Dawn of The Third Generation - Some notes on the Upcoming EOS Blockchain

With less than a week before the launch of the EOS mainnet I wanted to collate my thoughts on EOS and make up my mind about all the hype. Here are my tarted up notes from that process. None of this is investment advice but I hope it’s useful as background information.

What is EOS?

EOS is touted to be a platform that will support the creation of decentralised applications (DApps.) These are applications that run across a network, utilising the computing resources of participant nodes (computers.) So then we can think of EOS as Windows for blockchain. It will act as an operating system with a suite of tools handling the distributed-computing and blockchain processes in the background, as well as tasks such as user account management, authentication and database operations, to name but a few. Developers will be able to build applications running on the EOS operating system without having to worry about the low level nuts and bolts of working across a globally networked blockchain system.

Why is it called the Ethereum Killer?

The Ethereum blockchain ecosystem already offers some of the above functionality and is currently the main platform used for issuance of tokens that underly the vast majority of ICO projects. However, it has some drawbacks that EOS promises to improve upon. The main ones are:

1. Fees. Every computing operation on Ethereum incurs a cost and therefore, to run an application or a transaction on Ethereum a user has to pay in ‘Gas’ each time. This is cumbersome and not user-friendly. EOS won’t require transaction fees to run applications on its blockchain. This is a massive deal! So for example, a social media website run on EOS would not require a fee to be paid every time a user ‘liked’ a post or made a comment, as would be the case with Ethereum. A recent video on LinkedIn shows a game of space invaders running on the EOS testnet in which every move and shot is registered as a transaction. This could not be done on Ethereum due to the transaction fees that would be incurred and also, for the next reason…
2. Speed. The Ethereum network currently runs at around 15 transactions per second (TPS.) EOS will run at 1000’s of TPS just to start with and will eventually scale up to millions of TPS. Ethereum developers are working on scaling solutions (such as Plasma) which promise to solve the issue but these are still experimental and use bolt-ons (off chain solutions), whereas EOS is fast by inherent design. It uses the graphene blockchain model used and proven by two previous projects, Bitshares and Steemit. Low latency will mean that near real-time applications will be possible on EOS whereas they are not possible on Ethereum at present.
3. Ease of Use. Ethereum programs (called smart contracts) can only be programmed in Ethereum’s proprietary language, called Solidity, whereas EOS will allow many popular programming languages to be used, making it easier for more Developers to create applications on EOS. Furthermore, data stored on EOS will be human-readable through the use of self-describing databases and self-describing interfaces. A useful web-toolkit for creating interfaces will also make the life of developers easier.
4. Flexibility. One major problem with Ethereum’s Solidity is immutability. Once a program (contract) Is deployed on the Ethereum blockchain, that’s it. There’s no going back to change it, ever. So developers tend to keep Ethereum contracts simple and have to do a huge amount of testing to iron out bugs and vulnerabilities. The EOS system promises to be very flexible – you can freeze misbehaving programs and fix them, so a situation such as the DAO hack of 2016 where 3.6m Ether were stolen and the only way they could fix it was by forking the whole blockchain and creating an entirely new coin in the process, would not happen with EOS. The malicious code could be frozen by an EOS block producer and any hacked funds could be returned to the owner by a process of arbitration.
5. User-friendliness. One major problem with current blockchain applications is that they usually require end users to be tech savvy and know about public/private addresses, transaction fees (e.g. gas), wallets etc. There is a high degree of danger for the non tech-savvy including getting hacked or sending funds to the wrong address.
   EOS promises a paradigm where applications running on it will look no different to end users than current websites. They will not realise that the website is running on a blockchain. Also, if they lose their password or it gets stolen, it will not necessarily mean that their funds are lost forever, in much the same way that if a thief were to steal your housekeys it would not mean that they now own your house. This is the situation currently in blockchain. EOS hopes to solve this crucial issue that makes adoption by the masses fraught with worry and difficulty.

About Dan Larimer

Dan Larimer is known as a visionary programmer and systems inventor within the Blockchain industry. He received his Bachelors degree in Computer Science from Virginia Polytechnic Institute and State University in 2003. Dan became interested in Bitcoin in the very early days and began communicating online with Satoshi Nakamoto, albeit eventually getting a stern reprimand from the enigmatic Bitcoin creator!

He is the main software engineer behind the Graphene software platform and founder of two previous mainstream blockchain projects, Bitshares and Steemit, both currently around no.30 on coinmarketcap. He also invented the Delegated Proof of Stake consensus algorithm (DPOS) which is used by Bitshares, Steemit and will also be used by EOS.

Dan’s stated mission in life is “to find free market solutions for securing life, liberty and property”. By this he means to create trade and social solutions, including money, using technology, outside of control by third parties such as governments, banks or multinational corporations. He has studied Austrian Economics and uses those principles to try and “engineer the economic incentives which make freedom and non-violence profitable.”

According to Larimer, the first big project he founded, BitShares, proved that a Graphene blockchain could run at far superior performance and scaling levels than existing blockchains. Its aim was also to overcome the problem of exchanges that can fail, e.g. MtGox, or that can be shut down by governments. Hence Bitshares was created as a decentralised exchange (DEX) and was the first blockchain to achieve 10s of thousands of transactions per second. It had named accounts and built-in governance through DPOS. However, mass adoption was troublesome as people still had an issue with trusting the platform and it was aimed only at people wanting to make financial trades, which not everyone wants to do. Also, it still used transaction fees which was another roadblock. Larimer eventually left the project as he felt his input was no longer required, although this is a bone of contention amongst his critics. However, as a proof of concept to demonstrate performance and scaling it succeeded.

He then moved on to his next project, Steemit, which aimed to solve the problem of marketing and discovery – by being free to use, having no transaction fees and rewarding users for creating content. Steemit was very successful and remains one of the top 2000 websites globally. Larimer also left this project – again stoking his critics. Larimer himself explains that he decided to move on because his “ability to continue to add value became relatively small” since most of the ongoing work was front end web development and making it more user friendly, which is not really his main strength. Steemit accomplished the goal of showing that a social media platform could be bootstrapped to mass scale using a transaction free blockchain model and also proved that a website could be built where the users did not realise that it was running on a blockchain on the back end.

Larimer is now the Chief Technical Officer at Block.one, the Cayman Islands based company running the EOS ICO. He believes he and his team will create a multi-purpose platform assimilating all the useful ideas learnt from BitShares and Steemit and learning from the mistakes. Due to the performance levels of Graphene, Larimer envisages that almost any application could be built upon EOS and the end-users’ experience would be identical to an application running on centralised servers. But due to the nature of decentralisation these applications would be outside of 3rd party controls, and also, blockchain/token economics would mean unique incentives could be implemented to engage and reward all users. EOS DApps will be scalable to hundreds of thousands, even millions of TPS, with very low latency and easy maintainability and upgradability.

According to Block.one’s website:

“EOSIO is a compliant blockchain protocol that enables horizontal
scaling of decentralized applications, allowing developers to
efficiently create high performance distributed applications. The
EOSIO software provides accounts, authentication, databases, and the
scheduling of applications across multiple CPU cores and/or clusters.
This allows for horizontal scalability, replaces user fees with an
ownership model, and powers simple deployment of decentralized
applications.”

Decentralised Governance

A key component for the long term success of a blockchain is the governance system. Who is in charge, who decides what direction the software evolves in, and who is susceptible to attack or control. Good governance enables the system to adapt in the face of technical challenges, evolving use cases, and ideological differences.

Ideally governance should be totally decentralised and democratic, whereby all participants in the blockchain are able to have a say through a voting mechanism. With the original daddy blockchain, Bitcoin, governance was not explicitly baked in by Nakamoto, but instead is an emergent quality of how Developers and Miners interact. Developers suggest improvements and miners will support or reject the changes. Sometimes different groups of large-scale miners will disagree and two competing chains can ‘fork’, creating new coins such as Bitcoin Cash. The trouble with the Bitcoin kind of governance is that it becomes very political and the competing interests of mining groups overtake what is good for the community. Large, industrial miners such as Bitmain will be at odds with smaller miners and sometimes with the Developers. They will promote decisions that protect their sizeable investments in mining equipment.

Although the original idea was that mining would be spread globally and the “one cpu. one vote” model would ensure decentralised governance for Bitcoin, it has turned out that an oversight in the Proof-Of-Work mining algorithm used by Satoshi Nakamoto allowed the emergence of ASIC mining hardware that has caused the centralisation of hashing power into the hands of a few large industrial level miners. In fact, just the 3 largest Bitcoin mining pools account for over 50% of the hashrate of the Bitcoin network - not so decentralised.
The situation is similar for Ethereum, where the top 3 mining pools account for almost 60% of the hash power.

Building in decentralised governance is therefore difficult but Dan Larimer believes his DPOS consensus mechanism is the way forward. EOS will have an in-built governance system consisting of 21 Delegates, known as Block Producers (BPs). These 21 BPs will be voted for by normal EOS token holders and will then be the only ones who can finalise new blocks and earn EOS for doing this work. 15 of the 21 will have to agree on a block for it to be accepted onto the EOS blockchain and a new block will be created every 0.5 seconds through this process. A transaction will have 99% confidence after 0.25s. Within 1 sec the majority of BPs will have verified the transaction.
The BPs will have to work together to achieve consensus on blocks, otherwise none of them will get paid. Also, there will be a disincentive for any BP to be a bad actor as they would just get voted out and replaced by a different standby BP.

From a security point of view, Bitcoin and Ethereum could potentially be disrupted by seizing a few pools or taking down some large miners, since they are resource hungry and difficult to relocate due to their large physical scale of operations.
DPOS on the other hand is agile – new BP servers can quickly be set up anywhere in world – this is because DPOS does not require a wasteful overhead of burning electricity to mint coins.

Why such a long ICO (1 year)?

One big criticism of EOS, Block.one and Dan Larimer is the amount of money they have accumulated during the unusually long ICO of EOS, which has taken place over a full year of daily token sales. Overall, Block.one will have raised over $2billion USD and probably closer to $3bn. Do they really need such a huge amount of money to develop their software, or is this an elaborate scam?

In interviews, Larimer suggests that Bitcoin and Ethereum are themselves one long ICO. He argues that Bitcoin and Ethereum miners ‘buy’ tokens by way of burning huge amounts of capital in the form of electricity. But in the case of the EOS ICO, instead of all that capital going into heating up the planet, it has been captured (by Block.one) and will be put to good use by funding the ideal of the free society, through educational initiatives and by funding new ventures. The EOS.io website has a section for the new funding vehicle called EOS VC that states:

“Block.one has committed to investing over $1B into funds focused on
the growth of the EOS ecosystem.”

They will invest in people and companies globally and will use their blockchain expertise to help these companies directly to reach their full potential. They will also partner with venture capital firms around the world, who can then work with founders in all markets at a local level.

Thus far, EOS VC has announced deployment of capital through partnerships with:

1. Michael Novogratz’s Galaxy Digital
2. Derek Rundell of Eric Schmidt’s TomorrowVentures
3. Finlab AG
4. Michael Cao’s EOS Global

Timeline for the main net launch

Current EOS Tokens on the Ethereum network will become fixed (non-transferable) within 23 hours after the end of the final EOS Token distribution period which will occur on June 1, 2018 at 22:59:59 UTC. Therefore, anyone who still has not registered their tokens must do so asap or they will become worthless. For details on how to do this visit https://eos.io/instructions and click on the ‘Register’ tab.

Soon after this deadline, a snapshot will be taken of all the ethereum addresses holding EOS tokens and the EOS public blockchain will be launched. Whether this process will go smoothly or require more than one attempt is unclear but all eyes will be on EOS.

Price predictions

At the time of writing, the EOS ERC20 token is trading at $12.50, having seen an all time high of $23 in April. Concurrent with general market sentiment, the price fell back sharply since then but is starting to recover slightly as we near June 2. There are many future price predictions being thrown out over the internet of anywhere from $30 to $200 by the end of 2018, but the simple truth is that nobody knows what the short term price action will be. Often, there is a sell-off after much-anticipated events which wrong-foots inexperienced speculators, so watch out for that next week after the launch happens.

Bear in mind, a project like EOS will truly bear fruit over a longer term investment horizon. But also note there are also going to be lots of token airdrops by new projects launching on the EOS platform, so token holders can pick up free bonuses along the way.

Also, if the price of EOS reaches a high level and it becomes prohibitive for new projects to buy enough EOS tokens to run their application on it, then a rental market will emerge, whereby existing token holders can rent out their slice of the EOS network, thereby providing a passive income. There are some good videos on YouTube explaining this ‘real-estate’ model of EOS.

Final Take

With so many ICOs launching on a daily basis, does EOS stand out from the crowd at the end of the day? My conclusion at this stage has to be yes. It could all be a huge scam and Dan Larimer could be just a restless dreamer/scammer jumping from one project to the next.

But what makes this project stand apart is that unlike most ICOs that have good intentions but no experience, or are using tokens where they just aren’t needed, Larimer and Block.one have a proven grasp of economics, game theory and software engineering. The team has built successful blockchains already and brings to bear a huge wealth of experience learnt the hard way, by actually doing. Block.one has a well thought out strategy and a huge stash of money which they appear to be deploying in an organised and productive way.

Dan Larimer states that he is making a long-term commitment to Block.one and EOS. He says that building communities is a lot of work and he doesn’t start over lightly, but only when there are intractable problems. He sees EOS as an extensible, programmable, high performance platform upon which his future ideas will be built. In other words, his next projects will be on EOS, not instead of EOS.

Only time will tell whether this vision will be realised – and please remember there are other general-purpose platform projects out there competing for this space – examples include TRON, NEO, Cardano and of course, Ethereum.