Final Review of Steem Economic Changes
Today we are summarizing the changes we have implemented and are prepared to release on Tuesday and hardfork the following Tuesday. If there are any issues discovered this could be postponed. Change is always controversial, but we feel the changes we have implemented reflect the wishes of the wider community.
Changes:
- Power Down in 13 weeks instead of 104
- Inflate at 9.5% APR narrowing to 0.95% APR by 0.01% every 250,000 blocks (Roughly 0.5% per year).
- Steem Dollar conversion takes 3.5 days (down from 7)
- Witnesses receive 10% of inflation, runner up witnesses get paid 5 times the top 19 per block produced.
- Miners get paid the same as the top 19 per block produced
- Witnesses / Miners will continue to be paid in Steem Power
- 75% of inflation goes to authors and curators
- 15% of inflation is allocated to Steem Power
- Switch to Equihash Proof of Work algorithm.
- Remove rewards for including Proof of Work transactions
Equihash Proof of Work
By changing to Equihash we hope to make ordinary computers more competitive relative to high end GPUs with optimized algorithms. Over the past couple of months one or two people have dominated the Steem mining queue. They have either identified a computational shortcut or implemented a private GPU algorithm. The Equihash algorithm was adopted by the creators of Zcash, a recently hyped privacy-oriented crypto-currency. This algorithm is designed to make heavy use of randomly accessed memory. The computations are very simple, but they require 128 MB of ram accessed in a manner that should saturate the memory bus.
The nature of this Proof of Work algorithm is that solving a single instance can take over 2 blocks (validating is instant). We have updated the Proof of Work to allow them to build on top of any block after the last irreversible block. In practice this means miners have about 30 seconds to solve the Proof of Work.
Steem Dollar Conversion
The moving average of STEEM / DOLLAR price has changed to 3.5 days to match the conversion delay. This should dramatically reduce the price volatility risk of converting Steem Dollars to STEEM which in turn should strengthen the peg. 3.5 day median should still give 1.75 days to detect and correct any bogus price feeds before they can have a negative impact on the protocol.
Paying Witnesses in Steem Power
Witnesses have always been paid in Steem Power. In our last post on proposed changes, we suggested paying them in STEEM. After talking with prominent witnesses, we decided to keep it in Steem Power. The 3 month power down significantly improves witness liquidity and paying as Steem Power will prevent the need to manually power-up after every block as we believe most Witnesses will/should choose to do.
Paying Runner Up Witnesses (Time Shared Witness Slot)
We have increased the amount paid per block to the runner up witnesses. These witnesses produce blocks a couple of times per hour instead of every single minute. Even though they get paid 5 times the other witnesses for their block, on average they will still earn less per day than the top 19. By increasing the pay-per-block we enable more people to be runner up witnesses which helps keep a healthy supply of standby witnesses.
Miner Pay
Miners will make the same per block as the top 19 witnesses. After watching the miner queue be dominated by one or two individuals, some people were rightly concerned that a mining monopoly could consume a disproportionate amount of the block rewards. By paying the miners (collectively) the same as the top 19 witnesses, we are assured that even if mining is fully dominated by a single individual that said individual would have no more power or income than the top 19. After factoring in the cost of mining, their profits would be less than the elected witnesses.
Modeling Difference in Steem Supply
The above chart shows the difference between the old and new model by comparing the current Liquid Steem vs the new Total Steem ( Liquid and Vesting). What this shows us is that within 18 months there will be a net reduction in supply. This assumes the worst-case that everyone but Steemit powers down at the maximum rate.
We would like to caution speculators that these charts do not tell the whole story when it comes to projecting future price. Supply is only one side of the equation and there is no easy way to measure how the market currently perceives Steem held as Steem Power relative to the liquid STEEM. The other side of the equation is demand. Small changes in demand can have dramatic impact on price. There are too many variables to make any projection on where the price will move in the short term.
Witness Pay
One of the great services that our witnesses have been providing our community is the funding of development from their witness pay. The Witness role was not intended for this at genesis, however, it seems the delegated project funding has been very valuable, and these witnesses should be appreciated for their hard work and dedication. While the budget for witness pay, less the POW inclusion reward (.55% inflation currently), is increasing from 0.75% to 0.95% inflation. Effectively, witness pay decreases in the short run and increases in the long run.
Long View
These changes are intended to open doors for Steem as a currency and provide greater synergy between the currency, social and content aspects of the network. We look forward to comments. We will consider comments before setting a date for offering the potential upgrade to the network’s witnesses.
I would like to remind everyone that Ned and I will not be powering down during the first 3 months after the hard fork.
I haven't powered down yet, and will refrain to do so until STEEM succeeds.
What's your definition of success for this purpose?
If STEEM is used to solve some real world problem, in assertive way. Say if STEEM becomes de facto solution for problem Y in market X. Either, or both problem Y and market X need to be significant.
Example:
Someone builds a next gen marketplace on STEEM, and due to its "10x" benefits it becomes the new eBay, reaching out to millions of people (market X). Now the chain holds all of this massive intrinsic value.
Plus imagine what you can do with your votes if that time comes. Really make a change with giving value where you think is deserved!
Me either Dan, me either...
What % of inflation goes to authors and curators now?
It is comparatively tiny, around 6.5%. However, the economic models are so different that is not really comparable.
This is gonna be nuts! :D
@smooth wrote:
Afaics, the system is almost identical for authors and curators but very different for SP and non-SP holders...
Per my prior blog (c.f. also @arhag's comments) about the preexisting inflation math, existing SP holders were not debased (i.e. not diluted, effectively a stock split for them) when the ratio of SP to the total money supply including non-SP was ~87%. Above that ratio, SP holders were being debased and below that ratio they were experiencing a positive interest rate. For this new proposal, the stock split ratio can be approximated roughly as (which doesn't account for keeping the ratio constant or the fact that existing SP holders are diluted by new SP holders, but this is a small difference in most scenarios):
x × 0.15 × 0.095 = (1-x) × 0.095Which is again ~87% but that is 87% of ratio of non-SP to the total money supply, so ~13% if comparing the prior system. The other difference is that above and below that ~13% ratio, the effects are transposed, i.e. above is a positive interest rate for SP holders and below they are being debased but never more than 9.5%.
So if comparing the prior system to the proposed one, at the equivalent stock split ratio scenario, then SP holders are debased the same in both proposals and the author+curator rewards are also roughly the same at
0.75 × 9.5% = 7.1%. But it isn't likely that SP holders will only be ~13% of the money supply, although power down has been reduced to 13 weeks from 104. Thus the new proposal is likely to be much more dilutive. At the 95% ratio (what it was historically), SP holders are debased at roughly 8.8%. It is unlikely for the ratio to drop much unless the whales power down, but powering down wouldn't collapse the price if they aren't selling. So the real effect of this change is to debase the SP holders, so there really isn't any reason at all to power up. So we can expect everyone who can power down to do so, until the ratio reaches some level where the debasement rate on SP holders is much less than for non-SP holders. Although there might be a Prisoner's dilemma which every whale wants their SP to be powered up if the debasement is less. So the homeostasis is likely to be some where at a positive level of debasement for SP holders, but less than that of non-SP holders. Thus free market and kudos on a good design decision (actually is what I had planned to do, except I would not have dropped the 104 weeks to 13 weeks because it could create enormous selling pressure collapsing the price).The huge difference is that non-SP holders are only debased at 9.5% instead of in excess of 100% in the prior system.
The other huge difference longer-term is the debasement 9.5% APR decreases by 0.5% per year.
Agree but that wasn't the question that was asked. The question was asking about something relatively unimportant, to me, which was why I didn't actually do the math to answer it (nor did you) and just gave a rough guesstimate, but apparently not to the person asking it.
I think I will avoid this too. Solidarity and all that!
There is a new feature, you can now add a profile picture :-)
I'am sure that you have enough on both exchanges Polo and Trex ;). How low you can go. AGS greetings.
Thank you sir, good luck propagating platform adoption (upgrades / promotion) without too much need for heavy cashout sessions in the future.
Dantheman could I ask you where I can find what the reasoning is behind changing the 2 years to 13 weeks?
I don't see how Steemit can succeed if we are just hoping on the fact that whales/ dolphins are not powering down than the system can handle in 13 weeks.
We seen in the past that a relatively small group of whales powering down could plunge steems value, so I don't understand why the choice was made for 13 weeks?
Why don't we change the rate based on the health of the economy for instance.
At least until Steemit is fully grown. Now with the 13 weeks an unstable timebomb is created, that places all hope in the whales/ dolphins making sensible deicisions on powering down whilst right now nobody knows what the current system can handle in terms of steem flooding in.
There could be many better options than the 13 weeks and I haven't heard many solid arguments that seem to backup the 13 weeks.
I want to invest heavily in steem but this is a major breakpoint for me.
If anyone could point me in the right direction and put my nerves to ease I would be grateful!
Good Afternoon Everyone!
Charlie Shrem, currently #2 witness spot.
We're really excited about these new changes and will install and upgrade as soon as its released.
I'm really proud of the developers at Steemit, Inc and my fellow witnesses for the open minds and keeping an open dialog with each other and the community.
Full STEEM ahead!
-Charlie
Thanks for your hard work witness #2. Much appreciated. I'm going to have to fire up my clean machine and give you a vote. :-)
Overall the Steemit team did an excellent job listening to input and preparing this revised proposal. Despite the fact that it doesn't specifically address my concern of abrupt and unnecessary cuts to both witness and content rewards, nor my concern about changing too much at one time, I recognize that there are several compromises woven into the proposal and in its totality there is lot to like about it. I expect to be supporting the update.
I agree with @blocktrades in favoring a significant period for publicizing the changes before they go into effect, which should include an active effort by both the team and community to reach out to cryptocurrency news sites, bloggers, podcasts, etc. to make them aware of the changes and what they mean for Steem going forward. While there can indeed be a greater synergy between the currency, social, and content aspects of the network, that synergy can't be realized if no one knows about it.
Thanks. I'm 100% in favour of the changes. I think a serious platform needs to evolve in line with market and community needs.
That is why I have spent thousands of my own dollars in powering up (a lot of money for me) and why I will continue to support the platform.
I know that some people have suggested eliminating mining altogether but I feel that would be a mistake both perceptually and also in increasing centralisation.
The solution to one or two people dominating mining is not to remove it but to implement measures which make it more difficult for this to happen. Equihash should make it easier for regular people to compete in mining.
I would suggest that if Equihash becomes similarly compromised/dominated in the future (quite likely) we should look at changing again and keep evolving the mining to keep it fair and decentralised.
I would also suggest we need a simple mining solution to allow non technical people to get mining on Windows and OS X.
A 1-click miner for Windows/OS-X would be fantastic PR for Steemit and also allow new users to get involved and support the network in a different kind of way.
I would be happy to even contribute some of my own limited bitcoins/SD towards development of this. Sadly I don't have the technical skills to develop this myself but I know we have many people in the community who can.
We will no longer support windows due to limits on 3rd party libraries in use.
There must be some way around it though? I think this is an important issue as the vast majority of people will never use linux. It would be a mistake to exclude 90% plus of the users from mining. It simply does not look good from a marketing/PR perspective and gives more ammunition to the detractors.
Using docker or another virtual machine is an available workaround.
Someone could create a mining pool with a standalone miner.
Interesting - like a Minergate type setup?
That would be ideal.
I have a number of friends who were very non-technical and wouldn't even dare using regular miners (even on Windows) but they use their spare CPU/GPU time to mine using Minergate.
It makes it really easy for new users so something like that would be perfect.
Obviously the issue of centralisation comes up if there is a single pool doing it though.
Agreed. Something like Minergate would be great. It's so easy to get started mining on Windows with their software.
I was just thinking @Minergate should do this. Would be great.
You have got to be kidding.
Hello please tell me the huskies have a limit? how much can you put in a day? and if not difficult to you tell me how to climb with the help of likes ?
I agree, we must support Windows users as part of the mining activity. This will convey a mainstream (mainsteem) appeal and include more people into the overall front and backend of Steem/Steemit. Otherwise, the perception is this is a tech experiment on the fringe. The 'beta' moniker also reinforces this.
Yes perception and image is just as important as the substance behind the platform. We must not give the appearance of being exclusionary or elitist.
That is the exact appearance this change gives, because that is exactly what it is.
This change, (in addition to breaking every vesting\interest promise people invested for) removes any attempt to include windows miners. This makes it a purely developer and in-crowd project, mainly because the damn miner builds have never had:
This change just ensures the linux developer in-crowd control the mining, just like every other change they make around here.
Another nail in the coffin. There are not room for many more.
Cry more, dirty windows users
Seems the steemwhales creator sold his account??
you should ask for a refund!
I've been mining Zcash on Windows for a few weeks now, easy to set up. Equihash is awesome. Great move switching to that algorithm.
Witness Riverhead will be participating in this fork.
After the changes if you update the white paper, I will put it to my reading list.
Excellent! Witness xeldal fully supports the above changes and will be upgrading when released. Great work on listening to the valuable feedback from the community. I'm very please with the direction we're moving with this.
Hello, I imagine that you'll never get this message. Nonetheless, I'd like to make a business proposition:
I've noticed that you're not really doing anything with much of your Steem Power right now. Would you care to delegate some SP to me for a fee?
I'm building a new voting application and I hope to fund it with something on the order of 50,000 SP, for which I'd gladly pay 100 STEEM per week.
My application will not be used for self-voting; it will only vote on others' comments.
If you get this, let me know what you think.
To be honest I'm totally lost on all this. I am new to this world and have much to learn. As long as it proves to be beneficial to those posting on Steemit I'm all for it.
I will power down during 3 months when this is implemented.
I'm probably going to power down 33% of my SP the first month, not sure if I am going to sell any though. 1.425% APR on STEEM Power is a lot better than any bank CD / IRA.
If you don't mind expanding a bit, I'm curious why you will do this.
I'm not powering down squat. In fact, I just invested $700 into Steem Power from my Bitcoin wallet about 2 hours ago. The price of Bitcoin dropped $10 just 5 minutes after the transaction cleared too. Talk about perfect timing. Lol. And now Bitcoin has dropped ANOTHER $10 in just the past hour. Wow. :)
I saw that drop, and was exchanging during that, so good for you!
Thanks! Yeah, it was shocking to see BTC drop $10 just minutes after my final deposit cleared, and then keep dropping. It's still down at about $738 right now.
Awesome! Thank you! And good news for the 3 month power down in effect when this is implemented. Good job developers!