Blurt Hard Fork 6 Revisited

Human memory and perceptions are fragile things that need constant reminders. I have written many posts about the Blurt economy - I suggest people read those, going back 8 months. Some posts were qualitative while others were quantitative; I make no apologies for the details as a blockchain is one huge construct of logic and mathematics. Words are mere approximations that need to be refined into precise algorithms that can then be encoded in some machine language.

Thanks to @Megadrive for posting while I was still writing this. I have edited out some redundancy but is sometimes good to read the same thing in different words.

It has been unfortunate that the one functioning platform, Blurtter, was showing the wrong pending payouts for a very long time. The Blurt Wallet has always been the reality of the Blurt chain and its rewards. The "pending rewards error" has now been fixed by correcting the way the API sends out data. If people hadn't noticed that their wallet rewards were different to the values on Blurtter, then that is a matter of perception; many have asked about this and accepted the front-end error. The Ecosynthesizer website has actually been showing the correct values for some weeks.

But what is of concern is how many users just have not perceived the slow and steady INCREASE in vote yields between HF5 and HF6. I refer to here, dated April 1st, you will see vote-yields of 58% for large accounts. By May 16th, that yield had increased to over 71% APR. Your own vote-yield will depend on both your BP and VP, but the trend remains the same.

Smaller accounts saw a slightly different pattern in that we removed what was the vote-dust-threshold so that their votes increased even more in terms of their effect - many went from zero to a real positive reward, albeit modest, so a meaningless infinity in percentage terms.

Had nobody noticed any of this? A sharp drop is easy to see, but a slow steady rise back up has been happening for some 2 months. This is even ignoring the "doubled" values on Blurtter.

Now, HF4 was indeed, mis-coded and we had some days with no rewards; this was then fixed and named HF5. This means that the Blurt economy really jumped from HF3 to HF5. These things happen, they happened on Steem at HF20 when most people did nothing as their RC recharged - one week lost.

I wrote many times about what to expect at HF5. The initial effect was a drop in rewards followed by an increase in the reward pool that would then increase those rewards again. This is exactly what happened - rewards INCREASED between HF5 and HF6.

Sadly, I did not tabulate Blurt vote-yields going back months, but the formulas are easy to reproduce. The average vote-yield between HF3 and HF4 was about 60-70% APR; for many, this was far less due to the vote-dust-threshold. Firstly, how sustainable is a 65% APR? Secondly, note that the 71% in May is almost the same, indeed higher. My last report showed an estimate of 69% APR. Not really much of a change in the last 2 or 3 weeks, but a small drop due to more activity.

That value will change in response to activity - to posting and voting - and as activity increases, so the yield decreases. This is how the reward pool reacts to recent claims - always has done on both Steem and Hive. On Blurt, as staking increases this will nudge the yield slightly upwards. Indeed, if there is ever a time when, say, 80% of BLURT is powered up and voting, the yield will move towards the blockchain minting rate that is assigned to the reward pool: approximately 5% APR at today's rate. This has never happened on Steem and is unlikely to ever happen, but it is a useful floor to keep in mind.

So, what was HF6 about? Well, the reward curve has gone back to almost what it was before HF4. So why have vote values dropped now!? Well, the reward pool and recent claims have to adjust back to the new economics. This is what they do. They take 15 days to clear out the old posts and replace them with data from new posts. Both are algorithmic and hence predictable. We shall then be back to what many people wanted - or think they wanted. This is exactly what happened after HF5, albeit for different reasons.

The whole point of a proportional reward system is that everybody gets the same yield. You will notice today that this is the case. I refer you to this post that describes the curve in detail. Did whales really not notice that their votes were higher than average since HF5? The price for going back to pre-HF4 is that the rewards need time to adjust upwards from now so that they all realign. Everyone now has a fair share, but the system is still using the 15-days forward data. It thus takes 15 more days for that data to reflect the new equitable rewards.

We already see the reward pool has grown from about 2.02 to 2.07 million BLURT in one day. It will now continue to grow and hence distribute more rewards per vote until it reaches equilibrium again. If you read my weekly posts on Blurt Economic Indicators you can keep up to date with the macro changes to the economy.

One last historical detail may illuminate the dynamic self-adjustment on the blockchain. My work with Blurt started back in October 2020, a few months after the launch. One of the first things I analysed was a mysterious sudden drop in vote values back then. Read the post for the whole story, but the point is that the macro parameters of the chain are reactive. In this particular case, an attenuating function was put in place that lasted some weeks. But those reactions normally take days to fully manifest - at least 2 weeks and sometimes 2 weeks more as humans react to the changing values.

A final thought. This is like trying to change a tyre... while still driving.

References to the Paying Attention Economy

My, My, Mine... No... Ours!

Why Have My Pending Rewards Changed? The Economic Effects of HF6

Blurt Economic Indicators - 30 May 2021

Blurt Economic Indicators - 23 May 2021

Blurt Economic Indicators - 16 May 2021

Blurt Economic Indicators - 12 May 2021

Blurt Economic Indicators - 2 May 2021

What's really holding back crypto social media?

Blurt Economic Indicators - 25 April 2021

BLURT Trading Note - 23 April 2021

Blurt Economic Indicators - 18 April 2021

Blurt Economic Indicators - 11 April 2021

Foundation Backing of Curation Accounts Has Ceased - Time to Build Blurt Better

Helping Blurt Curators and Communities in the Future

Blurt Economic Indicators - 6 April 2021

Blurt Economic Indicators - 2 April 2021

Blurt Better - Do Not Vote a Post Before 5 Minutes

What is the Current Vote-Yield on Blurt?

Upvotes and Downvotes Are NOT Equal and Opposite

Moloch and Blockchain Economics

Blurt - the Feast After the Famine

Brief Summary of Blurt Economic Indicators - 24 Mar 2021

Why are my reward payouts about half?

HF5 WORKS!

An unposted post

A Tale of Activity During Hard Times

Error in Blurt HF4 Payouts to be Fixed ASAP

Changes to economic formulas used by Blurt front-ends

How Blurt's HF3 Will Affect Vote Values and the Economy

Measuring the Success of Cryptos: Price or Principle

Blurt Tip - Vote! Vote! Vote!

Dependencies are more important than decentralisation

Graphs of Proposed Solutions to the "Investor v Blogger" Dilemma

Towards Resolving the "Investor v Blogger" Dilemma on Blurt

Making the Reward Fund Distribution Sensitive to the Vested Blurt Economy

Blurt, Ethereum and Proof of Stake

My Fantasy Curation System

DRAFT Blurt Curators Guidelines v.0.1

Abuse mitigation plans for Blurt: A discussion

Metcalfe's Law is Wrong - This Has Consequences!

Why Cryptoeconomics is Important

How Can We Best Design the Curation Rewards on Blurt?

Why Are Curation Rewards on Blurt Almost Always Half the Vote?

Why Have My BLURT Rewards Fallen Recently?

Blurt Brainstorm: The Economic Landscape, A Personal Perspective

The Economic Changes in Hardfork 0.3.0 and Their Rationale

Blurt Economics: The Convergent-Linear Reward Curve Made Equitable

Blurt Proposal: to Change the Vote Dust Threshold to Zero