**
This Thread may be Removed Anytime!**

Twitter may remove this content at anytime! Save it as PDF for later use!

- Follow @ThreadReaderApp to mention us!
- From a Twitter thread mention us with a keyword "unroll"

`@threadreaderapp unroll`

Practice here first or read more on our help page!

State of #Avalanche: a) number of validators at 947, π b) TVL at 211.7M $AVAX, π€ c) GINI at 85.1%, d) min. number of approvers at 947 - 858 = 89, e) min. number of rejectors at 947 - 925 = 22:

..where the GINI inequality coefficient is defined as the sum of total stake difference *over* the sum of max. possible stake difference among validators. It's 0% for total equality (everybody having the same stake) & 100% for total inequality (a single one having all the stake).

..where the *approvers* control 70% of the stake and _need_ to collectively sign-off a TX. It's a minimum number, because one of the mega-validators among the *approvers* might get replaced by two or more smaller validators.

State of #Avalanche: a) number of validators at 944, π b) TVL at 213.0M $AVAX, c) GINI at 84.8%:

1/ My long terms followers will have noticed, that I've added a 2nd vertical bar to the graph. The 1st *left-hand-side* vertical bar is the 30%-vs-70% split, and the 2nd *right-hand-side* vertical bar is the 70%-vs-30% split w.r.t. to stakes.

2/ Why did I do that? Well, the LHS 30%-vs-70% split tells us that 853 out of 944 validators control 30% of the stakes, and the remaining 944 - 853 = 91 validators control 70% of the stakes.

State of #Avalanche: a) number of validators at 965, π₯³ b) TVL at 220.5M $AVAX, π€ c) GINI at 84.6%:

1/ GINI is a measure for *inequality*: A min. value of 0% would be very good for decentralization, and would imply perfect equality among validator stakes. A max. value of 100% would be very bad for decentralization, and would imply perfect inequality among validator stakes.

2/ So, a GINI of 0% would mean all validators have the same stake, and a GINI of 100% would mean a single validator has all the stake.

1/ I've been looking to poke holes into #Avalanche for *two* years, writing an entire simulator for 1 million+ nodes in the process. Anything you throw at it, has either been easily solved or is easily solveable.. it's just unbreakeable:

2/ Protective ephemeral centralization by the #Avalanche foundation? That's easy to fix: distribute $AVAX via sales, tax larger validator rewards, subsidize smaller validators or modify staking to voting relationship w/o affecting safety too much.

3/ Liveness suffering due to theoretical fat-tails distributions like Pareto or Cauchy? Easy to fix: apply adaptive staking vote shaping to measure & recognize such distributions in real time. Applt counter measure by dynamic staking power adaptation.

1/ Why #Avalanche is even better than I initially thought (part 2)?

2/ In our previous thread we discussed how *fast* the distributed #Avalanche *consensus* mechanism can sync all honest nodes:

3/ Above you see how after *only* 3 rounds the entire set of honest participants are in sync: Despite 15% being faulty (or malicious), the system manages to achieve the max. possible consensus level of 85% (for the overall network).

State of #Avalanche: a) number of validators at 982, π b) TVL at 258.7M $AVAX, π€ c) GINI at 85.6%: