Beyond Ethereum (Part 1: Intro to Blockchain)

I found out about the blockchain early in 2013 when the price of Bitcoin surged past $200. It was exciting technology and potential a financial opportunity through mining, but was also commonly viewed as a potential scam and tool of black marketeers on the dark web and silk road. My research at the time suggested it would be difficult to break even on electricity costs at that price point, and my friends advised me not to throw away my money.

I listened, but who could have envisioned that a few short years later, despite scams like the DAO, multi-billion dollar hacks like Mt. Gox, and general skepticism of legitimate uses for the technology, the price would continue to rise. Average Joes (and Joans) became millionaires overnight, and millionaires like the surprisingly prescient Winklevoss twins would become crypto-billionaires.

The core concept of blockchain is the distributed ledger. In the case of bitcoin, this ledger is used to record transactions between users, which are verified then stored immutably (can’t be changed and stored forever) on the blockchain. The distributed aspect is that anyone around the world can use their computer to “mine” or add new blocks to the blockchain, by solving a cryptographic puzzle using Proof of Work (POW). It takes about 10 minutes on average to solve this puzzle even with advanced mining equipment, and each block written to the ledger is linked to the previous one.

The end result is data can’t be forged or altered without using massive amounts of computing power required to rewrite one or more blocks and extend the chain while solving the POW puzzle for each altered block. Because the system trusts the longest chain and verifies the POW for each block, an attacker would need to compute new solutions and extend the chain faster than the legitimate miners do, which is virtually impossible without having control of more than 51% of the computing power in the network (the infamous 51 percent attack).

By the summer of 2017 when I enrolled in Draper University, the price of bitcoin was over $2000, and I was kicking myself for missing the wave. Little did I know it would continue to rise another 10x to reach a peak of $19,783 by December of that year, as blockchain broke into the mainstream and became a popular topic in barbershops and taxicabs. Accompanying this explosion of recognition was the phenomena of the Initial Coin Offering (ICO), which uses cryptocurrencies like Bitcoin as fundraising vehicle for new technology projects based on the blockchain. And at the eye of this crypto-storm was a new type of blockchain “Ethereum”.

The key innovation of Ethereum is in understanding that distributed ledgers can track more than financial transactions (just like normal ledgers and databases are useful to people who aren’t accountants). Ethereum uses the blockchain to track ‘state’ for a distributed virtual computer. For anyone who has taken a class in automata theory, you might recognize the power of this concept, as state machines can be used to model complex systems and create Turing complete programming languages. When combined with blockchain technology for tracking payments, this generates a powerful new capacity to programmatically link financial incentives to changes in information stored on the blockchain.

In layman’s terms, Ethereum allows users to write programs on the blockchain that can update information stored on the ledger and react to changes in that data. This is the concept of the “smart contract”: a program that is stored on the blockchain (and therefore can’t be altered or amended) that can automatically react to changing events in a predetermined fashion. Just like “dumb contracts”, the smart versions specify outcomes linked specific events, but instead of using lawyers and judges for enforcement, the actions happen automatically according to the programming. The details of the contract are published on the blockchain for anyone to verify before entering into the agreement, and all parties can trust the outcomes because the code can’t be changed.

One of the most common uses of smart contracts on Ethereum is to create new types of cryptocurrencies that can be used as Bitcoin alternatives (altcoins). These most commonly follow the ERC20 token standard which allows anyone with basic programming ability to easily create their own currency in a format that can be easily added to exchanges where they can be bought and sold. This is the innovation behind the explosion of ICOs seen over the course of 2017, when around 870 companies raised over 6 billion dollars (yeah that’s billion with a b). For everyone keeping score at home, thats nearly $7 million dollars raised on average per company. (source: https://www.icodata.io/stats/2017).

For many startup founders, this is an amazing innovation for fundraising because it grants access to unprecedented amounts of capital while sidestepping the notoriously painful process of raising money from venture capitalists and other traditional investors. However, it’s not all sunshine and roses, because those traditional funding sources also serve the important purpose of vetting ideas and filtering out projects that aren’t feasible or that don’t have real markets.

ICOs allow companies to raise funds with little more than an idea, a website, and a whitepaper to explain the concept, by going directly to consumers around the world who can each contribute small amounts towards funding the project. But many of these projects are half baked ideas or outright scams, and there are limited protections for consumers who invest in these projects when things go sideways. While there are plenty of legitimate companies creating real innovations in the space, there are even more hoping to cash in on easy money through crowdfunding with tokens and slick marketing campaigns, and real, working products are unfortunately few and far between.

After increasing enforcement actions by the SEC in the US and in other goverments around the world (China outright banned ICOs in Sept. 2017), the wild west days of unregulated ICOs might be behind us, but the ICO phenomenon is here to stay. In just the first 4 months of 2018, more than 600 companies have raised ~$4 billion dollars (source: https://www.icodata.io/stats/2018). Much of this demand for ICOs is driven by speculators looking for the next big coin to make a quick flip of their investment, chasing the outsized returns enjoyed by early investors in the most successful blockchain projects.

Despite the crash of Bitcoin prices in early 2018, where the value dropped by almost two-thirds, many are still bullish on the long term prospects of cryptocurrency (such as famed venture capitalist and blockchain investor Tim Draper, who predicts prices soaring to $250,000 per Bitcoin by 2022). This optimism is despite the notorious lack of consumer adoption for blockchain projects, which is largely due to the fundamental technical limitations of the most popular blockchains Bitcoin and Ethereum.

Blockchains based on Proof of Work are slow by design. The 10 minute delay in minting new blocks for Bitcoin is essential to the security of the network. As mining equipment becomes more advanced, miners are tasked with solving more and more difficult POW puzzles, which prevents uncoordinated miners from aggregating enough computing power to write corrupt data to the blockchain. Unfortunately this also means increasing amounts of electricity are required to operate mining equipment, and much of this power comes from cheap but dirty energy sources like coal that cause pollution,

There is also a trend towards consolidation of mining power in the hands of a few large conglomerates who can afford to develop specialized processing chips for mining. Currently the four biggest mining pools (BTC.com, AntPool, ViaBTC, SlushPool) control over 60% of the computing power in the Bitcoin network (source: https://blockchain.info/pools). This is a dangerous situation, because if these miners were to collude together, they would have the power to rewrite the blockchain’s history. While this is theoretically unlikely because such collusion would likely cause an immediate loss of trust in the network and a massive crash in the price of Bitcoin, this concentration of power in the hands of the few goes directly against the decentralized vision underpinning cryptocurrencies.

Ethereum is currently a proof of work system and shares many of the same challenges, though efforts are being made to convert the network to use Proof of Stake (POS) consensus. With POS, new blocks are created and validated by members of the network according to the number of tokens they stake to confirm the block. Instead of being rewarded for solving puzzles, POS block creators depend on transaction fees to earn revenue and offset costs. If users are dishonest and attempt to confirm invalid blocks, they can lose the tokens they staked. While this eliminates the need for power intensive mining and mitigates some of the mining pool concentration risk seen with POW, a minimum stake is required to become a validator, which prevents some users from participation in mining new blocks.

The bigger issues with Ethereum are speed and scalability. It currently takes about 20 seconds for new blocks to be created on the network, and transaction volume is limited to about 5 transactions per second. When compared to credit card processors like Visa who have regular process nearly 2,000 transactions per second and have capacity for up to 50,000 tps, it is clear that Ethereum has a long way to go to be ready for consumer friendly scale. These problems are readily apparent when large ICOs are launched on the network, which create a backlog of transactions that can cause massive delays in transaction processing, as seen with the launch of the popular CryptoKitties virtual pets game.

If these limitations don’t create enough headaches for developers who want to build consumer facing applications on Ethereum, there is also the issue of high transaction costs for sending transactions to the network. The limited capacity for transaction processing on the world most popular smart contract platform leads directly to increased competition between projects, who must pay progressively higher gas fees or risk waiting longer and longer periods of times for their transactions to be picked up.

Ethereum supporters point to innovations such as the Raiden network and Plasma project to solve these scalability issues, however there are unresolved technical issues with both of these approaches and neither has a clear release date. Estimates for the availability of scalable solutions for Ethereum range from months to years. Ethereum has a significant advantage in developer mindshare, but it isn’t the only game in town. Developers shouldn’t give up hope, because there are a number of alternatives to building on Ethereum, ranging from promising new blockchain projects that are scheduled for release in 2018 (checkout EOS, DFinity, NEO, and others) to existing alternatives that have been available for years but haven’t been well publicized (Ark, Bitshares, Lisk, Steem). I will explore a few of these projects and compare them with Ethereum in upcoming articles.