and miner rewards immediate.

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

While the queue is being dominated by at most 2 people, of course it doesn't make sense. Given that they get queued just as much as a top 19 witness, I'd say logically they should be given the same pay - effectively 1/25th.

However, let's assume the move to Equihash happens, and open source Equihash miners catch up to closed ones. (which should happen given the amount of attention Zcash has received - I notice Claymore has boosted hashrates by nearly 2x in a week.)

Would someone investing a ton of computational power into mining not deserve a high reward? If someone goes out and buys 100 GPUs to dominate the mining queue, I believe there's an argument to be made for the objective contribution. But that seems some ways off and a lot of assumptions at this stage...

·

I forgot to address the Equihash proposal, I just put an addendum at the bottom of the post.

Excellent post, valuable information thanks for sharing

Concrete proof that this idea of The Devil!

See the 666? ;)

·

Oh boy, @noganoo is gonna have something to say about it! :P

·
·

Haha! Thanks for the heads-up, I'll keep a close eye on my reputation score

I'm confused man.. what is happening?

·

Good question! Much of this is still way over my head.

·

The two proposals I linked in the first paragraph put forward some proposed changes to the Steem network in its next hardfork. A hardfork is a cryptocurrency term for when the network rules are changed by majority or supermajority consensus of its block producers. The proposed changes include a major adjustment to block producer rewards and I think those changes need some critical feedback, some of which I have provided here.

·
·

Thank you for posting this. I love reading the lexicon but I also need a layman's version on simple things like hard forks.

Oh my! The strange things that went through my head, hmm @beowulfoflegend ?

·

One/two people right now are taking all of the rewards from STEEM mining. It's not because they are the only ones trying, it's that currently with how things are setup, no one else is able to compete.

This new proposed update is set to increase mining rewards by 5x.

That's what is happening, though we're all confused as to why it's happening :)

·
·

This new proposed update is set to increase mining rewards by 5x.

This is a bit misleading, and I thank @pfunk for explaining this to me. Witness rewards are proposed to be so drastically cut overall that even with the 5x bonus, miners (and runner up witnesses) will be making less than now.

The overall cuts are extreme, harmful, and unnecessary, but if they are made then there is actually an argument that the 5x bonus is not large enough. For example runner up witnesses now struggling to cover server costs will find themselves overnight even more underwater or will quit.

·
·
·

Very true - it is going to decrease overall (like everything else), but proportionally it will increase. It's all the way you spin it.

I suppose a better way of explaining it is that it will be 5x the amount an elected witness will receive.

Either way - I'm still not for it :)

·
·
·
·

5x the amount an elected witness

Why does this matter?

·
·
·
·
·

I don't see the logic in going from 1x the amount of an elected witness to 5x.

I'm baffled as to why it's being increased as opposed to remaining the same.

I suppose it "matters" because it's an illogical choice in my opinion.

·
·
·
·

I don't see the logic in going from 1x the amount of an elected witness to 5x.

I'm baffled as to why it's being increased as opposed to remaining the same.

I would guess it is an attempt to reduce the inadvertent impact of the huge reduction on overall witness rewards (clearly directly primarily at top 19) on miners and runners up (i.e. keep their rewards closer to the same as before). It misses the mark a bit -- the miner and runner up rewards still ends up with a net reduction -- but not nearly as much as without the multiplier.

How else would you propose to reduce only top 19 witness rewards (presuming that is a good idea)?

I guess the real problem... How to split up the mining so the GPU wielders don't dominate everything.

This post has been linked to from another place on Steem.

Learn more about and upvote to support linkback bot v0.5. Flag this comment if you don't want the bot to continue posting linkbacks for your posts.

Built by @ontofractal

To be honest, the one thing I do respect is that they took a hiatus. They didn't have to do that, and they did. That's nice and humane, and shows some character.

So whomever is running rabbit, supercomputing, et al... thank you for taking the temporary hitatus.

I also believe it is the same person who has cleverly outwitted how to dominate the non-voted witness list by using names to fool everyone. But I don't fault them for that. It's smart. Unfortunately, a lot of people are upset by that... and that's not the fault of the exploiter, it's a fault of the system.

But he did take a hiatus, and for that, we should be happy about that, so he considers doing it again.

·

To be honest, the one thing I do respect is that they took a hiatus. They didn't have to do that, and they did. That's nice and humane, and shows some character.

They are mining Zcash because it pays more. Nothing to do with character.

·
·

BTW, I miss seeing you read my posts.. been awhile since you checked me out. I wrote about a whale tonight in a nice way, you might be interested. Just read it, and don't vote up if you don't want to.. :)

·

The hiatus wasn't out of kindness, it was because when ZCash launched. There was a LOT of money to be made right when it launched, each coin was trading for anywhere between $1k-$20k.

It's since dropped into the sub-$100 range and is a bit less profitable now :)

·
·

Yes, I did remember seeing ZCash worth +$900USD, and it fell by the second everytime I reloaded it. So this person running a botnet sent his bots to go mine ZCash, I get it.

However, the first time it happened for a single day, I had written a comment, which I think he saw, and he did take his botnet offline for a day. Not saying I'm responsible for it, but I think he did do it for non-making money reasons for that particular day. People like this aren't 100% bad and evil. They really aren't... at least I like to think of it that way.

If we are careful and responsive enough, maybe he'll even make a steemit post, explaining why he does what he does, and give some insight to it. He doesn't have to stop.. but he could at least speak to the rest of us so we can understand more about him. I'm interested.

To be clear... if he's that heavily invested in accumulating STEEM or SBD, he also pays attention to what happens in the community, so he reads posts and comments, and will probably read this post, and my comment (and yours) too.

I want to reach out to him, to see what he has to say. I don't want to stare at his bots dominating the non-voted witness list. I am super curious on what he has to say.. because he's pulled it off well. While we all sit here wondering how he did it. He doesn't have to tell us.. but he could say "something".

·
·
·

No one's saying these people are bad or evil - I think the main thought here is that the mining system is broken and needs to be fixed. The point of mining is decentralization, and only having one or two miners is the opposite of that.

As for talking to them, one of them hangs out on steemit.chat and has explained a lot of his/her operation over the last few months. Goes by the name of nobody in the #mining channel, and I believe is the one responsible for the gxt- accounts. Seems like a nice enough person, and knows exactly the kind of monopoly they have on mining right now.

Basically from what I've read of the conversations, this person runs 2 nodes - a miner and a witness. That single mining node (likely a custom GPU rig) is outperforming the hundreds of other people participating, combined. The only other person who stands a chance when nobody is mining is the owner of the rabbit accounts.

There were also conversations about how nobody did switch to ZCash, and still is mining there, which is why you don't see gxt names in the mining queue right now.

The idea that a botnet exists is pure speculation.

No one's going after the one or two people that are dominating the queue. This is all about NOT increasing the reward into that system. It's a system that right now is viewed as flawed by many people within the steemit community.

If you really want to talk to them, just go into steemit chat :)

·
·
·
·

Thanks. I do believe the GTX- something was just a ploy for people to believe a GPU miner was out (which really isn't). When it first showed up, everyone speculated that a GPU miner exists, and that's why there can't be any possible way of competing against them. If you convince people there is no way to compete, they stop competing, which will cause people to "stop trying". Once you do that, you can dominate the place. Which is exactly what has probably happened. :) If I were that guy, I probably would have named my bots gpu-miner-1080 instead. But by naming them GTX- something gave the illusion to people who "figured it out" that it relates to a video card, and therefore must be a GPU miner. It may not be.. it might just be a lot of computing power by a botnet with lots of cpu miners (with RAM) who use the same naming scheme to scare people off, thinking a GPU miner exists in the wild (which doesn't)..

P.S. Dan & Ned might not be too concerned, because blocks are still being solved, the chain is running fine, and new account creation by pow is now down to a minimum by the dominate rabbit and gtx and supercomputing botnets... So it's less of a problem to them than to us.

I hate it when people end a post like this by the cliche "Just sayin..."

But this time, "I'm just sayin..."

P.P.S... the alias "nobody" refers to Apache, the webserver user. Which indicates he is using hacked website accounts to pull this off. So yes, it does indicate a botnet.

Just sayin again. :) But to the botnet owner, don't attack me. I have respect for what your doing, because the design allows you to do it. If we don't lock it down, someone else would. We should lock it down, and that's our ongoing issue.

I don't think the word 'monopolize' means what you think it means.
What's to stop someone, anyone, from mining? All it takes is a computer?
so...apparently they are just taking advantage of an opportunity...which ANYONE can do if they want to spend the time and money to do it.
I see no problem with that.
In fact I'm considering mining my ownself....once I get some other stuff done. I have this old XP laptop just sitting there.

·

You would be astronomically lucky to enter the miner queue with an XP-era laptop versus a person using a private GPU miner with multiple modern GPUs. When both the supercomputing and rabbit accounts were competing in the queue, it was rare to see even one account belonging to someone else in the queue.

·
·

(shrug) I dunno how it works. I have some learning to do.
I suspect that if it was easy anyone could do it.
I also suspect that if it was easy it wouldn't be as worthwhile TO do it.
I think the word is 'dominate' rather than 'monopolize.'

·
·
·

When the two groups of accounts have competed, they have duopolized the queue. If the accounts are owned by the same person, as the timing of their block production might suggest, it is a potential effective monopoly.

·
·

Hmm.. You know - We don´t really need outside miners nor the mining queue imho.
I´d say we should drop the whole mining-thing all together and just witness.

·
·
·

I've had similar thoughts, but the miner slot does provide some value such as anonymous account creation. I'm pretty sure Dan wants to keep it for diversification of block production, which is valid to me. But I don't agree that it should be paid equal to the backup slot.

Here's what the Steem whitepaper has to say about it, from page 22:

This process is designed to provide the best reliability while ensuring that everyone has the potential to participate in block production regardless of whether they are popular enough to get voted to the top. People have three options to overcome censorship by the top 19 elected witnesses: patiently wait in line with everyone else not in the top 19, purchase enough computational power to solve a proof of work faster than others, or purchase more SP to improve voting power. Generally speaking, applying censorship is a good way for elected witnesses to lose their job and therefore, it is unlikely to be a real problem on the Steem network.

·
·
·
·

But seriously, thanks for the real research and dedication to Steemit.

I'm guessing one of the reasons the miner's reward is left this high could be to motivate more people to develop a GPU miner and end the current domination. The bounty on that is rather thin and this would incentivize the development.

Btw..didn't Golos say they will develop/release a public GPU miner if the same queue domination happens there too?

Mining is a racket. No responsibility, no accountability, no commitment to the platform, and no expectations are laid on miners. They provide very little. If the majority of miners got lost tomorrow, there would be no negative impacts. Their earnings should be far less than the lowliest backup witness.