CHAPTER 5

Digital Cats Will Change the World

On the 30th birthday of the World Wide Web, Tim Berners-Lee confirmed what we all secretly believed. Asked by journalist Samira Ahmed,¹ “Did you think there would be so many cats on the web?” Berners-Lee replied, “That was the plan all along, to fill the web with cat videos.” Joking aside, the social aspect of the internet is something that was not immediately foreseen when the technology started to infiltrate our homes. The earliest use of the World Wide Web reflected our habits and activities at the time, making them arguably more efficient and accessible, but fundamentally unchanged. We sent e-mails or set up web pages instead of sending letters (in my case and for many others these did, in fairness, feature a reasonable volume of cat pictures). We gradually came to trust stores on the internet to take our credit card details and keep them safe and secure, moving shopping online. The rise of the cat video came from the new social constructs of MySpace, Facebook, Twitter, and the explosion of platforms that now surround us, and from the peer-to-peer file sharing which preceded peer-to-peer payments. Our behaviors changed because of the technology and would not have done so without it.

The Innovation Game

One of the dangers of enterprise-led innovation is an underlying requirement to maintain the successful business model. Following the principle of “if it ain’t broke, don’t fix it,” even the most inventive companies can find themselves simply iterating for lower costs, or for the faster completion of existing processes which remain fundamentally unchanged. The risk associated with radical changes in behavior is sometimes too much for investors who are keen to realize a return in a short timescale. Return on investment can be difficult to quantify when technology is emerging. Project assessors will struggle to prove estimates of long-term cash inflows for traditional rate of return or net present value calculations. On the other hand, as we have seen, there is equal risk in innovation for its own sake. Some excellent ideas fall by the wayside when solutions are developed before clearly defining the problem. Workplace transformation is notoriously unsuccessful if it imposes new ways of working on unwilling subjects. If one end of the industry is hampered by risk aversion, and the other by overenthusiastic adoption of new technology regardless of its relevance, where can genuinely new concepts take hold and flourish?

There is one sector whose great strength is its imagination. Some of the most exciting innovation in the blockchain space is coming from the gaming industry. This is hardly a surprise: gaming has been at the root of other major shifts we have already experienced in the technological landscape. User interfaces such as touch screens and virtual, augmented and mixed reality were adopted in gaming before getting a foothold in the mainstream of enterprise. As one of the steppingstones leading to the development of the internet and World Wide Web, the multiplayer Empire game on the University of Illinois PLATO system is recalled fondly by Eugene Jarecki² who played it as a child growing up in the 1970s. The explosion of artificial intelligence in the 21st century owes much to gamers’ demands for faster gameplay and more detailed graphics. Hardware stepped up to keep pace with the needs of a lucrative industry, and the new generation of graphics processing units (GPUs) produced for gaming also enabled the complex data analysis and algorithmic learning at the heart of artificial intelligence.

Gaming has inspired the development of concepts that could not have been otherwise realized without blockchain, and new behaviors are emerging which will have an impact upon the future of our everyday lives, at home and at work. Asset ownership has taken on a life of its own by using nonfungible tokens (NFTs) and their unique, indivisible properties to represent individual items. Where industry has been exploring the use of NFTs in asset twinning, gaming has taken the concept and run with it to introduce decentralized ownership and lay the foundations of a new and vibrant economy. At the same time, developers have had to address the challenge of capacity for the volume of transactions processed in a game and the amount of data that can be physically recorded on a blockchain. Games are remarkably complex when you break down the rules. Few of us realize how much processing we carry out internally when deciding on our next move. For effective automated gameplay there are hard decisions to be made about the value of each transaction and whether it is necessary to record everything on-chain, and the smart contracts governing each move must be highly efficient, costing gamers as little as possible in transaction fees and taking the minimum time to verify.

Gaming innovations are moving rapidly and have implications for business models and commercial behavior in the future. Let’s look at them in more detail.

A Gaming Mindset

Ownership in a game is deceptive. When you have walked miles to catch or hatch a rare Pokémon, earned or paid good money to buy special outfits for characters in anything from Angry Birds to Fortnite, or mined the materials to construct your diamond castle, it feels as if you own something. You can compare Minecraft worlds with your friends and strut with pride in a multiplayer environment, but in reality ‘your’ assets remain firmly on the servers of the gaming company, or static in your licensed copy of the game. As Deckbound’s Gareth Jenkins explained at the 2018 Blockchain Game Summit³, gamers are ready for blockchain: it fits their mindset. They have an expectation of ownership of the assets they create and win during gameplay, including the ability to buy and sell items. Marketplaces already exist for the exchange of items in many role-playing games (RPGs), but the assets themselves do not belong to the players who are trading them. Shifting to a state where there is individual ownership feels like a natural step and opens new doors for gamers. Blockchain provides the medium in which this can be achieved.

This was part of the impetus behind long established gaming giant Atari’s quest to create a token. Speaking to Atari Token cofounder Daniel Doll-Steinberg, we discussed the virtual world theory underpinning this and other developments in blockchain. Value in the real world is created through selling your skills, your time or your assets, and similar value could be created in the virtual world if games have the mechanisms in place to exercise the same economic and monetary principles. The skills honed through hours on a console are starting to deliver transferable real-world value at the highest level. We are seeing the rise of e-sports with the English e-soccer competition, the ePremier League (ePL)⁴, and among the new generation of Formula 1 racing drivers, 19 year old Lando Norris ascribed some of his success and his ability to read other drivers on the track to his years of sim racing⁵ and participation in the McLaren Shadow Esports competition. Even Fortnite has stepped into the arena with its 2019 World Cup Finals in New York streamed to millions of fans. However, deriving value directly within a gaming platform from skills, time or investment, rather than in a discrete competition, is an entirely new opportunity. If this can be harnessed—and blockchain offers the means to achieve it—the subsequent creation of value will kick-start a sustainable, democratized, worldwide economy. This is the disruptive vision of gaming which drives development in small, agile companies and which Atari has recognized.

The development of the Atari token has been a much slower and less public process than other more well-known blockchain game projects. As a NASDAQ listed company, Atari could not jump into the ICO space with the same agility as smaller enterprises and start-up companies, but their vision led them to plan for a token sale under complex NASDAQ regulations. Doll-Steinberg reflected on the pressures of developing a structure for the project where the founders were comfortable with the innovation environment while also meeting regulatory requirements. He recalls the frustration of watching some teams with little or no business experience raising money and failing fast while Atari Token’s highly skilled founders were taking the long road. The more agile independent companies were liberal with public information about their plans, but Atari was bound by listed company restrictions on releasing anything which could affect their share price or breach stringent insider trading laws. The press picked up on the story thanks to a few lines in a mandatory report for the Paris Bourse in February 2018, but by this time the 2017 bull market had collapsed and the ICO gold rush was declining rapidly. Plans for an ICO were abandoned in the summer of 2018 but Atari have continued to invest internally.

The bulk of development in blockchain gaming has been focused on small companies. This is what we have come to expect with disruptive innovation. As the internet started to gain traction, existing market leaders failed to catch the wave, and Google from the garage and Facebook from the dorm overtook them. In gaming, the larger companies don’t yet have an incentive to change their strategy. Their business models are profitable and sustainable, and they are in the same position as large enterprises across all sectors who are not ready or able to experiment and risk affecting their bottom line. When the tipping point comes, though, Atari at least will be ready to react.

What disruption is emerging right now from innovators unfettered by regulatory requirements? Value creation from ownership of digital assets has been the first source to be realized, and rewards for time invested and skills developed are not far behind. Let’s now turn our attention to digital cats and see why they are going to change the world.

Cryptokitties and Nonfungible Tokens

If cat videos are the emblem of the modern World Wide Web, then digital cats are the flagbearers for the future of blockchain and cryptocurrency. When Cryptokitties⁶, the first high-profile blockchain game, launched in November 2017 it immediately grabbed the attention of the blockchain community. Such a large volume of transactions was carried out on the Cryptokitties marketplace in the first few days after launch that the Ethereum network struggled to cope with the load⁷. This excitement was not due entirely to the fact that, at last, here was a fun outlet for your cryptocurrency. Sharp eyed developers reading the open source code realized that once you purchased your cartoon kitty, you had absolute ownership of this cute feline asset. Closer inspection confirmed that when you breed two kitties together you own the offspring, and that every kitty has a unique combination of traits inherited from its parents and defined by the smart contract that executes the breeding transaction. Every kitty’s genetic code is recorded as a discrete digital asset using a nonfungible token (NFT). We touched on programmable money in Chapter 2: let’s recap.

ERC 20, ERC 721, and ERC 1155 Tokens

A token is not simply proprietary cash but can be a measurement of value, a nonmonetary asset, or a degree of influence in a community. Most commercial applications based on the Ethereum Virtual Machine and incorporating their own native currency are using what is known as an ERC20 token. ERC stands for Ethereum Request for Comment and refers to the relevant discussion in the open source community which led to the development of each type of token. It is clear from the numbering that many proposals are discussed, but only a handful are viable and go on to be adopted by the developer community. ERC20 is digital cash: these tokens are often known as alt coins and they behave like money. All the tokens are the same and are divisible, much like a dollar bill. Thousands of ERC20 tokens have been created. Many have been used commercially for crowdfunding through ICOs, some retaining and gaining value as their related applications gain traction, and others worthless. Such tokens are often referred to as “shitcoins”, and a valueless holding is known among investors as “dust”. The majority of blockchain based games use Ether (ETH) directly for in-game purchases and actions, although a few have their own ERC20 currency built into reward and player retention mechanisms.

Gaming has harnessed two other ERC tokens with special properties. The Cryptokitties game and others like it have adopted ERC721 as their token of choice. ERC721 is a “one of a kind” token. They are still freely exchangeable, but each token’s individual relative value will vary according to its specific characteristics, its uniqueness and rareness. In technical terminology, this means that the token is not “fungible”. Fungibility is a property of currency whereby, for instance, any dollar bill can be used to buy a drink worth one dollar or less, with change making up the consistent exchange value. By contrast a nonfungible token (NFT) represents a unique, discrete item, a collectible which has its own value based on its own properties and the perception of the buyer.

It is quite simple to visualize an NFT in the context of Cryptokitties. Although all the cats have the same basic cartoon outline, every cat is slightly different thanks to the genetic traits ascribed to them. Some people find certain traits attractive and will assign subjective values to a cat with purple paws, or another driving a go-kart. They are also indivisible: you can’t slice them up, as that would be cruel.

NFTs represent a groundbreaking concept which is transferring to other sectors outside the world of online gaming. What assets in the real world are unique and could be represented accurately and immutably in the parallel digital world? This is something that is being investigated by both the private and public sectors and will drive changes in provenance, investment, property ownership, asset twinning, and other applications.

The ERC1155 standard, finalized in June 2019, combines both fungible and nonfungible token types under a single banner. Championed by developers Enjin and supported by the wider Ethereum developer community, the token allows a single smart contract to manage different classes of asset within a game. As Enjin’s Witek Radomski explains, in a fantasy role-playing game the fungible type is “useful for stackable items that don’t need to be differentiated, like a bundle of arrows for a bow” whereas “a pet dragon… could be an NFT and have a unique name of its own, a power level, and a rich history in the game.”⁸

Now that assets of all types can be owned, we start to see value creation from investment. Mechanisms in a wide variety of games allow collectors to enhance their kitties and improve their resale value, train their pet dragons for higher value and greater game success, or even rent out virtual land to other players. The value that players ascribe to assets is another aspect of the unique gaming mindset: investment in the game.

Paying for Play

It is immediately striking that gamers in the blockchain space are willing to pay to play. Every token must be purchased, or transaction fees paid to generate new tokens within game play (for example, when breeding new Cryptokitties). In a world where we have become used to getting things for free, this may seem a counterintuitive development, but the free model is not sustainable for economic success in the long term, as many platforms and publishing houses are belatedly discovering. Bringing a monetary culture back into the mainstream has positive implications both for rewarding creators and for the sustainability of an alternative economy. Crucially, it shifts the focus from spending purely for gaming enjoyment (paying for upgrades in a standard game, for instance) to potentially earning a return on the investment of time and money made in playing.

Cryptogames entrepreneur Gabby Dizon of Altitude Games defined three characteristics of gamers in the blockchain space when we spoke in May 2019.

Collectors are most interested in buying and holding some of the rarer NFTs in a game. They are the equivalent of real-world hobbyists, enjoying their collection and adding and divesting assets for pleasure as time goes on. They may well profit from their activity in the long term, but making money is not their primary motive.

Investors have a return firmly in mind. They want to make a profit from the portfolio of assets that they buy, or at the very least make enough to break even on their investment and then play without any further risk as fiat to crypto exchange values fluctuate. The return could come from straightforward buying and selling or equally from mechanisms in the game which generate income from the assets held. In Cryptokitties, owners can hire out their kitties for breeding and earn siring fees. In Axie Infinity, land parcels can be rented out, and there are similar gameplay constructs in most emerging blockchain games.

Players will include people in both of these other groups but cover a much wider spectrum, from high-school kids through to seasoned gamers who simply play for the fun of it. Any earnings from their investment in the game come as a bonus. These are the people who are more likely to start creating value through their gaming skillset and the time they spend on a platform. As e-sports have started gaining traction and value, so we can expect the top players of fantasy games and RPGs to have an opportunity to develop online careers. The implications for the future of work are staggering.

It is vital to the emerging game economy that there is a balance between these different types of users. When a game launches, the investors and collectors are the first to engage, but it is the regular play and volume of activity among the community of gamers that determines the success of the platform and the returns to be made.

Cryptokitties, as the first game to hit the headlines, is fundamentally a collecting game. The smart contracts that govern the genetic code in generation after generation of cats reveal rare traits and incentivize owners to continue breeding and buying to increase the value of their portfolio. The game suits the collectors and investors who seek long- and short-term returns on their ETH purchases—and there is money to be made. Shortly after launch, one kitty sold for 253 ETH, and in September 2018 another was purchased for 600 ETH,⁹ although these figures are unusually high.

The static collectables format is unlikely to hold the attention of hardened gamers in the long term. Hard on the heels of simple structures like Cryptokitties and Gods Unchained we start to see games coming through which combine the ownership of NFTs with active gameplay, for example Axie Infinity, which features collectable characters, battles between teams, breeding of new Axies, and land ownership and management, or the groundbreaking Neon District. Bringing such complexity onto the blockchain requires and inspires even greater innovation in terms of ease of access, scaling, transaction processing, and overall user experience. We will address these challenges later in this chapter: for now, let’s look at the final piece of the jigsaw for successful game development.

The Importance of Community

Social media started to change our communication habits from around 2008 onward, and enterprises in multiple sectors have homed in on community engagement as a key marketing tool. Community management is a new but important discipline for holding the attention and loyalty of consumers. Advertising practice has shifted over the decades along a continuum from “telling”—one-way communication on billboards, in print and on broadcast media—through interaction between the enterprise and the customer, to “listening,” which encompasses everything from review sites to semantic and sentiment analysis of social media posts. Good community practice takes this one step further, creating a networked structure with layers of engagement, which holds the interest of consumers or employees and makes them feel a valued part of the whole. Achieving a truly mature community made up of diverse individuals is often difficult. The Community Roundtable’s maturity model¹⁰ defines the expected development stages of a successful community and is based on a large body of research on engagement and management strategies stretching back to the earliest days of corporate intranets.

According to the Community Roundtable’s annual research, members of a community range from a minority of eager contributors to a silent majority of listeners and lurkers. On average, one in 10 members of a community will actually be engaged in the day-to-day discussions, and the holy grail of community management is that the activity of the network holds the attention of the core while drawing the lurkers into the discussion. It’s a tough job in the noisy world we inhabit with so many different platforms and social groups competing for our attention. My own daily round includes Facebook feeds and groups, Twitter, LinkedIn, and Instagram, multiple Slack channels, WhatsApp chats, Telegram groups, and Discord, and I cannot follow all the discussions. Asking people to engage in yet another medium can be met with everything from disinterest to hostility. The bottom line, though, is that communities form most naturally around a common interest and a need. Gaming has a head start on enterprise, with mutually supportive groups clustering around games from the earliest days of standing together in arcades through to the Steem, Reddit, and Discord groups that exist online today.

Community is an essential aspect of blockchain and cryptocurrency adoption. Where an enterprise chooses to use ICO crowdfunding, the first community that must be built is that of investors and collectors who have the wherewithal to support fundraising for development. They must have ready money, trust, and patience. Once the funding round is complete, the established community needs to evolve in order to attract and engage potential users who will generate anticipation and excitement for the launch. This is true for all enterprises regardless of sector: the engagement effort does not stand still once funding is in place. Once launched, new users of any decentralized application will find themselves in a new and potentially complex environment where there is no central authority to pick up the pieces in case of a mistake, and the support of their peers is vital. When this need for community support is combined with the gaming culture of cooperation around the interpretation of complex rules, there is a great resource for gaming companies to listen, react, and refine their offering.

Gabby Dizon reinforces the impact this has. “Creating a game economy from scratch needs a strong player community,” he told me. Not only is community important within the context of a single game, but also for building engagement with the players of other games. Nothing develops in a vacuum, and interoperability and collaboration are vital in building interest, excitement, and ultimately engagement with each new game as it launches. Dizon and other leaders in blockchain gaming have a vision of meta-events across platforms, leveraging the power of community to enhance, not compete. This is something which is enabled for the first time by sovereign ownership of assets through the mechanism of NFTs.

Nifty Moves

Outright ownership of game assets using nonfungible tokens is a groundbreaking concept because it represents something that could not be done without blockchain technology. At first sight it seems to be a simple nice-to-have feature which can take buying and selling activity outside the game platform, but what else could you do with your NFTs? Gamers may wish to take the asset and use it in another game: to be fair, a cute cat in a two-dimensional cartoon may not be useful on a Call of Duty battlefield, but how about a fully tooled-up Overwatch character?

The way an NFT is represented across applications differs according to the environment, but ownership is undisputed. A token could be listed in an inventory application, used as a basis for bonuses and rewards, or even act as collateral for borrowing. Coming back to the economic and monetary principles of the virtual world, here we have the mechanism for value creation from skills, time, and assets. Tokens can be programmed in a way that physical assets cannot, and managing interoperability opens more doors. NFTs have some very nifty moves.

Interoperability of Nonfungible Tokens

Cryptokitties creators Dapper Labs took the first step into the unknown through their 2019 partnership with a different game, Gods Unchained. Each Cryptokitty token generated a limited-edition card pack in Gods Unchained, and fancy Gods Unchained cats appeared in Cryptokitties.¹¹ The interoperability of tokens in a collaborative ecosystem is yet another feature peculiar to blockchain technology. These cartoon cats are leading the way toward trustless co-ordination and cooperation between proprietary systems across multiple sectors, not simply in gaming. Rather than using APIs to bring assets in from different sources, NFTs could move between systems independently, subject to formalities such as intellectual property rights on design elements. What other assets could be represented, recorded, exchanged, or licensed in a decentralized structure? We have already seen in Chapter 4 that work is underway toward establishing a persistent identity for assets and equipment in capital projects, and some companies are exploring the use of tokens to represent each unit, unique by serial number or physical characteristics. The aspect of using digital assets as a financial security also merits closer scrutiny: they have intrinsic value and ownership is clear. This is a new way of thinking and may represent a behavioral shift to come.

Intellectual Property and Asset Ownership

One of the considerations for enterprises using NFTs in their application is protecting their intellectual property. Cryptokitties founders spent time working through the pros and cons of NFT ownership, and eventually published the Nifty License.¹² This recognizes the company’s ownership of the generic cartoon style of the cryptokitties but grants to the owner of each token broad rights to use the cartoon with that token’s unique markings and features. The Nifty License is a way of linking the physical and the virtual: when challenged, the owner of the digital asset can authenticate their rights to use the related imagery. Released for use by any developer, this license is a blueprint for collaborative creators who don’t want to lock down their work but still need to retain some control and set up a path to monetization.

It seems logical that the principles behind the Nifty License and the ability to encode the metadata of a unique asset into an NFT could be applied to sectors outside gaming. In music, artist Imogen Heap has developed Mycelia for Music, a research and development hub for music makers. Part of the Mycelia mission is to ensure that all artists are “paid and acknowledged fully” for their part in any work. This is especially relevant to the music industry where tracks are commonly sampled and incorporated into other works. Most such activity is above board, but lawsuits for music plagiarism have been part of the landscape for decades. The Mycelia Creative Passport was launched in 2019, and the Grammys blog¹³ reported on Imogen Heap’s plan to protect music makers by giving them “proper credit, digitally.” However, researchers from the University of Ulster, in a report commissioned by the UK’s Intellectual Property Office, sound a note of caution.¹⁴ They acknowledge that “blockchain looks like a partial solution to ease the strictures of the royalty framework” but recognize the complexities of the industry and the wide range of activities in the music value chain. The report identifies a number of projects at varying stages of development and adoption who are all tackling different aspects of the industry’s challenges. These include include the aforementioned Mycelia; the Ujo project from ConsenSys; the Musicoin¹⁵ streaming platform; dotBlockchain, a partner of Intel, whose mission is to eradicate losses from rightsholder information as file types change; and Custos, whose copyright protection system and Interplanetary File ­System (IPFS) are attracting considerable interest for musical content tracking and attribution. As in our global commercial supply chains, there are many pieces of the jigsaw coming together.

In those supply chains, too, there are plenty of uses for an NFT. The Durian fruit producers we met in Chapter 4 are identifying each unique fruit and encoding its metadata in an NFT for tracking all the way to the consumer. California-based Centrifuge are addressing a different problem in the supply chain using the same methodology, allowing financial business documents to be created as NFTs. Ultimately, the technology that arrived in the mainstream as a digital cat is now representing a wide range of assets across the world.

Scaling the Blockchain

In every game there are clear rules governing each move that a player makes. Consider straightforward board games such as Othello, or checkers, or Connect4, before even thinking of the variations in chess or go. Every move involves knowing the way that pieces are permitted to act and thinking ahead and calculating the potential outcomes. When you consider the complexity of the reasoning involved, you see why games have been the training ground of artificial intelligence from Deep Blue to Deep Mind.

Once a move has been made, by human or AI, we can write this action to a blockchain so that it can be authenticated by other parties. But could we go further? It has already been established that smart contracts on a distributed ledger enable the execution of transactions. Coins are transferred from one wallet to another, registry entries are created for shares and property, life events are written to identity records, and a Cryptokitty with a unique ancestry-derived genetic code can be created. Surely a move in a game is a simple smart contract? It seems logical that blockchain can become the referee, checking that the move made is legal and the outcome correct.

Of course, there is a catch. Take a simple game like Connect4. There is only one possible move: a colored disk is dropped into the frame as two players attempt to make an unbroken line of four in any direction. What is required to confirm that the move is valid and to check whether a player has won on that move? The smart contract would need to establish:

1. Is there exactly one more disk in the game than at the end of the previous move?
2. In the additional disk of a different color to the one previously played?
3. Has a line of four been formed horizontally, and / or:
4. Has a line of four been formed vertically, and / or:
5. Has a line of four been formed diagonally?

To verify this simple move there are several functions, each with a cost. As the complexity of a game grows, the gas fees for smart contract processing would be eye-watering, to say nothing of the energy burden of data processing, the wait for validation, and the overwhelming noise of all these entries rendered immutably on the blockchain.

Addressing the challenge of scaling is critical in both gaming and wider enterprise blockchain and essential for bringing the speed and throughput of cryptocurrency transactions to the standard of existing centralized payment processors. There are numerous solutions emerging across many different protocols, all at varying stages of development. These complement wider work in both technology and software design, which is needed to bring blockchain and cryptocurrency into the mainstream at large scale.

Scalability and Speed

When blockchain was first introduced as the technology underpinning digital cash, the ultimate vision was of a global decentralized payment system. Thanks to the development of electronic payment processing by banks, the Visa system, and private mechanisms such as PayPal, the first thing most people think of when they consider the processing of a high volume of transactions are the twin metrics of speed and throughput. How fast does the blockchain propagate, that is, how quickly is each new block created, and how many transactions are processed on each block? One measure of this is how many transactions per second (TPS) can be achieved.

In terms of throughput, global mainstream processor Visa reports that in 2018 it processed 124.3 billion transactions on its networks for 3.3 billion card carrying consumers.¹⁶ This is approaching 4,000 TPS at a vast scale. PayPal, whose 267 million active users made 10 billion transactions in the same period,¹⁷ comes out at approximately 317 TPS. By contrast, Bitcoin trundles along at a sedentary seven. Successive cryptocurrencies have bettered this, and although a few have reported throughput close to Visa levels this has not been tested at the scale and vast capacity of established centralized systems. Ethereum reportedly reaches up to 30 TPS and Litecoin closer to 50 TPS. The fastest established crypto payment processor is Ripple with around 1,500 TPS for the XRP coin, although as this is achieved through the Ripple protocol consensus algorithm and not on a public blockchain it may not be a valid comparison. The all-time high for EOS, according to the EOS Network monitor, is 3996 TPS, but day-to-day live transactions run at a fraction of this figure, showing under 100 TPS when accessed in August 2019. EOS has been designed for high scalability in a deliberate effort to stand shoulder to shoulder with Visa, but the structure of the EOS blockchain and its delegated Proof of Stake consensus mechanism result in a more centralized structure than traditional cryptocurrencies.

How can transactions per second be increased? In basic terms, this involves either increasing the number of transactions that can be completed each time a new block is generated, or the speed with which the blocks propagate, or both. Of course, it is not so simple. In Chapter 2 we addressed the introduction of Segregated Witness (SegWit) to the Bitcoin blockchain, which changed the structure of the block contents, and the disputes over block size, which led to successive forks of the chain into Bitcoin Cash and its derivatives.

Transaction speed, or latency, is a different side of the scalability equation. According to the Kraken cryptocurrency exchange, the actual time taken to move a coin from one wallet to another varies across a broad range from the near-instant Ripple and Stellar to a full hour for a Bitcoin payment. A significant part of the speed equation is the consensus mechanism used by each blockchain. Proof of Work is pure but slow. Any attempt to speed this up, for instance by making the mining algorithm easier, compromises the security of the blockchain. The Ripple protocol consensus algorithm, the EOS delegated Proof of Stake consensus, and Stellar’s federated Byzantine agreement system (FBA) are all examples of effective, fast mechanisms, but there are concerns over centralization where validator numbers are limited. This suggests that there is a trade-off between true decentralization and transaction throughput.

How will the speed and volume of processing grow? New blockchains are being developed with speed and scalability in mind. There are plenty of teams waiting in the wings and reporting test processing at hundreds of thousands of transactions per second. The market is skeptical of many of these claims simply because showing fast processing on a test net and a limited network of servers is very different to transactions in the real world. However, there are some exciting projects underway. Among these is Algorand, launched in July 2019, which has attracted a lot of attention with its platform claiming a throughput of 1000 TPS and a five-second latency at a test volume of half a million transactions. It uses a new Byzantine Agreement protocol for consensus by committee with verifiable random functions to select the user who will propagate the next block as the chain grows.¹⁸

Sharding

The Ethereum network is considering a concept known as sharding. This is used in traditional database management where large data repositories can be partitioned for faster processing. It is used for huge databases where users would otherwise be waiting a long time to retrieve their information from one source: multiple data shards speed up the process. Search engines, large e-commerce sites, and CRM systems all use sharding techniques. The parallels with the needs of blockchain are clear. The Ethereum sharding project aims to split validation workloads among different groups of miners, breaking the whole blockchain into shards or microchains carrying only a fraction of the transaction data. Since the nodes would no longer need to process each and every transaction, this can theoretically increase the speed and throughput of the network. There are complexities to be addressed with sharding, not least in security, ensuring that a particular shard cannot be compromised. However, randomizing the allocation of each shard and changing nodes regularly is one solution.

Ethereum is not the only protocol looking at sharding. It is also of interest to Hyperledger, and some new blockchain projects are adopting sharding structures from the outset. As the concept has been proven in centralized databases, it is only a matter of time before the additional considerations for implementation in a decentralized setting are resolved. However, for the existing blockchain ecosystem, simple linear scaling is not going to put Bitcoin on a par with Visa, and sharding may not be the answer to every question.

Let the Side Chain Take the Strain

Bitcoin, Ethereum, and other truly decentralized blockchains run on what is known as a “gossip” protocol. This means that every transaction is broadcast to every node. This is one of the great strengths of the technology but as transaction volume grows, there are no easy economies of scale to be found. Developers are working on solutions that sit alongside the established system without compromising its structure. One of the most straightforward approaches to scaling has been the development of side chains as a second layer on the established protocols.

Side chains were first proposed in a 2014 paper by Back et al.,¹⁹ which introduced the concept of “pegged sidechains.” It suggests the “transfer [of] an asset from the (original) parent chain to a sidechain, possibly onward to another sidechain, and eventually back to the parent chain, preserving the original asset.” These side chains, the authors continue, would be independent blockchains and as such could host innovative development and experimentation. It is a concept that has been adopted and taken forward successfully by several developer groups. Let’s look at the innovations around the Lightning Network on Bitcoin and the gamer’s favorite, the Loom Network.

Lightning Network

The first generation of live side chain projects began with the Lightning Network on the Bitcoin protocol. In their 2016 paper,²⁰ authors Joseph Poon and Thaddeus Dryja proposed “scalable off-chain instant payments” for the Bitcoin blockchain. The authors recognized that as every change in state of the Bitcoin ledger is broadcast across the network, this could never be scaled to compete with the centralized payment systems currently servicing the world’s commerce. They proposed that transactions could be processed in separate channels with only the final change of state being broadcast. This decongests the network and reduces the costs to the user as fees are only payable to miners when the change of state transaction is processed on the main blockchain.

A good example of how this works in real life is simple day-to-day micropayments. Want to buy your coffee in Bitcoin? Waiting for validation and confirmation from the network means a long line and cold coffee. Instead, such micropayments can be managed through a Lightning Network app. You set up a small balance in Bitcoin, much like a prepay debit card, and buy your coffee using this balance. At any point, either when you no longer need this one-to-one link to your vendor, or at a point in time such as the end of a month, the change of state, not the individual transactions which led to it, can be written back to the blockchain. In accounting terms, which are a good analogy, the side chain holds individual journal entries for each coffee purchase in a ledger account, and the Bitcoin blockchain shows only the opening and closing balances of that account at a point in time.

Lighting Network developers have addressed a number of security concerns, including how to manage disputes about the account balance on a side chain without compromising the main blockchain record. Lightning has inspired a wave of side chain innovation.

Plasma

Plasma is a development layer on Ethereum with a similar goal to sharding, but a structure not dissimilar to that of Lightning. It enables the creation of “child” chains with their own distinct properties, which use the main Ethereum blockchain as the trusted source. Special smart contracts on Ethereum known as “root contracts” contain the rules that govern the behavior of each child chain. Plasma is still in development as a standalone solution, but it has been adopted to reinforce asset security in the Loom Network.

Loom Network

How do you deliver a decentralized gaming experience without slowing to a crawl? The answer, according to the Loom Network, is to develop application-specific side chains.

Loom’s software development kit (SDK) was designed to help non-blockchain developers to build the blockchain element of their systems in a straightforward fashion. Loom’s team is also behind Crypto Zombies, the definitive smart contract coding tutorial for those who want to dig deeper into development skills. The SDK allows the construction of a “DAppChain” sitting as a Layer 2 on top of a number of existing blockchain protocols in the same way that the Lightning Network is a Layer 2 on Bitcoin. The team behind Loom are aiming for it to become the universal Layer 2 connecting all major current and future blockchains. This is a significant step toward the seamless collaboration and interoperability that we have enjoyed for decades with the mature internet, and Loom integrations are in place for Ethereum, EOS, Tron, and Cosmos among others.

Increasingly popular with developers of games, Loom is also used for social networks and other applications that require speed and high levels of user interaction. The default SDK uses Delegated Proof of Stake consensus for all the individual side chains, although developers could choose other mechanisms, and this means that transactions on the side chain incur no gas fees and have the fast confirmation time that a Proof of Work blockchain lacks. Compare this with applications native to Ethereum, such as Cryptokitties, which incur fees for every transaction and take time to validate. Launching a game on a side chain enables the use of highly complex smart contracts for gameplay. The first DAppChain went live in March 2018 and Loom is used by several growing games including Axie Infinity, Battle Racers, and Neon District.

Loom has adopted Plasma as a method for users to transfer ERC20 and ERC721 tokens onto sidechains with the knowledge that should there be any problems with the sidechain, such as a hack or fraudulent activity, that their assets are still recorded on the main Ethereum blockchain. This bolsters the security of side chains and makes them a realistic and reliable platform for scalable development.

Practical Considerations for Scaling Blockchain Projects

Speed is not by any means the only measure of a successful system. There are plenty of other factors to consider, not least a loaded question asked by Blockchain Game Alliance’s Alex Amsel: Do you need to scale if you don’t have users? As we have seen before, having the “best” technology does not guarantee that that market will follow. Competitors on slower frameworks could easily win greater market share, and there is nothing to be gained by future proofing an application which has no users.

If the system features any type of interoperability, it should remain in the same blockchain ecosystem as its collaborators: this is one of the strengths of Loom in widening collaboration networks. Software development kits (SDKs) vary in their thoroughness across frameworks, and the more support that is available for technical teams the greater the integrity of the eventual platform. Anyone seeking a new, faster option needs to look carefully at the team behind the development, doing the due diligence that is required with any early stage business.

Add Just a Pinch of Blockchain

Software developers using blockchain can meet all these scalability efforts halfway by reducing on-chain processing. Amsel takes the view that if something does not have value (whether monetary, commercial, or personal) then it does not need to be on a blockchain. This is a good rule of thumb to follow in developing sustainable applications.

Take an objective look at the processes that are being automated. Be honest about what elements of the user journey really require an immutable, distributed record to be created. There will be considerations beyond simple functionality and authentication. Businesses often rely upon valuable data as a commodity, therefore broadcasting even metadata to a network can compromise the business model. An example of the way blockchain has been used to good effect for this type of business is the Gospel Data Platform discussed in Chapter 4: the introduction of a distributed ledger has been carefully managed to solve a problem while protecting the data which participating businesses hold and value. If there is a genuine business reason to incorporate complex smart contracts into the software, then the next step is to consider the needs of both the users and the developers.

Focus on the User

It is all too easy as a software developer to lose sight of the user and focus instead on the technology, pushing to have the cleanest processing, the fastest validation, the best experience from the very particular point of view of a developer. Users are not generally concerned what is running under the hood. They want the software to execute the functions for which it is designed through a straightforward user interface with suitable security and resilience. In the past, when running a growing software business, I learned to spot the innocent statement “just refactoring something…” This normally flagged up a developer falling too far down the rabbit hole into a wonderland of new tools, and future proofing far beyond the roadmapped scope or needs of the product. Designing for the expected growth in users according to a realistic business model is important; chasing the fastest blockchain experience at the expense of a good user experience and proven resilience is foolhardy.

Marguerite deCourcelle, CEO of Blockade Games, is clear that blockchain developers need the complementary expertise of application developers to achieve good user experience. She points out that, “when Epic Games made Fortnite, they simultaneously developed the Unreal Engine to power it. They made development choices in real time based on the requirements for the gaming application.” It is the high-quality user experience that drives adoption and access to a mainstream user base. Blockade, the company behind Neon District, released their blockchain game developer platform in September 2019 as a tool for the industry to achieve this collaboration. It turns the process on its head: established game developers can use the platform to build in blockchain features which will enhance their application. The goal is to make it so that players don’t realize they are playing a blockchain game. This frictionless adoption is something that every industry should strive to achieve.