Monday, July 2, 2018

Brief Lessons from Being a CEO in the Cryptocurrency Space

Professionally, I've been in the space since 2013 and almost all of that experience has been in some form of a leadership role (usually the CEO or director of something). After some reflections- thanks to my many long haul flights crossing either the Atlantic or the Pacific oceans-  I've decided to write down some thoughts on lessons I've learned.

The cryptocurrency space is an unusual animal in that it's not entirely a field of technology adhering to some leadership variant of a silicon valley startup or an established research group like DARPA. Rather it's an aggregation of a political movement, disgust of current systems, academic research, frustration over middlemen of necessity, a religion with cyperpunk roots, and a series of open source projects- I probably missed a few others.

Thus, one is left with contradictory leadership demands. For example, research is a slow, methodical process often laced with regressions and many alternative threads subject to strong opinions by brilliant, but somewhat difficult, people. Whereas the political and quasi-religious elements of the space demand sweeping statements, adherence to strict principles and a more emotional side that can sometimes bend facts in strange directions.

It's utterly unsatisfying to say it depends or suggest that decentralization might not be the best solution. One has to wrap this up in prose, obfuscated arguments or some other method (from Bitcoin to Burning Man is one such example). Yet, in practice, we all seem to converge to this pragmatism when building products and protocols. It's seldom said, but always felt that a large percentage of all cryptocurrency is held by third parties at exchanges or cloud wallets.

The point of our movement has been to develop better tools to identify when and how we can remove middlemen we don't want; not to blindly kill all middlemen for the sake of killing them. Centralization or federation can dramatically reduce costs, improve efficiency and sometimes is necessary to improve privacy.

For example, Lastpass provides a great solution for me to manage my many, many password, but doesn't actually have access to my passwords. I'm sure such a solution could be decentralized, but I don't see a point. Replicating an encrypted dataset, which I could already store locally anyway doesn't improve my security. It just seems to drive up costs and reduce user experience.

And frankly there are hundreds more examples. Abstracting my point, it's not about decentralization; it's about control. People want more control over their data, identity, reputation, assets and commercial relationships. Decentralization is a tool that can be used- with other tools- to achieve these ends, but isn't an end in itself.

The political side of our movement touches a different thread that gets conflated in the discussion. There are many whom hold libertarian or anarchist philosophies. They view this technology as a way of fundamentally restructuring society. Yes control over data, money and identity being decentralized is nice, but the point is these things are stepping stones in gradually reducing the relevance and power in institutions these factions see as immoral or even dangerous.

This philosophy has nothing to do with running an open source project. This philosophy has nothing to do with building more resilient and better customer experiences for consumers of digital products and services, but it's bundled into the movement itself and the business requirements of many of the cryptocurrencies.

Therein lies the conflict we are facing in this space. CEOs are forced to either pick a side or wear masks as they travel from crowd to crowd. Either you're a decentralize everything fanatic or a blockchain advocate who's here for the technology. Or you publicly moderate with some winks and a I feel your pain smile.

I cannot see such things being sustainable long term. Protocols being biased to political ends seldom has a good outcome. It's almost a design by political committee. We've already seen the impact with Bitcoin's evolution where very sensible proposals couldn't gain traction and very sensible people were demonized, silenced or even driven out of the space.

Furthermore, this statement might surprise some, but it turns out that many of the brightest minds in the world disagree with the decentralize everything movement. They don't share libertarian or anarchistic ideology. Should we be so quick to dismiss their contributions with such zeal because they aren't in the club?

During my tenure as the CEO of IOHK, I've been fortunate enough to carve out a fairly diverse enclave of thought. Some of our employees are hardcore socialists. Some are conservatives. Some, like myself, are libertarians. But we share common goals in trying to understand the problems our space can actually solve.

A particular thread has been our quest to develop sustainable, secure and incentives reasonable proof of stake protocols. This effort has now spanned millions of dollars, a half dozen papers, peer review at Crypto, EuroCrypt and other venues as well as collaboration from people at more than ten institutions.

As one could imagine, there is a massive difference of opinion on where and how such a protocol ought to be deployed. Even the Hyperledger Fabric guys want to get a piece of the project. Yet the common goals are clear and non-philosophical. We have a series of real attacks that have been discovered by engineers, researchers and those who study incentives that need to be addressed from nothing at stake to the honest majority problem.

Questions like what incentive does one have to be in the honest majority and how should slot leader selection be decided are meta to the underlying model. They are specific to the deployment of the protocol and the realities of the users of the protocol. The team deploying the protocol has to make a choice that will ultimately determine the level of centralization and the economics of the system.

The point is that such things ought to be explicitly discussed. They shouldn't be embedded within the underlying design of the protocol and silently accepted by those who adopt it. It's perfectly ok to say we are ruled by a plutocracy if the users of the protocol accept that structure. It's not ok to say your protocol is decentralized, but then economic realities provide all but the richest and best connected from participating.

Thus my first major lesson has been to try to extract the politics and philosophy from the objectively scientific and engineering problems and then be honest about choices we have made in our particular deployments. I wish more projects and their leadership would be honest here. 500,000 transactions per second is not a revolutionary new capability. It's either an artifact of deceptive marketing or architectural choices leading to massive centralization and trust.

Second, I've learned that it's extremely important to clearly understand relationships and commitments as early as possible. The nature of our space is that we are scattered throughout the planet. Modern communication tools are incredibly good at keeping us all connected and informed, but they are useless for promoting empathy, building real relationships or proactively communicating implicit concerns.

With Ethereum, I spent a huge amount of time worrying about the legal structure and how the project was going to be lawfully funded. Only a handful of people were directly involved in this effort out of commonsense and expertise. This meant that I only interacted with a handful of the hundreds of people involved with the project on a daily basis. The rest had to just assume I was doing stuff and hope it was for the best. In the absence of communication, opinion and speculation dominated yielding the aftermath many today are all to familiar with.

With IOHK, I invested much more time in simply building relationships with the people I work with directly or indirectly. This includes traveling throughout the world to see my employees in person. Sometimes a beer and a good conversation is all that is needed to build common ground and empathy. It seems simple, but it's so powerful and undervalued.

We also clearly define the roles and relationships for each new member of the team. It's a huge overhead for HR and line managers, but it resolves tons of issues before they even exist. Yet this model does not work for open source development. One cannot ask an engineer independently fixing an issue listed on GitHub to sign an NCA, NDA and report to a line manager.

The point of open source development is to move forward with both professionals and volunteers. Many of the most successful projects are completely built and maintained by unpaid volunteers. When a traditional open source project forks, it's usually a cultural, technological or philosophical difference (Robles and Barahona even wrote a paper about it). But what happens when one introduces concepts like tokens, premines and ICOs? Incentives and relationships change.

Investors of cryptocurrencies are addicted to the prospect of strong, centralized and reliable management of the development of the protocol as it has the potential to yield the greatest returns, but this management style is at odds with the point of cryptocurrencies, which ultimately require governance by the many and in the open to be sustainable.

Furthermore, the existence of premines, ICO funds and other stores of value create a goldrush of crypto-lamprey fish seeking to attach themselves to a project to extract tokens rather than make contributions. Add some demagoguery and buzzword salad and you have your standard crypto-evangelist.

These actors bring me to my third lesson. Conflict, pain and principles are a prerequisite to success. I learned this most directly with the Ethereum Classic (ETC) project. To me, ETC is Ethereum. The original social contract was that we were proposing the world's most expensive and inefficient computer in exchange for absolute certainty that the code deployed wouldn't be modified due to outcome on some group whether it be a company or government.

Suddenly the DAO hack occurs and the social contract was changed to well it will run as written except this time because it's just too painful to this particular group. Nevermind that group blindly gave $150 million dollars to an unaudited smart contract with no checks and balances. Nevermind the lack of consumer protection put into the offering- likely due to a crude attempt to evade regulation.

People made a poor choice. The point of cryptocurrencies is to let people make poor choices and to force them to accept the consequences of their actions. This clearly is no longer the point of Ethereum. So those that wanted to keep it that way created ETC.

During the early days, there was tons of chaos. Some of the more colorful community members even attempted to hijack ETC for some purpose or another. But yet, the community stuck to its principles and now has stabilized around a very decentralized group of leaders. No one is in control. Reasonable software is being maintained. ETC just keeps living.

The value of ETC is no longer about carrying out the mission of Ethereum. It now seems to be a sandbox to figure out how to move forward without anyone having a mandate to do so. Unlike Bitcoin, there isn't a 100 billion dollars to protect nor thousands of companies pulling in different directions. Rather it looks much more like a normal open source project.

There was a lot of conflict and pain in getting to this state, but it was achieved only by keeping with a set of principles. Maybe they are too harsh for normal people. Maybe they won't allow the protocol to grow beyond a certain critical mass. Ultimately the market will decide, yet that doesn't make ETC any less meaningful as a project to the people who use it.

This last point I feel is the most valuable our space has demonstrated. Ultimately cryptocurrencies are composed of people who agree to something. While these agreements might be entombed in code and protected by clever mathematics, they are used by people.

People decided that the original vision of Ethereum made sense to keep alive and thus we have ETC. People can just as easily abandon it as they have hundreds of failed projects. A good leader in this space has the ability to rally people to a common cause and give them the tools to decide what to do about it. A bad leader tries to inflict his personal view on the masses for their greater good.

In a way, I think this gives some justification to the enormous market capitalization we are seeing alongside the cult like enthusiasm. From a practical sense nothing has been produced by Bitcoin or any other project to justify a cumulative valuation greater than some of the world's largest companies and even countries. From a social view, how much is the freedom to re-imagine the entire fabric of the world's marketplaces worth?

Money and governments are artifacts of promises and future commitments. They don't actually exist, but the belief in their power is what actually gives them power. It seems the belief that Bitcoin is worth a 100 billion dollars or that one social contract is superior to another is all that is really needed to actually realize it.

Thus to all future leaders of our space, I'd like to say separate your objective from subjective, clarify relationships as well as understand conflicts of interest and stick to your principles even in the face of conflict. And try to have fun while riding the roller coaster, it's sure to be an exciting journey.