Leaked Huobi Document Reveals Centralized Domination of $EOS: Governance is Currently Broken
In another gem of crypto news that hit the space recently, there was an alleged report that stated that the leak of an ‘internal excel document from Huobi’ was circulating.
Obviously, the allegations of; “collusion, mutual voting, and pay-offs that occur amongst the Chinese BP community” is something that any investor in $EOS and frankly, in general, should be taking seriously. Especially when considering the fact that $EOS is currently the 5th-ranked cryptocurrency in the blockchain sphere at the time of writing.
What are BPs?
If you aren’t familiar with $EOS, then you probably won’t know the meaning of the acronyms that the user, ‘MapleLeafCap’ was using in their original tweet.
A ‘BP’ = Block Producer
What is the Significance of a Block Producer?
This is a great question. In order to answer it, we need to head over to the $EOS whitepaper:
While the information contained in the excerpt above does not cover everything there is to know about Block Producers, it should give enough information to give you as the reader a gist of what we’re talking about here.
If you read the above excerpt and came to the conclusion that these individuals are more than likely the most essential facets of the entire $EOS protocol, then you’d be correct.
Therefore, evidence of collusion/nefarious activity among them would essentially be akin to confirming that the entire blockchain itself is a fraud.
Now, there may be some people that are reading this that have felt that way about $EOS since its inception that are screaming ‘DUH!’ at their monitors — and we understand (really, we do). But in lieu of concrete evidence (up until now) to suggest that this was the case, these were merely unsupported presuppositions that we, along with others, carried and harbored toward $EOS.
What is Huobi Exchange?
Oddly enough, we haven’t had the opportunity to really talk about Huobi Exchange yet in this article, but they are also an integral part of this entire ‘reveal’.
For those that do not know, Huobi Pro (their formal, ‘business’ name), is a massive (in terms of volume) cryptocurrency exchange, that is currently ranked third among all exchanges indexed by CoinMarketCap.
They are based in Asia and have been publicly accused of engaging in very high levels of wash trading by researchers/analysts of crypto exchange activity.
In addition, the Japanese government’s financial agency (FSA) had sent the exchange a letter directly when it was operating within the jurisdiction to question its practices and inform them that some of the things that they were doing were in direct violation of Japanese financial regulations.
Thus, given Huobi’s relatively terrible track record (especially in the year 2018), there is already enough room for some individuals to presuppose the truth of the Twitter user’s revelations before even investigating the claim itself — and those people would not be unreasonable to do so.
However, we do not conduct research in such a manner, so we will start to dig in!
Diving into the Alleged Evidence of Huobi Malfeasance
The link to the article that addressed the alleged leak is in Chinese, but we will still post the link here for your own edification:
The fortunate thing though, however, is that there are always Google Translate options. So, we’ll still be digging into the article’s contents to see what revelations we can discover.
What’s interesting is that the article translates from Chinese to English remarkably well…A little too well for a website that was allegedly written in Chinese orginally.
Why do we say this?
Well, because Google’s translation swap of English to most other languages is rough at best. It becomes downright abysmal when Asian and Eastern European languages are involved.
The Article’s ‘Preface’
In the preface, the article states,
“ Recently, an Excel spreadsheet of “Fire Coin Pool Node Account Data 20180911” was circulated in the EOS circle. Because the data in the table is detailed, it is not like external action, and the author of the data remarks part is Shi Feifei, the fire money employee. Therefore, it can be inferred that this data sheet is the internal data of the accidental leakage of the fire currency.”
After some research, its unknown what the authors are talking about when they make statements about:
- ‘Fire Coin’
- Shi Feifei
- ‘fire money’
- ‘fire currency’
It turns out that these are errors are due to an error in Google Translate. Using another online translator (Yandex), the proper meaning can be deciphered.
What the document actually states is that:
Recently, a “Huobi pool node account data 20180911” Excel Data Sheet spread in the EOS circle.Due to the detailed data in the table, unlike the external, and the data notes part of the author is Huobi employees Shi Feifei, it can be concluded that this data table is internal information Huobi inadvertently leaked.
So, this means that:
- ‘Fire Coin’ = Huobi
- Shi Feifei = Employee of Huobi
- ‘fire money’ = Huobi
- ‘fire currency’ = Huobi
While the data sheet is not directly provided in the article for users, it states the characteristics of the alleged data sheet that has been circulating around the crypto sphere.
According to the article, this alleged data sheet contains:
- Node mutual voting table
- Control node voting situation
- Node income statement
- Ticket position and account situation
The author of the article then goes on to state that,
It is these four data detailed data table, fully exposed Huobi and other nodes vote each vote, control other nodes split node revenue Insider!Although EOSONE analyzed Huobi“1 Zone 7”on August 16, 2018, the monthly income of 1.7 million?”, But with the Huobi this confidential internal information as a direct evidence, used to see the world of EOSONE still feel shocking!
To make the above translated quotation more understandable, here is how this should be interpreted (per our best, un-biased understanding with the assistance of a native Chinese speaker):
‘The four data categories on the Excel document show that Huobi uses its nodes on the $EOS network to organize votes, collude with other nodes, and fix the revenue that the involved parties will receive. Despite the fact that EOSONE had drawn attention to the unusually high amount of revenues that these nodes have received through the protocol, it is still surprising to learn this new information that has been released via the leaked spreadsheet.’
Who is EOSONE?
EOSONE has been confirmed (via outside research and sources) as the user behind this Twitter account:
The latest Tweets from EOSONE_io (@eosonebp). EOSONE is a non-profit international organization that provides services of proxy, DAPP research and development to the EOS ecosystem. People’s Republic of China
As is stated on their account, they are, “A non-profit international organization that provides services of proxy, DAPP research and development to the EOS ecosystem.”
This organization’s main purpose appears to be monitoring activity among BPs on the $EOS chain, specifically, and blowing the whistle on any and all activity that it perceives or concludes is nefarious in nature.
Their Steemit account (where their findings in detail/depth are produced) can be found here:
The latest posts from eosone. Follow me at @eosone. Join thousands on steemit who share, post and earn rewards.
They post articles in both Chinese and English.
A cursory glance of their profile shows that they actually have been effective in exposing some of the collusion occurring directly on the platform among other nefarious activities:
Something you need to know before reading: Our declaration is: not for the capital, but for the facts! Please indicate the source when you reprinting. The data in this article is as of September 6, 2018, 11:00. Who is Bitfinex? People who focus on the EOS must be familiar with it.
The above information in this sections should serve as a sufficient enough breakdown of the entity, EOSONE.
Lack of Physical Link/Upload to Spreadsheet Hurts Argument Slightly
Again, without the spreadsheet in front of us, it is difficult to validate the assertions made by the author. However, the fact that this article was released under a (somewhat)-reputable journalistic source in China (they aren’t the NYT or anything) makes it something worth investigating a bit further.
Without an obvious motivation for the author to fabricate the story or a denial from the individuals in question within Huobi’s organization (at the time of writing), the article must at least be considered to be tentatively true.
Also, the fact that collusion of this sort has already been observed (conclusively) among BPs in $EOS yields even more credence to the accusations that have been leveled against the platform.
“Mutual Vote Exchange Benefit Sharing”
Under this section of the article, the author shares the following image:
This chart will be explained in a moment, but first there is some background information on how $EOS works that must be shelled out first before continuing any further in order to ensure that all readers are on the same page.
If you consider yourself to already be knowledgeable on all things $EOS-related, then you should skip the next section and head directly to the heading labeled, ‘Back to the Chinese Exposition Article on Alleged Collusion Among Huobi and EOS Actors’.
What the Hell is All of This About?
Okay, if you’ve read this far into the article and you’re scratching your head because you can sense that this is a relatively ‘big’ revelation but the ‘dots’ aren’t necessarily clicking — don’t worry.
We’re going to help explain what the issue is below.
Understanding Block Producers on $EOS’ Protocol
So, if you don’t know about $EOS’ consensus method, here’s how it works in a nutshell (all information obtained from the whitepaper):
#1 — The $EOS Consensus Mechanism is Called ‘Delegated Proof of Stake’
‘Consensus mechanism’ refers to the way that blocks are produced on a protocol. For example, on Bitcoin, this is the ‘Proof of Work’ (mining) consensus.
What makes ‘Delegated Proof of Stake’ different is that it is devoid of the ‘mining process’ (there are no computers on the network that are completing calculations in a race to see which one will complete the block faster).
“Under this algorithm, those who hold tokens on a blockchain adopting the EOS.IO software may select block producers through a continuous approval voting system. Anyone may choose to participate in block production and will be given an opportunity to produce blocks, provided they can persuade token holders to vote for them.”
So, that’s where the ‘proof of stake’ portion comes from. If you hold tokens, then you can vote for block producers and any individual (node) on the network can technically be a block producer so long as they are able to receive enough ‘votes’.
In this scenario, tokens = votes.
There are 21 ‘block producers’ in all that are selected per ‘round’. Each ‘round’ on the protocol consists of 126 blocks and each block is created approximately every 0.5 seconds.
According to the whitepaper,
“At the start of each round 21 unique block producers are chosen by preference of votes cast by token holders. The selected producers are scheduled in an order agreed upon by 15 or more producers.”
Now, here’s the most important takeaway one should have regarding how $EOS works:
“A blockchain that adopts the EOS.IO software will award new tokens to a block producer every time a block is produced. In these circumstances, the number of tokens created is determined by the median of the desired pay published by all block producers. The EOS.IO software may be configured to enforce a cap on producer awards such that the total annual increase in token supply does not exceed 5%.”
The Problem Should Now Be Obvious
In case it is not obvious for you, we’ll lay out the issue here point by point (as it relates to the alleged leaked Huobi documents):
- Every block producer is guaranteed a reward on the chain (that they can keep) .
- Block producers are selected via tokens being staked on the protocol
- Because of #2, those that already have a large accumulation of tokens can easily vote for themselves.
- However, the idea brought about in #3 would more than likely not be sufficient enough for someone to become a block producer in the protocol because there is (hopefully for $EOS holders) not a single entity that possesses so many tokens that they would be able to vote themselves in using only their resources.
- The issue in #4 can be circumvented by aligning with a network of high-resource individuals that can collude among each other to combine their vast resources in order to ensure that there are certain block producers that are consistently selected on the protocol.
- Since these block producers that are being consistently selected will continue to receive an ever-increasing number of tokens on the protocol due to the fact that they are accumulating ‘interest’ as a reward, they will become perpetually enriched — allowing them enhance their ability to ensure that they are the primary block producers that are selected for each ‘round’ on the protocol.
Another Major Issue Underpinning These Allegations
The problem here is not just the fact that block producers can virtually guarantee that they will be block producers on the protocol indefinitely.
Another issue is the fact that the protocol operates by allowing changes to be determined primarily by either block producers or token votes. Here’s the issue with this system:
- If there is collusion of the sort that is alleged in the Chinese article, then these block producers must hold a significant portion of tokens on the protocol (that’s the only way collusion could happen on this protocol based on what we know).
- If this is the case then, based on the principles of $EOS (as stated in their whitepaper), they would have de facto control over the entire protocol.
- The implications of #2 mean that they would have full control over any and all changes/rule alterations that are being made to the protocol and this could have immense consequences for token holders.
The consequences of BP centralization on the $EOS protocol was perhaps outlined best in this Medium piece.
In the piece, there are a few important points that are outlined about the powers that token holders and block producers and they underpin an even more important concept, which is this:
The same features of $EOS that were allegedly designed to grant it superior ‘decentralization’ and efficiency are the same tools that can be used by colluding parties to degrade any level of centralization on the network.
The author of the Medium article cited above makes a cogent point in stating that,
“Understanding this, it seems clear that the term ‘platform-grade censorship’ is in fact intellectually inane.
If a system has been designed that ultimately encourages plutocracy and collusion among BPs then there is absolutely no guarantee for developers and users that their applications and transactions will not be censored.
Contrary to the myth propagated by certain actors in the blockchain/crypto asset industry right now, censorship resistance does not exist on a dynamic scale: it is very much a binary property.
Because EOS cannot guarantee censorship resistance, it should be viewed as a centralized system, without the throughput advantages that openly centralized networks like AWS provide. Spencer Bogart of Blockchain Capital writes well on this.”
Back to the Chinese Exposition Article on Alleged Collusion Among Huobi and EOS Actors
Remember when we posted this chart from the article?
The author then writes:
“The above table shows the mutual voting data for Huobi from September 4, 2018 to September 10, 2018.As you can see,：
1) in addition to Huobi’s own control of the 5 nodes of eoshuobipool, cryptokylini, eosiosg11111, cochainworld, eospaceioeos (detailed analysis below), Huobi will vote for the remaining 20 nodes.These 20 nodes, 16 nodes and Huobi mutual vote, leaving eosgenblockp, eosbeijingbp, eoseouldotio, eospacificbp, eoslaomaocom these four nodes did not return to vote.
2) Huobi new starteosiobp, eosflytomars, eoscannonchn and other nodes of the vote, these nodes also increased the number of votes on Huobi, Huobi can be said to pay a“return”.
3) atticlabeosb is the only one to vote for Huobi, but did not get Huobi back node.”
List of Alleged Colluding Nodes
The alleged collusion between the aforementioned BPs in the screenshot posted above and the information provided below significantly implicate a few of $EOS’ (current) block producers.
In order, they are:
A Reddit thread that provides even more probative evidence attesting to the above accusation can be found here:
Update: Denny Wu from Huobi said it’s a fake document from an exemployee ( happened at 13th Sept) But before he lied, the “exemployee” said something in their wechat group yesterday, no one said she is an “exemployee” Check the claim record of those puppet block producers.
While Reddit is not necessarily the pinnacle of scholarly research, the links and information contained within the post are noteworthy.
Below, is a screenshot from the Reddit post (in case it is deleted/removed/redacted/altered in any way):
The links in the post that were provided are as follows (in order):
All links from the above post (including pictures) have been archived here: http://archive.is/7nlfX
EOSONE BP Rank Information
Research information provided by EOSONE (cited earlier in this piece) shows the top block producers as of August 17th, 2018 (a little over a month ago from the time of writing).
Extracting this information is important to analyzing the potential impact of the concentration of Huobi nodes because this will allow us to determine how many of the nodes are allegedly controlled by Huobi as well as quantify exactly how successful/unsuccessful they have been over time by supplementing information given in the Chinese article and by EOSONE with data logged and tracked by the website, eosauthority.
The Top BPs are as Follows:
The Report Also Outlines the Specific Criteria Used to Aggregate Transparency Information From Block Producers
According to EOSONE, they made the decision to give each of the 83 top BPs analyzed various scores across a number of categories for the purpose of assessing their levels of; transparency, internationalization, community construction, activities and projects.
Below, is the document produced by EOSONE in the piece that shows the specific criteria they used:
Now, if you’re a viewer and you’ve made this far into the article, you may be thinking to yourself:
‘What is the objective purpose of these rankings and why are they important?’
This is a fair question to ask.
The reason why these rankings are important is because EOSONE is a known and reputable source of information in the $EOS community. Since the T21 block producers are (supposed to be) decided by votes, and the amount of votes one has is determined by his/her supply of tokens, the publishing of such information has a direct impact on who should be/is seen as legitimate.
For the purposes of this research piece, this ‘ranking’ system is effective because it allows us to quickly parse out the characteristics of all of the T83 nodes on the protocol and, most importantly, the Huobi nodes that have been called into question by the Chinese article’s release and the leakage of the alleged internal excel sheet.
Examining the Spreadsheet
In examining the spreadsheet, it becomes immediately clear that the nodes that were listed in that leaked Huobi document are indeed the same nodes that are located at the top of the hierarchy for $EOS BP nodes on the protocol.
In addition, if you look at the voting data for these nodes on the transparent EOS voting site, https://eosauthority.com/voting, you can see that these nodes have voted for each other much in the fashion that the Chinese article dictates that they do.
So, while there is not 100%, concrete and undeniable evidence that these allegations are true and that the spreadsheet is real, there is a significant preponderance of evidence at this point.
Basically, if we were to make this analogous to a court case, we have the body, the blood, a suspect, a motive, and individuals that spotted the suspect at the scene of the crime, but we are somewhat in doubt about the murder weapon.
To continue this analogous metaphor, the ‘victim’ in this case would be collusion among BPs on the $EOS protocol. This is something that even the most ardent supporters of the protocol must admit is true. The ‘blood’ here is the impact of this centralization on the protocol (i.e., the rankings of the BPs that we see). The motive is obviously to monopolize the chain to unfairly bolster one’s financial rewards from working with it and the suspect, obviously, is Huobi. The murder weapon though, the spreadsheet, is something that we cannot positively confirm 100%. However, there is a lot of evidence to corroborate the assumption that the evidence should at least be assumed true at this point until Huobi is able to definitively refute it.
Now, that logic sounds backward, because in almost every scenario, we accept that the burden of proof, in a scientific and legal sense, must lie upon the accuser in the majority of societies around the world. However, in this case, there is such a backing of evidence behind these circumstances and the accusations being leveled against Huobi in addition to the fact that Huobi was caught in a lie in their initial response, that they must now issue a legitimate response to explain why:
A) They initially lied about the nature of the spreadsheet and the employee that allegedly leaked it
B) The evidence provided by independent researchers on the protocol as well as the Chinese article strongly corroborate their stories.
What is included within this article and the spreadsheet should be considered a major deal, because it essentially undermines the protocol’s core goal:
Providing a scalable, decentralized dAPP platform.
Instead, $EOS is running as a highly centralized, corrupt platform that is unfairly manipulating the protocol in such a way where only a few can truly benefit from the protocol.
The impact of this system of collusion also makes the protocol inherently untrustworthy as well. Users cannot assume that their interests or even the best interests of the protocol will be preserved under the current state of things.
Obviously, these concerns have not just been voiced within this article, but in several different corners of the cryptosphere — including Daniel Larimer (founder of $EOS) himself.
In this Medium article, the opinion expressed by Daniel Larimer that ‘code is law’ is expounded upon at great lengths to help users and general crypto spectators alike gain an understanding of how he plans to reconcile the vast governance issues on the protocol:
Time will tell as to whether $EOS will be able to effectively fix these governance issues or not.
Thus far, they have failed miserably.