RE: Blurt Economics: The Convergent-Linear Reward Curve Made Equitable

You are viewing a single comment's thread from:

Blurt Economics: The Convergent-Linear Reward Curve Made Equitable

in blurtopian •  4 years ago 

[EN] Thank you for the detailed explanation and definition. The proportional curve (c in the ratio one to one with r) is the fairest variant and should be our goal.

Every person may have only 1 account as @double-u suggests in his Wishes for BLURT, makes the unpopular convergent curve obsolete. For this we would need a way to mark accounts as "human", like BrightID does.

I suggest to go in this direction and announce this publicly. This way people who have or want to create subaccounts for abusive purposes will know right away that they have no future here on Blurt.


[DE] Danke für die ausführliche Erklärung und Begriffsbestimmung. Die proportionale Kurve (c im Verhältnis eins zu eins mit r) ist die fairste Variante und sollte unser Ziel sein.

Jeder Mensch darf nur 1 Account haben so wie @double-u es in seinen Wishes for BLURT vorschlägt, macht die unbeliebte konvergente Kurve überflüssig. Dazu bräuchten wir eine Möglichkeit, Accounts als "menschlich" markieren zu können, so wie es z.B. BrightID macht.

Ich schlage vor, in diese Richtung zu gehen und dies auch öffentlich anzukündigen. So wissen Leute die Unteraccounts für missbräuchliche Zwecke haben oder erstellen möchten gleich Bescheid, dass sie hier auf Blurt keine Zukunft haben.

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 BLURT!
Sort Order:  
  ·  4 years ago  ·  

There are legit reasons for a person to have multiple accounts.

I would focus on algos that define what a bad actor does rather than assuming everybody is potentially one and hence needs to be restricted.

BTW set A=B in the formula and it immediately becomes proportional :-)

Ok, we make A=B :-) (A=4 and B=5 is also a good direction)

Can you approximately say, where the point is where the current curve converges to being linear? Measured by the vote value in Blurt. (At Steem it was about 20 STEEM, if I remember correctly.)

  ·  4 years ago  ·  

Interesting how certain numbers just stick in the memory. That "20 STEEM" was another simple fabrication by the Steem devs to placate the masses and hide the truth.

I tracked down an exchange I had at the time here so that I can now replicate their fantasy. The curve is smooth, there are no cut-off points. That value of 20 comes from estimating c=0.7r, which I note with a smile is already very low. Had the original calculation been honest, it would have chosen something like c=0.9r, yielding a much higher value than 20 STEEM.

Anyway, if I plug that into the Blurt economy, I get a very large number that nobody is going to get! Honestly, it's huge, and that's because the rshares Blurt generates are smaller than Steem but it is using the same equations. On the plus side, everyone on Blurt is in the same boat! In terms of disparity between whales and redfish, the current implementation is being fair to all - apart from those posts below 50 million rshares.

So, at the moment on Blurt, everyone from whales downwards are actually experiencing a reward curve with a gradient lower than 1 - between 0.5 and 0.7 - and my 3rd graph above is actually the best representation of what is going on at the moment. Fixing the dust-threshold and the reward curve will change this dramatically. Given the calculations your question has prompted, my recommendation within the post feels even stronger :-)

In conclusion, I would advise to set the reward formula with A=4 and B=5 and the content_constant to 2000000000.

  ·  4 years ago  ·  

@michelangelo3 there are indeed legit reasons for having multiple accounts @rycharde pointed out, such as keeping assets safely spread across accounts, cold vs hot wallet, if you have a business or project on Blurt it might need its own account etc, I would like to see some form of proof of unique humanity and only those accounts are to receive rewards, but realistically such a solution is not in the scope of 2020, HF 0.3 will hopefully be in November with the other improvements and something like this would be more for 2021.

Yes, of course there are reasons for multiple accounts, I don't want to exclude that. I have two more accounts myself, one for testing purposes and one that I want to use as a recovery account.

I would like to see some form of proof of unique humanity

Preferably without having to prove his identity by document. That's why I invested some time to get verified by BrightID. The idea behind: Through a partnership, users from BrightID find to Blurt and in return we would also bring BrightID new users. Win-win.

Ah - new users, at the end of BrightID Applications is a graph with the number of users, rising very nicely for such a new project.

If we are interested in a partnership, maybe 2021, I can ask Adam or Alireza from the BrightID team if they are interested too. These two are very enthusiastic, reminds me of Blurt ;-)

  ·  4 years ago  ·  

Sir here is my current blurt account, Mashooq Ur Rehman