Weiss Ratings Decries EOS Centralization: A Case of Bad Breadth

Image

While the EOS community continues to work on finalizing its MainNet release, we have some serious questions about its centralization that remain unanswered.

EOS is a still great project. It continues to merits a good Weiss Rating overall. And it still has the potential to do what Ethereum promised — take the crypto world to the next level, and establish a foundation for the smart economies of the future.

Indeed, EOS is the first fully scalable, complete, third-generation, distributed-ledger platform ever to be released to the public. I know that’s a mouthful. But that’s exactly what it is.

That’s why we have been giving it a good rating.

That’s why so many eyeballs were watching its MainNet launch this week.

And that’s why we are also paying very close attention.

But there’s a stench in the room — a case of bad breadth:

The EOS dream will not come true until it fixes its centralization problems, which came into sharper focus with its launch this week.

This project is not new to controversy. It’s been criticized because of some of the people involved in its development. It’s been bashed for its scheme to raise $4 billion without even releasing the product.


Weiss Ratings does not usually get bogged down in those debates. Why? Because the computer programs behind cryptocurrencies are open source. They are open to the public to dissect. And they are designed to eventually be run by a decentralized community of peers.

So we don’t fret much about who does what in the early stages. At the end of the day, it’s the design — the protocol — that will make or break the project, regardless of who commands it in the initial going.

The classic example: To this day, no one knows who Satoshi Nakamoto, the inventor of Bitcoin, really was. Nor does it matter.

This is especially true now that EOS is moving into the hands of its community. The folks and the company who sponsored it, Block.one, have released it. They are not much involved in what happens next. And for the small amount of involvement that still lingers, we hope they can find their way to the exit doors ASAP.

In a moment, I will tell you more about the big issue that remains to be solved. But before I do, I want to reiterate this single point: The destiny of this project is now mostly up to its community.

That includes the teams across the world who are now running as candidates to become block producers (like miners) on the main EOS chain. They will now have final say on how this network is launched, and how its governance model works. It is up to them to make whatever changes are necessary to ensure the long-term success of this project.

And that long-term success could depend on one pivotal factor:

Centralization or Decentralization?

A decentralized project spreads influence and control over a very broad group of participants. It has good breadth.

A centralized project does the opposite, concentrating the power among a small elite. That’s bad breadth.

What’s ironic is that Block.one’s motto has always been “Decentralize Everything.” And indeed, EOS founder Dan Larimer is often quoted as saying that his main goal in life is to create “free market systems to secure life, liberty and property.”

Unfortunately, however, Block.one’s motto and Larimer’s goal are in sharp contrast to the reality of what EOS is today.

In fact, in a just-released report, we identified EOS as a platform with one of the most centralized distributed ledgers in the world today.

The Gini coefficient, used by economists to measure wealth distribution of countries, when applied to EOS, comes up at 97 (on a scale of 0 to 100).  Even if you recognize that crypto communities and countries are two different animals, the fact remains that any Gini coefficient above 60 is problematic.

EOS’ coefficient, at 97, is off the charts. Like a feudal kingdom with indentured servants.

Based on the way things are set up today …

• The top 10 EOS token holders will effectively run and operate this network …

• They will elect a group of 21 block producers (“miners”). And …

• Via the block producers, the top 10 will assert their dominance of the entire ecosystem.

Granted, wealth distribution in a free market is never perfectly even. Some wealth will always concentrate at the top, and I have no issue with that. But let’s dig deeper …

First off, Block.one organized an extended, aggressive, year-long Initial Coin Offering to raise $4 billion. And they did it all with the rationale of ensuring a “smooth and fair distribution” of tokens. Founder Dan Larimer said so himself many times.

In fact, whenever people objected to the big-money fundraise, the buzz was always along the lines of: “Hey, the reason they’re doing this is to make sure ownership is distributed as broadly as possible.”

Did this turn out to be the case? Apparently not, or at least not yet.

So why go through the hassle of a staging a year-long ICO? It seems to have served little or no purpose other than giving Block.one the chance to scoop up billions of dollars.

Second, EOS uses a “Delegated Proof-of-Stake” consensus algorithm. In other words, token holders get to vote on their block producers. This is a revolutionary concept that other networks are also incorporating. For example, Cardano, NEO and the soon-to-be-released Hedera Hashgraph use variants of this concept.

Trouble is, EOS lets each token holder vote for up to 30 different candidates — all with the same tokens.

So if I own one million EOS, does that mean I can vote for 30 different block producers with the full weight of my one million? Yes.

And since there are only 21 block producers, if my buddies and I have a majority of the tokens, could I effectively run the whole EOS network? Yes again.

This is what I mean by a case of bad breadth. The wealth distribution on this network is narrow enough as it is. Allowing each token to be voted for up to 30 different candidates makes it even narrower. It opens the Pandora’s box to a possible situation in which a small elite can call the shots on the EOS blockchain.

And this is supposed to be decentralized?

No. This community has got to come together and ensure the decentralization needed to let EOS’ true potential shine through. They need to …

1 — Get rid of the 30-votes-per-token scheme. It makes no sense in a network that aims to have a decentralized, on-chain governance model. Allowing large token holders to vote more times than there are block producers is an artificial instrument that throws the entire governance process into question.

2 — Allow for more than 21 block producers. Just 21 may make sense if each voter can actually impact the election process. But as things stand today, the overwhelming majority of token holders have little or no incentive to participate. They simply don’t count.

3 — Cap the voting power of large token holders to no more than 2.5% of the total token supply. In other words, if you’re rich with tokens, you have more power. That’s OK. But that’s where it ends. Being super-rich gets you no additional advantage.

4 — Large token holders need to identify themselves. This may sound counter-intuitive to some in the crypto community, but the fact that the EOS network is controlled by a few anonymous individuals is counterproductive. How many Fortune 500 business leaders are going to trust their future on an infrastructure run by unknowns?

Remember, the reason centralized corporations like Amazon or Google can grow so big is mostly because consumers and businesses know exactly who and what they’re dealing with. Although the crypto space is not the same, this is essentially the level of trust that’s needed to create massive growth.

Don’t get me wrong. We love the EOS project. As Ethereum and Bitcoin run into speed and governance bottlenecks, we think EOS has a solid chance of advancing public distributed ledger technology to new levels.

Now it’s up to the EOS community to do whatever it takes to ensure its long-term success. If that means diverging from the original design specs of the founders, so be it.

If you invest in EOS, you should be watching them like a hawk. That’s what we’re doing. And, if the bad breadth persists, downgrades are likely.

Best,

Juan

P.S. EOS currently has an overall Weiss Rating of “B-.” Get the latest ratings of all 93 cryptocurrencies that we currently grade, sent straight to your inbox each week. Click here to see how.

Leave a Reply

Your email address will not be published. Required fields are marked *

Comments 4

W June 9, 2018

But these top holding adresses are exchanges. I thought you would be smart enough to figure that out?

Reply

Vitalik Buterin June 9, 2018

This is a good review! I would never buy EOS.

Reply

Gloss Grey June 11, 2018

By putting their tokens on Ethereum, rather than running their own chain and exchange tokens there, EOS successfully collected money for itself and simply becomes non-responsible for anything of the chain…

This is all about money, not techniques.

Reply

firestorm June 11, 2018

1 — Get rid of the 30-votes-per-token scheme. –> 30 votes per user is actually whale deterrent. It means that unless the big whale or cartel has more than 50% they could end up with nothing. You should try some simulations but Steemit has worked well with this structure for a while.

2 — Allow for more than 21 block producers. – there are more than 21 – there will be at least double that who will count as standbys. Block Producers Candidates have been highly transparent and independent for the most part.

3 — Cap the voting power of large token holders to no more than 2.5% of the total token supply. – bigger holders could easily split their wallets, in fact some of the biggest exchanges (who actually are the most of the biggest accounts) have split wallets. Exchanges cannot vote without getting a very specific set of proxy instructions from the token holders. An exchange may have thousands of accounts but these are all grouped into one for your gini stats rendering them incorrect. In demographic terms it is the same as saying that I’m going to count all the residents of that city as one person. Over 25% of tokens are on exchanges

4 — Large token holders need to identify themselves. Some are known to the community and many are exchanges.

Reply