tl;dr It’s early in the game for crypto, and an idea maze helps us dig deeper and discern between compelling adoption paths and dead ends. Bitcoin then Ethereum catalyzed excitement for “crypto” broadly, and the industry is debating the merits of “sound money,” “web3” and “open finance.” While each share the same ethos of disintermediation, they require different features to succeed and are competing against different alternatives. Inheriting not only the ethos of decentralization but also the technical tradeoffs of its predecessors has resulted in an underlying cognitive dissonance. The way to resolve the dissonance is to forget the familiar narratives, and instead to examine: the novel features of the technology, who the core customers are, who is failing to serve them now and why there can be a compelling new product today. Let’s enter the crypto idea maze.
Table of Contents
- The Idea Maze
- Before the Maze: Bitcoin then Smart Contracts
- Enter the Maze: The Three Theses
- Cognitive Dissonance
- The Way Out: Customers, Alternatives and Intermediate Utility
- Waiting for Catalysts: How Badly Do They Need You?
- The Timing of It All
- Assessing What’s Next
- Sound Money: In Code We Trust
- Web3: UX vs Trust
- Open Finance: Bearer vs Registered Instruments
- Exploring Assumptions and Rebuttals
- How Can We Contribute Today?
- Right Values + Uncertainty = Opportunity
The Idea Maze
The idea maze metaphor stems from the fact that a technology trend has various adoption paths that lead to varying levels of success (h/t Balaji Srinivasan & Chris Dixon).
Good startup ideas are well developed, multi-year plans that contemplate many possible paths according to how the world changes.
In an exuberant time, it’s often tempting to run straight for the entrance to take the lead, but chances of success will be much greater for teams that have a rough map that guides them. Being a student of the technology, its history as well as learnings from other mazes will help us map out the maze. It will give teams more clarity in pursuing or avoiding certain target customers, product features, and to market strategies.
A good founder is capable of anticipating which turns lead to treasure and which lead to certain death. A bad founder is just running to the entrance of (say) the “movies/music/filesharing/P2P” maze or the “photosharing” maze without any sense for the history of the industry, the players in the maze, the casualties of the past, and the technologies that are likely to move walls and change assumptions.
For example, a sports fan and an experienced coach are both watching a game and see the same inputs, but they notice vastly different levels of detail. While a sports fan can be excited about what’s happening, the good coach can see how to navigate the opposing team, how certain changes of the roster or formation could be effective, and reliably affect the course of the game. Technology and adoption seem no different.
This feels especially salient for crypto because it is multidisciplinary and requires cooperation among many stakeholders across industries and geographies. Also, mapping the idea maze seems especially more important early in the lifecycle of a platform ( e.g. web, social media, mobile). Because vision and ethos are currently leading ahead of adoption, it is not clear from this vantage point which approaches are headed towards a market with strong demand and sustainable economics vs a dead end ( i.e. early).
With that, I’d like to share the beginnings of my crypto idea maze (open source all the things!). We will focus on challenges in reconciling core assumptions of various crypto theses, and we will save specific adoption paths (go to market + geo) and deep-dives on applications for another time. As such, this is not the “full maze,” but I hope you’ll reach out to continue constructing the maze together.
Before the Maze: Bitcoin then Smart Contracts
In the beginning**, there was Bitcoin.
The initial intent for Bitcoin is very clear.
A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution.
It was an experiment in creating a payment network that removed the need for mediation, in order to (1) reduce the fees of digital commerce and (2) remove the need for sharing more identity information than required.
Completely non-reversible transactions are not really possible, since financial institutions cannot avoid mediating disputes. The cost of mediation increases transaction costs, limiting the minimum practical transaction size and cutting off the possibility for small casual transactions, and there is a broader cost in the loss of ability to make non-reversible payments for non-reversible services. With the possibility of reversal, the need for trust spreads . Merchants must be wary of their customers, hassling them for more information than they would otherwise need. A certain percentage of fraud is accepted as unavoidable. These costs and payment uncertainties can be avoided in person by using physical currency, but no mechanism exists to make payments over a communications channel without a trusted party.
Bitcoin was created to provide a way to transact without identity or a trusted intermediary. It began as the currency of choice for technologists & hackers, libertarians, black market users, cross-border remittances, and the oppressed ( i.e. Venezuela). This was especially timely given the government bail out of banks that took on excess risk.
Prominent features that came from having no trusted intermediary were censorship resistance (very expensive to change or stop a transaction) and immutable monetary policy (can’t “print” more resources unilaterally). In order to do so, it uses a peer-to-peer architecture and a novel economic incentive mechanism (proof of work).
It is important to note that the innovation explicitly traded off efficiency in order to provide censorship resistance and flexible membership set of participants (anyone can join or leave the network at any time). We will come back to this, but in the meanwhile something else captured the attention of the community.
*** To be fair, Bitcoin was preceded by Digicash (Chaum 1989) and Bit Gold (Szabo 1998).*
Then there were smart contract platforms.
A smart contract is a computerized transaction protocol that executes the terms of a contract. The general objectives of smart contract design are to satisfy common contractual conditions (such as payment terms, liens, confidentiality, and even enforcement), minimize exceptions both malicious and accidental, and minimize the need for trusted intermediaries. Related economic goals include lowering fraud loss, arbitration and enforcement costs, and other transaction costs.
The idea of smart contracts have been around for quite some time (coined and popularized by Nick Szabo in 1994, quoted above), and the demand to implement smart contract capability had a resurgence through the Bitcoin community (RSK, Colored Coins, Mastercoin). One person working on Mastercoin was Vitalik Buterin, who thought the idea of extending the Bitcoin protocol was exciting yet a limited implementation of the full potential of smart contracts. His ideas diverged enough from the priorities of the Mastercoin team that it wasn’t being implemented in Mastercoin, and Ethereum was born.
While Bitcoin achieved consensus over the state of a transaction ledger ( i.e. accounting), this branch of the movement began experimenting with consensus on the state of a computer replicated in a distributed network around the world ( i.e. distributed virtual machine).
It’s worth noting that, while cloud computing focused on getting distributed computers to do things quickly and efficiently to serve demanding applications within a trusted setting ( i.e. datacenter controlled by one party), decentralized computing focuses on allowing computers that don’t necessarily trust each other to agree on a shared state of a computer. While adding computers to a cloud infrastructure makes the network more computationally productive (in varying degrees depending on technical architecture), decentralized infrastructure is instead optimized to maintain a certain level of “trustlessness” (or threshold for bad actors). Therefore, adding computers to a decentralized network–while adding complexity–doesn’t add much, if any, computational throughput of the network.
Despite those limitations, Ethereum was attractive for many reasons but in particular: (1) it was now very easy to create and deploy uncensorable smart contracts into production (the awareness around Bitcoin mining certainly was an added catalyst for building a lively early mining community) and (2) people could now imagine, “if it’s possible to transact without intermediaries, what other interactions could we also manage without intermediaries?” The simple analogy evolved from a digital “vending machine” that is owned by one party (and could thus be taken down, i.e. Paypal) to one that is not owned by anyone, is unstoppable and still accessible to anyone in the world with an Internet connection.
As Bitcoin was timely for the zeitgeist of the financial crisis, Ethereum was timely for consumers losing trust in various intermediaries, namely banks (1, 2) and Internet giants (1, 2). This distributed computer concept generated excitement because the architecture has deep implications on possibilities to rearchitect those trust-required intermediary industries. Some representative early ideas/examples were decentralized Uber and Airbnb (“web3”) as well as decentralized platforms for lending, derivatives and startup investing (“open finance”).
We now have experiments that replace the simple transaction ledger in Bitcoin with various forms of distributed computation ( e.g. Ethereum, EOS, Dfinity, Tezos, Stellar). This sets the stage for the Crypto Idea Maze (2018 vintage).
Enter the Maze: The Three Theses
We are in a fascinating part of the crypto idea maze in 2018; we have the proliferation of smart contract platforms, excitement around web3 & open finance, and the Bitcoin & privacy coin communities coalescing around the sound money narrative.
So what are the crypto theses?
Sound money – “Trustless money” that cannot be inflated by any trusted authority such as a central bank.
Web3 – “Trustless internet” where Internet architecture is free of trusted centralized data & service monopolies. Users have more control over their data and Internet usage. These networks also compensate participants for economic value generated in the network.
Open finance – “Trustless financial systems” that extend cryptocurrency to provide open software primitives for equities, debt, derivatives, checking accounts, remittances, work contracts, retirement accounts, property etc .
(Open to feedback).
As such, the term “crypto” means different things to different people, and yet we all refer to this one word as a catchall (sorry, cryptographers! :\ ). This has been a boon for a budding interdisciplinary community but also much miscommunication among blind men describing an elephant.
While they all share the same ethos of disintermediation (respectively of government banks, big Internet giants, and commercial banks/fintechs), we will find that there is significant cognitive dissonance in having a shared set of technical tradeoffs across implementing disparate theses with different needs.
Cognitive dissonance – The state of having inconsistent thoughts, beliefs, or attitudes.
Over time, many have come to realize that it is not easy to reconcile winning forms of sound money, web3 and open finance under one technology. This is likely due to divergent underlying assumptions of the theses.
Problem: Adding a feature on top of a platform optimized to remove said feature.
Cause: Inheriting tradeoffs designed for another need.
Specifically, we are building identity and “censorship abilities” on top of a platform optimized for not having identity and censorship . The potential problem here is: intuitively, that’s far from resembling the most efficient architecture (this is an uncontroversial statement, even to core devs of these technologies).
At a more fundamental level, this dissonance exists because many smart contract public blockchains inherit tradeoffs made by Bitcoin: the network architecture, replication scheme, incentive mechanisms. Those tradeoffs absorbed many orders of magnitude of replication inefficiency and “SLAs” ( e.g. throughput, availability, predictable pricing) to make it difficult to (1) not require identity and (2) remove the need for a central entity that can censor.
Now the question becomes why should things with disparate user end goals ( i.e. web3 / open finance) share the infrastructure tradeoffs of another product need ( i.e. sound money). There still may be compelling reasons why (as even one change in assumption can quickly change the outcome), but it’s definitely not a gimme.
A Means to an End
For the use case of money, those were excellent new features. No one can change the rules of the game on you, doesn’t matter who you are, where you are, the value you have won’t be diluted by unforeseen increase in money supply, is unstoppable by a single party and doesn’t require to know who you are.
However those features are not inherently valuable; they are valuable due to serving a specific user need. In fact, those features are good or bad depending on what service one is trying to provide. For example, the downside of introducing identity is discrimination, but the upside is reputation. Analogously, the downside of having central entity is censorship/liability, but the upside is quality service and convenience.
While the various theses share technological primitives and ethos of disintermediation, we will explore what this means in the context of the diverging needs of their end users. This is the crux of the maze today.
The Way Out: Customers, Alternatives and Intermediate Utility
As such, there’s never been a higher premium for the ability to think independently. We ingest narrative after narrative on a daily basis on Twitter, and conflicting assumption sets of these narratives are exactly what made this maze more complicated (be skeptical of this post too!). The only way out of the maze is to think independently about the customer need and what features those needs will demand.
We will discuss in more detail in the section below, but let’s first take a look at the criteria in the rubric.
Who is the customer and the stakeholders in their experience?
- Who is the core customer? (define personas / markets)
- What % of the world is affected?
- What is the level of “pain” per person affected?
- How do people join the network and why do they stay in the network?
- Customer’s counterparty
- Is it a one-to-one? Is it one-to-many?
- Does the counterparty need to have the same product need as the primary customer?
- Trusted intermediary
- Who do you depend on to enforce the rules? Who could rule against you?
“Atomic unit of success”
For an ambiguous and multi-faceted product like crypto, it’s important to explicitly define the fundamental metric for customer satisfaction. For Google, it’s a completed search. For Uber, it’s a completed ride. For owning a share of Apple, it might be return on equity and earnings per share.
But for USD, what is it? For crypto networks, what is it? We will discuss below.
Next best alternative
Ethos-agnostic customers (aka “most people”) will be driven largely by a new feature’s benefit over its next best alternative .
For example, web search provided a convenient and fast alternative to seeking information from distant places. Google search provided a simpler and faster Internet search experience than Alta Vista, which was powerful and customizable but slower (perhaps optimized for the wrong feature given the early average user). An early version of Uber provided a convenient way to call a black car to a location other than your usual office and travel in style. Uber pool now provides an alternative to driving day-to-day or taking the bus/subway (in SF).
The tangible advantage over the next best alternative in serving a user need is exactly the forward momentum a product needs to improve and expand into adjacent services.
Intermediate utility refers to the usefulness of a network before it is sufficiently saturated (“if it takes over just 5% of a target market, do those 5% of users still derive utility in the meanwhile?”).
Two things help the creation of a network at an early stage: (1) the discovery of active peers despite small network size and (2) how compelling those experiences are before the network fully matures. Those are crucial factors in the “intermediate utility” of an early network. Early adoption paths with intermediate utility are often the only ways to achieve a desired end state. Contrapositive: certain desired end states are impossible/expensive to achieve because it’s difficult to have intermediate utility to bootstrap the network.
What is an example of intermediate utility? For example, Facebook was fun to use when I was in school (to communicate, signal and stalk within the school setting). I didn’t use it too heavily but enough to keep me engaged. That continued engagement of its early users for the early use case allowed Facebook to have a chance to further serve those users as well as expand into new markets. If we had to wait years (or even just months) to get that initial usefulness, we never would have given it a shot–making more successful end states out of reach.
So the fact that small portions of the broader target network can get utility from the early iterations of the technology is a huge benefit, allowing it to sustain intermediate size of networks without having to reflexively and virally grow all at once (check out Tony Sheng’s recent post on reflexivity in crypto).
Intermediate utility serves as a “checkpoint” for the network to “save progress” as it grows. More broadly, a series of step-function gains in intermediate utility will ultimately set the stage for main stream adoption / “crossing the chasm.”
Waiting for Catalysts: How Badly Do They Need You?
While the theses are justifiably motivated by ethos, customer need and external catalysts will be the largest drivers of how this idea maze unfolds. Let’s see how the criteria above map to the three crypto sub-theses.
- Value Proposition : Money that can be used to buy goods and store value, unaffected (and unprotected) by idiosyncratic government policy. For example, shields user from incremental monetary easing, poor economic policy, corruption, and any other sovereign risk (issuer risk).
- Core Innovation : “Internet money” in general made it possible to be unbound by physical realms (Digicash and Paypal still had this property, but had centralized settlement). Proof of work distributed that settlement layer, making it unstoppable by any one jurisdiction.
- Core Customer : People that don’t have access to stable store of value (e.g. Venezuela/Argentina/Zimbabwe inflation. Another example is large number of Chinese investors buying up prime U.S. real estate to “park their money.”)
- Counterparty : Person who is providing them the good/service they need (rent, food, other).
- Next best alternatives : SWIFT/ACH/Western Union and cash (transfer of wealth). Gold and USD (store of wealth).
- Provider of next best alternatives : Government-registered banks, money transmitters, independent off-shore banks (store and transfer of wealth). Central bank + enforcing government (provides the instrument and trust).
- Tradeoff : Between acceptance by peers (merchants/peers) and trust of issuer.
- Atomic unit of success : (1) Ability to pay for good/service, (2) Peace of mind around saving for future goods/services (no loss, low volatility)
- Intermediate utility : Transactions only require two parties at a time. Storing value requires everyone to be bought in passively (other people believe it), but requires only “single player mode” actively (you hold it).
- Challenge : Game theoretic deadlock in getting merchant and supply chain buy in for accepting a new currency. Lack of stability due to difficulty in matching monetary policy (supply) to predicted adoption curve (demand).