The race to build the best blockchain

Billions of dollars are being spent, right now, to solve a problem we didn’t know existed a few years ago. 

The problem boils down to creating a decentralized app platform that’s secure, scalable and governed in a fair, transparent way. And the solution might change the world in a similar way the internet did a few decades back.

This is the race to build the next blockchain. Blockchain, often defined as a “digitized, decentralized, public ledger,” is already a proven technology. But it’s far from perfect, and improvements could extend its utility far beyond cryptocurrency.

Bitcoin and Ethereum, the two largest cryptocurrencies by market cap, are widely regarded as immensely promising, hindered only by a few obstacles. Unfortunately, these obstacles aren’t all that easy to overcome.

Bitcoin, a decentralized currency and payment network, is secure and robust, but very limited as an app platform. It also doesn’t scale well, and is largely at the whim of a couple of large entities known as mining pools. Ethereum is far more versatile and a bit faster, but otherwise has the same problems Bitcoin has, and is arguably less secure.

Name any blockchain-based system that’s live right now, and it suffers from the same issues, only to different degrees. It either isn’t scalable enough, versatile enough, or decentralized enough. Building a system with all three traits is the holy grail of blockchain tech right now, and a lot of very smart, well-funded people are scrambling to be the first to do it. 

The what

Blockchain, as implemented in Bitcoin, is a way of keeping records. Instead of having one bookkeeper, everyone in the system has a copy of the book. Once a page is full of records, it’s signed and becomes an immutable “block” in the blockchain. The signing is done by solving a math puzzle; the people that do this are special network participants called miners. Thanks to the wonders of cryptography, once someone solves the puzzle, the entire world can easily check whether the answer is correct. The book’s history cannot be changed; if you tear out or change one page, the math breaks down and the network discards your version of the book. 

This is typically slower and less practical than simply having one entity — a person or a company — keep the records themselves. But there are benefits to removing the bookkeeper; once he’s gone, it removes trust from the equation. When there’s no way to fake a transaction, there’s no need for a middleman. You can send and receive money without a bank. You can sign contracts without a lawyer. You can support artists without a publisher. 

The list doesn’t go on forever. You’ll find snake oil salesmen at every corner, telling you that the blockchain is the future of everything (and probably selling you some sort of token). It isn’t. But there’s something about this technology that promises to usher a fundamentally new way we do things. 

That’s because Ethereum and other decentralized app platforms let you not only keep records of financial transactions, but run entire apps on the blockchain. Applications on Ethereum, says the project’s website, “run exactly as programmed without any possibility of downtime, censorship, fraud, or third-party interference.”

This paves the way for a whole new level of corporate and governmental accountability. Imagine a social network which cannot mess with your data any way it pleases — at least not without everyone noticing — or a bank that cannot raise your interest rate at whim, because the code just won’t let it. 

Ethereum is by far the biggest platform in terms of adoption right now. However, Ethereum is not nearly scalable enough for serious business use, which became painfully apparent in Dec. 2017, when the network slowed down and transaction fees skyrocketed due to heightened interest in a game that let people collect digital kitties. 

Ethereum is working hard to make the system more scalable, robust, and less energy-intensive. But its foundations were laid when blockchain technology was still relatively immature; many of its competitors have the luxury of being able to build their technology from scratch. Who will win the race?

The who

Things move incredibly fast in the blockchain world. Ethereum is three years old. Projects like Cardano and EOS, sometimes called “blockchain 2.0” projects, are already considered to be giants in the space. They have a combined token market cap of roughly $11.8 billion despite barely being operational.

Cardano, which focuses on a slow and steady approach, with every iteration of the software being peer reviewed by scientists, is promising, but it hasn’t fully launched its smart contract platform yet.

EOS, an incredibly well-funded startup that launched in June, is another huge contender. However, EOS has a complicated governance process which caused a fair amount of trouble right after the launch, together with a slew of freshly discovered bugs. With an estimated $4 billion in pocket, EOS has the means to do big things, but it will take some time to see whether it can live up to the promise. 

But there’s already a new breed of blockchain startups coming. They’ve been working, often in the shadows, to develop new concepts and technologies that may make the promise of a fast, decentralized app platform a reality. 

A startup called Dfinity brands itself as the “Internet Computer,” claiming it will usher a new era of cloud computing, which will be more secure and reliable than today’s computing, with better interoperability and built-in privacy. 

It’s one of many blockchain-based startups penning such lofty claims, but Dfinity backs them up with an incredibly strong team of mathematicians, cryptography experts and computer scientists. Andreas Rossberg, for example, is an ex-Googler and co-designer of WebAssembly, a web standard that lets developers run complex, high-performance apps inside your browser. Ben Lynn, another ex-Googler, is a cryptography pioneer, the co-creator of the BLS signature scheme. 

At launch, the value of Dfinity tokens — its market cap —  will be just shy of $2 billion.

Dfinity’s president and chief scientist, Dominic Williams, claims that assembling a superstar team like this is hard. “We’ve got a lot of superstar researchers and engineers. We see building at Dfinity as semi-analogous to building at Google. And we believe that, if you really want to have an internet computer that runs the world’s software, it’s not just about releasing some protocol. It’s about having an organization that can support that network, that’s large enough that it can provide people with the confidence they need to actually a build production system on the top of the network,” he told me during a long, insightful conversation in June. 

It also isn’t easy to get hundreds of millions of dollars from top investors without a working product (unless you count the private beta), but Dfinity has done it. Williams told me the company raised roughly $150 million in funding so far from a number of investors that include Andreessen Horowitz and Polychain Capital. Even more importantly, the money raised values the Dfinity tokens — yet to be released — “just shy of $2 billion,” which will instantly make Dfinity a major blockchain project at launch time. 

If that’s not bleeding-edge enough for you, you can turn to projects like Ava. Led by Cornell professor and notable voice in the crypto space, Emin Gün Sirer, it’s a new type of decentralized system that’s based on Avalanche, a technology laid down just months ago by an anonymous team of developers called “Team Rocket” (Bitcoin itself and Mimblewimble protocol are also created by anonymous developers; notice a trend here?).  

“Avalanche offers three or four orders of magnitude of speed-up over most blockchain systems in use today,” Sirer told me over a coffee. Again, this is made possible with some smart math, with the result being, ideally, a network that’s incredibly fast and scalable — and does not waste energy. Like so many up-and-coming blockchain projects, it hasn’t been tested in the real world yet. 

The where

I caught Williams and Sirer during the Crypto Valley Conference, in the Swiss city of Zug (the capital of the homonymous Swiss canton), where the old world of Swiss banking is trying to reconcile with the new, propulsive world of blockchain tech. Zug, a tax oasis within Switzerland, has attracted thousands of businesses, and some of the most important blockchain projects, including Ethereum Foundation, which governs the development of Ethereum. The conference was opened by Zug’s mayor Dolfi Müller, who readily admitted that just a few years ago, a couple of youngsters taught the city leaders about the importance of Bitcoin. 

But a lot has happened since then. In June this year — in just one example how serious this blockchain stuff is in these parts — the city of Zug, together with a company called Luxoft and Lucerne University of Applied Sciences in Switzerland launched a blockchain-based e-voting system.

I did not go to the Crypto Valley by accident. Blockchain conferences are dime a dozen these days, but it’s not very often that some of the most promising, up-and-coming projects come together at one place. Besides Dfinity and Avalanche, visitors had a chance to see a presentation by the incredibly ambitious Aeternity, led by Yanislav Malahov, a blockchain expert who claims to be an early influence on Ethereum. Stefan Thomas, ex-CTO of the huge, biz-oriented cryptocurrency Ripple, was there as well, presenting his new project Codius, an open-source platform for hosting applications. 

In true decentralized fashion, however, things are happening for blockchain in numerous other places as well; Silicon Valley, while still a hugely important hub and source of VC funding, is not the only place to be. Malta has recently created a regulatory framework for blockchain technology. Luxembourg and Gibraltar are important European hubs as well; in Asia, it’s Singapore and Japan that lead the way. Each of these countries — most of them small, flexible, fast-acting and forward-looking — has attracted a number of cryptocurrency projects, looking to set up shop at a place that will support their lofty ambitions. 

The how

Are these ambitions based on reality, or is the vision of the ultra-fast blockchain that powers demanding apps with hundreds of thousands of users a mirage?

I asked Sirer whether a scalable, decentralized, energy-efficient and well-governed blockchain is even possible, or is it a case of “pick two, maybe three.” He laughed, implying that I got this one very wrong. “Just because a problem is hard to solve right now, doesn’t mean that it cannot be solved. There are things that I can give you an impossibility proof for, but this is not one of them.” he said. 

“Just because a problem is hard to solve right now, doesn’t mean that it cannot be solved.”

The foundation of his Ava project, Avalanche, is not technically a blockchain. It’s based on a DAG, or directed acyclic graph, a system in which data spreads laterally instead of being stacked in blocks. It has a novel way to achieve consensus on the network, with some cool math backing it up (check here for a thorough explanation), and it’s radically different than well-established consensus mechanisms. It promises to be much faster than most, with 1,300 transactions per second achieved in a test scenario, while staying reasonably secure from attacks. 

If you’re getting confused, I don’t blame you. Here’s a (very simplified) explanation of what these projects aim to achieve and the practical problems they face:

The DAG and the blockchain are just ways to organize data in a decentralized network. On top of that, you need a consensus mechanism — a way to make sure network participants agree on what’s happening within the network. In Bitcoin, for example, this is done through something called Proof-of-Work, in which special network participants called miners employ computing power to solve a hard math problem, whose solution “signs” a block of transactions. This makes it hard for someone to forge a transaction on Bitcoin’s network, but it’s also a terrible waste of energy. 

A permissioned consensus mechanism is easier to secure, as it only accepts transactions from pre-approved participants. But the types of blockchains we’re looking at here are permissionless, meaning their consensus mechanism needs to be resilient to certain types of attacks. For example, it needs to be Byzantine-fault tolerant, meaning it can withstand a certain number of nodes on the network being malicious (i.e., spreading false info). It also has to be resistant to a Sybil attack, which boils down to forging identities to gain an undeservedly large influence on the network. 

When you have scalability and security in place (having in mind that those are always moving targets), you still have the issue of governance. Ethereum, for example, does this in a centralized fashion, with the Ethereum Foundation calling the shots (though it’s always possible for the community to “fork” in a different direction, as it once did, when Ethereum Classic was created). But it’s also possible to bake a governance algorithm into the system itself. 

Dfinity, Williams told me, has invented novel ways to achieve each of these goals. The project’s whitepaper is a tough read even for experts in the space, but Williams sat with me (for far longer than I expected him to) and explained it until I showed signs of understanding the basics. Dfinity’s team invented a whole new way to create randomness, a tough problem in cryptography. This, in turn, enables for a new consensus algorithm, allowing for very fast finality (i.e. the amount of time needed for one block in the network to be fully verified). Finally, for governance, Dfinity employs something called the “blockchain nervous system,” which is a type of democracy in which network participants can choose to automate their decisions based on the decisions of other participants whom they trust. It should ideally result in a fair system that auto-regulates itself. 

“It’s not necessarily about being better than Ethereum, but doing things in a different way.”

For the layman, it’s impossible to decide which of these systems is the best – and there are dozens of others I haven’t even touched upon here. In fact, even experts disagree — follow people like Dfinity’s Dominic Williams, Ava’s Emin Gün Sirer, Ethereum’s Vitalik Buterin and Vlad Zamfir, EOS’s Dan Larimer, and Cardano’s Charles Hoskinson on Twitter, and you’ll see they constantly keep jabbing at each other, finding flaws in the other team’s tech. Real-life adoption and usage will ultimately settle this debate, but until then, all that most of us can do is follow it from the sidelines (ideally with some popcorn). 

Ultimately, it doesn’t really have to be a winner-takes-all situation. I’ve asked Williams whether Dfinity and Ethereum can co-exist, and he told me they can. 

“It’s not necessarily about being better than Ethereum, but doing things in a different way,” he said. 

The when

Figuring out when the debate will ultimately get settled is equally hard. But even with all the funding that’s gone into the space — ICOs raised a total of $6.3 billion in the first quarter of 2018 by CoinDesk’s count — it looks as if it’ll take a few years for the next generation of blockchain technology to truly mature.  

Ethereum is slated to implement major changes in two to three years, but its roadmap is quite fluid. After its big launch, EOS needs to optimize and start showing results it’s been promising; Cardano appears to be slightly behind, but its leaders hope that their slow, steady, peer-reviewed process will win in the end. 

Projects like Ava and Dfinity have far more obscure roadmaps. 

“We’ve got so much breakthrough tech, we don’t want to share all of it publicly until we’re ready. Who knows, we may just make one big, Apple-style event, when we’re fully ready, to show all we’ve got,” he told me. He never gave me a precise estimate, but Dfinity’s own documents mention Q1 2019 as one possible date for the launch. 

As for Ava, Sirer told me that the project is currently in the early stages of raising funds. There will likely be an ICO down the line, but no date is set. “People have been contacting me as soon as the whitepaper hit the airwaves. There’s a ton of interest already,” he said.

This sort of optimism is common in the space, even for projects which are years away from being market-ready. “Massively oversubscribed” is a phrase I kept hearing regarding funding — VCs have entered the space and are now looking for the next big thing and, often, capitalize on the ICO rush. 

But beyond the ICO craze, some of these teams may just succeed at really building something that deserves the name “internet computer” — a decentralized, democratic platform that lets you run apps resistant to censorship, their inner workings transparent to all. 

Disclosure: The author of this text owns, or has recently owned, a number of cryptocurrencies, including BTC and ETH. The author has no stake, and has not recently had a stake, in EOS, Cardano, Dfinity, Ava, Aeternity or Codius.