What are the biggest challenges facing the adoption of cryptocurrencies? I think the biggest problem facing crypto today is governance.
I used to think the biggest problem was scalability. The usual data point here is that the Visa payment network can handle up to 56,000 transactions per second at peak load.1 The two largest cryptocurrencies by market cap, Bitcoin and Ethereum, max out under 10 transactions per second by comparison. We can investigate what parameters of each system’s implementation leads to those numbers but the more productive path to take right now is to realize that each of these cryptos is an early-stage experimental technology that still has major outstanding research questions that need to be addressed before either can compare to Visa and its ilk. An important point to keep in mind is that blockchains are not databases2 and the “currencies”3 they let us model in the underlying software are not very much like traditional fiat money at all. My take on it is that it is still too early to say what exactly these technologies actually represent simply because it is still so early and the types of applications they enable are that ground-breaking.
Taking a look at major efforts in the field, it does seem that most developers are focused on scalability. A common marketing point of some of the more exotic cryptos is usually their transactions per second. We can also see the ongoing debate around how to scale Bitcoin with SegWit (and the various forks) and the ongoing work with Casper in Ethereum (along with the sharding research) as indicators that scalability is a major concern. Given this state of affairs, I’m confident that scalability has a good chance of being solved eventually and as crypto-enthusiasts we should ask if there is anything else that may hinder widespread adoption.
Another common concern you will hear when discussing this topic refers to the volatility of various cryptocurrencies. The neighborhood corner store can’t price their offerings in BTC if the price is subject to swings of 15% or more day-to-day. I agree that volatility is an issue for the use case of crypto as a medium of exchange (and secondarily as a store of value) and it would be helpful to have a “stablecoin” that is not subject to such large swings in price. There does seem to be a hedge from the scalability camp in that the expected volatility of a given crypto should decrease as the transaction volume increases; in the limit, the more people using your currency, the more stable the price. All things considered, scalability improvements simply give you the option for more users, so you could imagine something like 10Mb blocks that are mostly empty. Assuming we only had increases in scalability, existing cryptos would still need to execute a successful marketing campaign to garner bigger adoption. There are a few “stablecoin” projects in the space and I think this work is a pretty important complement to scalability efforts. These projects constitute a very useful set of experiments to figure out the best way to achieve stability in a way that resonates with the market.
The more I talk to other people working in this space however, the more I keep hearing the same sentence over and over and over again: “yeah… governance is hard”. I think we can all agree that governance is hard. Humans have been trying to find good governance structures for, like, ever. If you view governance as simply processes for organizing groups of humans, you see how widespread and also how crucial good answers to governance problems are. Given how important governance is to the human process, I see a lack of compelling answers here as the biggest risk to growing adoption of a cryptosystem. Moreover, I think the true innovation of crypto (and these pesky blockchains) is trustless organization.4 Emulating currency is actually a fairly trivial application of being able to coordinate asynchronously across space and time in a permissionless way. We all hope to avoid repeating past mistakes as we learn and grow and it would be a shame if our community did not utilize such a powerful tool in its own journey.
The reason a lack of explicit governance structure is particularly dangerous can be summarized as a lack of consensus. Proof-of-work algorithms are great for generating on-chain consensus amongst decentralized parties. The problem arises when a particular system has to answer questions of meta-consensus, or equivalently how do we come to consensus off-chain? How do we agree on, and then safely implement changes to how we come to consensus? There is no standard answer to this question and consequently, projects are left to invent their own extra-protocol protocols. This fact has immediate consequences on (expected) protocol utility as inefficiencies with governance lead to slower meta-consensus which holds back the rate of experimentation. We still have no clue how to (as a society) operate computers, much less deal with the fallout of global exchange infrastructure which implies that the most valuable action to take in this space is rapid, prolific experimentation. Another major concern lack of governance introduces is an increased attack surface which lowers the cryptoeconomic security of a given protocol. While more theoretical than the first concern, if you can embroil your project’s leadership with some dispute, it stands to reason that important actors on the network (whether on-chain or not) are partially (if not completely) compromised.
We can again look to the top two cryptos to see how governance has played out so far. In the case of Bitcoin, we see a “civil war” between various parties who have different ideas over how to best scale the Bitcoin chain. I would expect some form of algorithmic governance would help prevent the events that have led to this situation. Note that I’m not saying different factions within a crypto project should not be able to disagree; if we view blockchains as a nascent technology that can and will evolve as we keep working with them, then the best mechanism we currently have for their evolution is the hard fork. If you closely analyze the underlying issue with the impending Bitcoin debate, you’ll see it comes down to defining (and thus protecting) the brand. We can also analyze the operation of the Ethereum project in this framework. Here, the situation seems much better as evidenced by the Byzantium hard fork (and general development efforts) that seems to have been successfully pulled off. What Ethereum has that Bitcoin lacks is a single founding member like Vitalik who is still actively guiding the project, keeping a unified vision and consequently a unified brand. While this speaks volumes to the leadership abilities of the Ethereum core team, I do not think we want our industry’s answer to governance require a single leader in the long term, even if they are a “benevolent dictator.” We would hope to keep with the spirit of decentralization to find solutions to governance that are not so centralized. This present state of Ethereum is very inspiring in terms of what we can hope to achieve with respect to governance. The story doesn’t end there, however, as the long tail of crypto projects (i.e. the rest of them) have not been as successful in navigating governance challenges across the board.
What can we do then? When all you have is the hammer of decentralization, everything tends to look like a nail. One possible answer here is to bring as much of the meta-consensus process on-chain. I only know of one project, Tezos, who is actively exploring this answer5 although if you follow the media they have apparently been dealing with their own internal governance issues even before launching the mainnet. Their approach lets users vote on protocol upgrade proposals and assuming they launch the mainnet may turn out to be a very interesting adventure in cryptogovernance.
It is critical we all think more deeply about how governance plays a role in the teams and projects we are involved with. These challenges facing our community will need to be tackled head on as we continue to push blockchain adoption. The potential here is too great to lose time battling over short-term concerns at the cost of longer-term success.
- I hear the average operating capacity of Visa is more like 20,000 transactions per second. [return]
- blockchains are collections of cryptographic proofs; it doesn’t cost millions of dollars to insert a Gb of data into Google Spanner. [return]
- tokens are not money! 2017’s ICO craze has led the horse to a small puddle of water when an oasis is just around the next hill… [return]
- decentralized consensus is really just a synonym for trustless organization [return]
- and I would love to hear of any others! [return]