11 Mar Lane Rettig
Lane Rettig is a writer, technologist, and entrepreneur with a background in computer science and quantitative finance. Lane received bachelor’s degrees in computer science and Japanese at the University of California, Berkeley, received his MBA from The Wharton School of Business, and his MA from The Lauder Institute. He spent time developing a healthcare startup called Seratis before beginning his work on blockchain. Lane founded Crypto NYC, a blockchain-centered coworking community in New York City, and worked as a core developer for Ethereum from 2017-2019. He is currently working on Spacemesh, a next-generation blockchain platform, and is engaged in independent research and consulting in the governance of cryptoeconomic systems.
In this interview, we cover Lane’s views about the state of Ethereum’s governance, new and dynamic approaches to blockchain governance, the rise of third-gen blockchain technology, and “autonocracy”.
Conducted December 26, 2019, by Daniel Kees
How did you get interested in blockchain and blockchain governance?
It’s quite surprising to me that it didn’t happen sooner. One of the things I love so much about it and find so appealing about the space is that it’s the intersection of five or six or seven or eight distinct disciplines, depending on how you count. This is what makes it so impenetrable to a lot of people because, let’s say, you’re brilliant technologically. You’re a great programmer. That’s great, but you may not understand the legal side. Or you may not understand the social side, the regulatory side, or the economic side. It has a legal foundation. There’s economic stuff, a lot of business stuff, and obviously all the technology stuff. There’s cryptography, math, etc. These are all areas of inquiry that I’ve been deeply passionate about for a very long time.
I would call myself a generalist. I’ve done a bunch of different things in very different industries. I did not go super deep, career-wise, spending a decade in any one particular, narrow vertical. So, in retrospect, I’m kind of surprised I didn’t discover blockchain sooner.
I feel that it was in the cards when it did happen, though. I think the reason it didn’t happen sooner is that I was very heads down, very focused on a startup. I had a health care technology startup, called Seratis, that I was working on from 2013 until 2016. I founded Seratis with a classmate at Wharton over the summer between my first and second year of MBA. I was doing what founders are supposed to do, which is a mono-maniacal focus on one thing, and it was in the year after that project wrapped up when I finally came up for air. I cast a wide net and was exploring a number of different cutting-edge ideas in different spaces and trying to figure out what I wanted to focus on next when, in early 2017, I stumbled upon blockchain and Ethereum, specifically, at a conference—a technical, but not the blockchain-specific conference.
The thing that fascinated me was the concept of a smart contract. That’s less about the technology, actually. Of course, technology is deeply interesting and challenging, and it’s what I spent the most time thinking about and working on the past couple of years. But initially, it was this beautiful, Platonic idea that you can have a “contract” in the legal sense (the word “contract” is not used precisely the same way in these two disciplines) written in “dry” code—to borrow this term from Nick Szabo—interpreted by machines as opposed to human language, like legal-code. As a programmer, someone who’s been writing computer code since I was a child, that idea was very appealing to me.
I usually like to preface this by saying that, we had classes on governance and legal studies at Wharton. And in the context of corporate governance, I never found it a particularly exciting or interesting topic. It’s just not a sexy topic, to be blunt.
But what I found as I went deeper and deeper, especially in the Ethereum world, was that the technological side—as challenging as it is, and as much work as there is that needs to be done there—is in pretty good hands. There are a lot of brilliant engineers, mathematicians, cryptographers, etc. who are working very diligently on that side of things. However, for a hodgepodge of reasons, there’s almost no one who has expertise in things like governance and related skills like philosophy, conflict studies, legal studies and jurisprudence, and all these kinds of things, who are deeply engaged, at least in the core Ethereum ecosystem.
I’m not an expert by any means, but I felt drawn to work on the aspect of it which I thought was the least well-served: governance. I had at least a little bit of experience having run a startup before and having gone through business school and having worked at a big, successful company, which funnily enough, doesn’t sound like a particularly unique set of experiences. But in the blockchain world, it happens to be. So that’s what I found myself working on, governance.
How would you explain “blockchain governance” to someone who’s not familiar with that area?
Governance is very difficult to define, a very “fuzzy” idea. There are a number of different approaches to it, and Prof. Kevin Werbach’s work has helped a lot with this. When we held the original Wharton Cryptogovernance Workshop in San Francisco a few months ago, he presented an introduction. The first thing he did was to poll all the participants and ask them for a definition of governance. As is so often the case, rather than asking one person, I find it helpful to get different perspectives.
At the highest level, there are three or four angles to it. One that I like to look at is really just decision-making. There’s some funny joke that says if you put two people and an apple in a room, that’s governance. The basic requirements are people and resources. If there are no resources, there’s nothing to govern. Governance is how a group of people arrive at the process by which decisions (usually about the disposition of resources) can be made. If no decisions are being made, we could say that, undeniably, governance is not working. Again, these are things like the allocation of time and money. That’s probably a requirement as well if there are no resources in question whatsoever. So at the very minimum, we’re talking about how people spend time and money.
So, there’s decision-making and resource allocation. There are things like collective action problems, which is when you have very large groups of people and very interesting game-theoretical dilemmas, which are very real in the blockchain world. There are issues like a tragedy of the commons, which is literally a textbook example from ECON 101, and yet it’s very real and very manifest in the blockchain world today because blockchains exhibit many of the characteristics of commons (in the Ostrom sense of the word). They are rivalrous in some sense, and they are nonexclusive because we can’t prevent people from accessing blockchains like Bitcoin or Ethereum by design and yet they have limited capacity and limited throughput and that sort of thing.
There are power dynamics at work as well. How do the various parties and actors—both individuals as well as organizations that are involved in all these things like decision-making and resource allocation processes—position themselves relative to one another?
Finally, when you go a bit deeper, it has a lot to do with a frame of mind. If you have this group of people or bodies acting in their own self-interest, is it possible to align the desires and thoughts and values and principles of all those different parties to put the collective interest ahead of their own interests? Can you get them to agree on the basic ground rules? There’s actually some interesting psychology and organizational dynamics at work here, too.
Given your unique professional background, were there any “best practices” or governance themes in those spaces that you find lacking in blockchain?
There are a number of tensions in blockchain governance. The one that I’m thinking of right now is the tension between efficiency on the one hand, which tends to imply centralization, and, I’ll just call it “decentralization,” on the other hand… I was exposed to governance issues in the startup world and the tech world and, broadly speaking, the “Silicon Valley startup mindset”—because there is this culture that I think has emerged in Silicon Valley that has been reinforced by a lot of the top VC funds and incubators and accelerators, a lot of this I attribute to the “lean startup” and Paul Graham schools of thought. I’m not sure that they speak directly about the centralization-decentralization question, which is a more recent phenomenon, but it’s all highly centralized by design.
In the startup world or Wharton for that matter, I was never exposed to this concept of decentralization or to the idea that we can achieve optimal outcomes or that we can better optimize for certain desired properties through a process of decentralization. This centralization-decentralization question is, of course, one of—if not the—core “tension” in the blockchain space. In many ways, blockchain is a reaction to the centralization of Silicon Valley and tech giants and governments and all those things. In the context of governance, it’s really wonderful to have an ecosystem where everyone can have their voice be heard and participate and transact as purists in a permissionless fashion. All these nice properties we get with decentralization.
However, there’s a cost. Everything in the universe—everything we choose—is a tradeoff, and by opening one door we’re by definition closing another. In the case of choosing decentralization, what we’re giving up, of course, is efficiency. Things cost more; they take longer. I’ve seen this acutely, fascinatingly enough, both in the software layer as well as in the human, social layer. Something I’ve noticed in a lot of the teams I’ve been a part of in the blockchain space, including in the Ethereum world, is that the teams themselves operate in a very decentralized fashion, in a sort of leaderless fashion. Even if there is a leader, it tends to be a more nominal-type thing… The teams I’ve been a part of deferring a lot of the important decisions to the “will of the team” and the nominal leaders don’t want to go out on a limb and make decisions on behalf of other people, etc.—so you lose something. You lose a lot of efficiency and it takes a very, very long time to make even relatively simple decisions.
All of the startups I have ever had a hand in or been a part of being run in a very centralized fashion for a very specific reason. Startups are by definition operating in war mode all the time, and they need to be agile and make decisions quickly, pivot quickly, update products and factor in customer feedback, and respond to rapidly changing market conditions. A decentralized system can’t do that, by design.
While decentralization is nice, there’s probably a balance point, and I’m not sure that we’ve found it yet in the people and processes and teams and things that are building blockchain today.
In talking about the differences, particularly the efficiency losses in decentralization, have you been able to see ways in which those issues have been addressed (e.g., through formalized governance mechanisms or on-/off-chain solutions)?
Ethereum is like “decentralization maximalism,” a lot like Bitcoin. It may have been the case that, before my time, the Ethereum Foundation and other folks had a much more central role in directing the project, but that’s not been the case since I’ve been involved. And as a result, I think it’s led to a lot of inefficiencies in Ethereum. What we’re seeing in a lot of the “next generation” blockchain platforms and projects is a reaction to that, more towards trying to optimize for efficiency.
Polkadot—led by Gavin Wood, the Parity team, and the Web3 Foundation—is a great example. Gavin was one of the Ethereum co-founders and one of the core three or four people who had a hand in turning Ethereum into what it is. So he has as much experience as any human in the world today with building a successful blockchain platform.
Polkadot takes the exact opposite approach to governance that Ethereum does. It has a very rigid, highly structured governance system. I’m not completely up to date because, of course, there’s active research happening, and the network hasn’t launched yet. But they have, on the one hand, a council of actual humans, something on the order of 11 people or so last I saw, who have the ability not to make decisions unilaterally on behalf of the network but to accelerate the decision-making process. There are formal proposals that are written in code. This is the most important point about it and is something Polkadot shares with Tezos, another network that maximizes the efficiency of governance by encapsulating the actual changes in code itself.
Whereas, in Ethereum, when someone makes a proposal to change something, it’s kind of this fuzzy, hand-wavy thing. It’s written in English and is published as an issue on GitHub. In the case of Polkadot or Tezos, there may be some description associated with it, but basically, it’s a piece of code, which by some formal structured mechanism will or will not be incorporated into the actual running nodes in a partially or fully automated process. That is the most interesting experiment, I think, that’s happening right now.
Tezos has already completed at least one upgrade, so they’ve proven that in theory, this model works. I don’t think Polkadot has launched its main network yet, but I’m very closely watching it. There are many exciting things about Polkadot, but I think seeing how governance plays out will be really interesting.
Decred is a very interesting project which has taken a bit of a middle road. It has a system of voting so people who hold tokens can purchase these things called “tickets” and use them to submit or vote on proposals. But they do not have that automatic upgrade trigger mechanism of Tezos and Polkadot, so there’s still human intervention and a social layer there. You have Polkadot and Tezos at one extreme and Bitcoin and Ethereum at the opposite extreme, where there’s no formalized processes whatsoever or very minimal formalized processes. Decred is sort of a middle path.
The really short answer is that it’s very, very early days, and there’re a lot of very fascinating experiments that will play out over the next 12 to 18 months. We all need to sit tight and see how these things work and how well they work.
There’s a super short thought process I have about evaluating the quality of governance. Governance is not one-size-fits-all. There are probably some properties of governance that we could debate about and most of us would agree are highly desirable in almost all cases. We probably want a system of governance that is representative of—and encapsulates the needs of—the majority, or ideally the entirety, of the stakeholders of a system. That’s just one property of a governance system that is likely desirable in almost all cases, although even then I would say it’s never 100%.
It’s very subjective. This is actually one of the areas I’ve been doing research on and been publishing a little bit about in addition to some other values we may think are desirable in most governance systems: transparency, accountability, representativeness.
So successful governance can look different in different contexts. Have you seen any novel approaches or trends in the so-called “third-generation” blockchain projects?
There are a lot of projects that are in various stages of the research-development launch pipeline. And for that reason, even governance aside, the next 12 to 18 months are going to be really, really fascinating to watch. There are dozens of networks such that, even if only 10% to 20% of the networks that have said they’re planning to launch during this time period actually end up launching, it will still completely change the landscape (and even if we only have two or three successful launches, though I think we’ll see more than that). Some projects are quite close, to be clear.
On the one hand, we have the “on-chain governance maximal projects.” Tezos is the most obvious, and it is actually live and running. Polkadot is the next most obvious example. They are certainly optimizing for a slightly different set of desired properties than Ethereum, for example, which is totally fine. One of the funny things about the way the blockchain space is playing out is that technology is becoming increasingly commoditized. Ideas like “sharding,” for example, which Ethereum pioneered in the research sense, will be part of the upcoming Ethereum 2.0 launch. But that’s still quite a ways away, and other chains have actually launched with it already. Polkadot has their version of this, called parachains.
Things like governance, values, and “constitution” (i.e., what a community stands for, what are its foundational principles) are actually the social things that I think projects are going to differentiate themselves upon, more than the technology.
If you look at the history of technology, whether it’s airplanes or air conditioning or any of these things I was exposed to at Wharton, it all goes through the same process: there’s an early flood of ideas, then eventually standards emerge. This process is going to play out in blockchain. We’re going to see technological standards emerge, and projects will differentiate themselves more along the lines of things like governance. For that reason, I think these experiments are all really, really important.
So while I personally am not a fan of on-chain governance—because it basically reduces to plutocracy, and I don’t think that that’s the kind of system that I personally want to be a part of or transact on—it has its upsides, like (potentially) efficiency. It’s an important experiment to watch, and we’ll all learn a lot from it. There’s probably an opposite side of that spectrum, which would be minimalist governance, and that’s like Bitcoin’s or Ethereum’s style.
Spacemesh, the primary project I’m contributing to right now, is focused on shipping code and getting our public testnet up and running over the next few weeks, and so governance is something that I and many others on the team are thinking a lot about. But you could make the case that it’s less important until the code has been shipped, and there’s actually something to govern. So we’re not sure yet. But if I have my say, Spacemesh will steer far away from the “on-chain plutocratic voting” direction.
The most interesting experiments that are happening in the next generation blockchain platforms are projects like NEAR Protocol. They’re good friends, and I’ve had an ongoing dialogue about governance with them for a while. Unlike many projects, they recognize the importance of governance and have invested quite a bit in getting it right for a long time. I think NEAR sees both the inefficiency issues that are plaguing Ethereum and how it’s becoming harder and harder to upgrade the network or get anything done for lack of formalized processes. On the other hand, they also see the risk of maximizing plutocracy, as can happen with on-chain voting.
So I think the form of governance we’ll see emerging in a platform like NEAR is going to resemble more closely some of the best practices we have in the off-chain world. Things like boards of directors tend to work reasonably well for corporations. And we have legislative bodies and multi-house structures for national governments that try to encapsulate and represent the needs of very diverse groups of stakeholders, which blockchains will ultimately need to be able to do as well. I know that the NEAR team is thinking along these lines, and it will be very exciting to see some really novel structures emerge that, rather than throwing the baby out with the bathwater, don’t completely jettison all the best practices of the off-chain/default world but actually borrow some of them and update them to work in the on-chain world. That’s what I’m the most excited to see.
I’m curious if you think there is a role for formalizing corporate governance concepts like the fiduciary duty. I know that’s something a few people have written about in this space. Or, do you think that these blockchain-powered communities can develop these things internally and execute them on their own?
I think it’s definitely the single most important question on the legal side, but to me, it’s actually among the topmost interesting and exciting questions in blockchain and in the world in general. It relates to the so-called “Szabo’s Law,” a concept coined by Ethereum’s Vlad Zamfir (named in honor of Nick Szabo, though the articulation is Vlad’s). Vlad has thought a great deal about blockchain governance, arguably more than anyone else that I know of.
Szabo’s Law says that the only change which should ever be made to a blockchain protocol is a technical change that is required by the network, which implies that there is no social governance layer whatsoever. For a platform community network that follows or subscribes to Szabo’s Law as articulated by Vlad, blockchain itself is not something that can be negotiated with or reckoned with. It’s not a legal person or state actor. To borrow another phrase from Nick Grossman at Union Square Ventures in New York (from this podcast): It’s like air. Bitcoin is air. Bitcoin is in the atmosphere, and it’s not controlled by anyone. It’s the most decentralized thing in the world.
Basically, a given blockchain will never try to be legal. If it subscribes to Szabo’s Law, it is a jurisdiction in and of its own right. It doesn’t care about the United States or New York or any particular place. It spreads, and there’s nothing you can do to control it.
The reason that Vlad articulated it in this way is that he very strongly disagrees with this approach. He says, “No, no. Blockchains need to be legal.” The reality is blockchain exists in a social context. There is a set of human beings who give rise to it, who come together, who build this thing and maintain it, etc.. Those humans themselves live in particular jurisdictions and are subject to particular rules. If we don’t make any attempt whatsoever to reconcile the “crypto-legal regime” (as he calls it) with the existing legal regime, then blockchain will never work because either government will shut it down or people will get thrown in jail or worse things will happen. I’m massively reducing Vlad’s argument, but we don’t have time to go further into it here. I urge people to look into this concept of Szabo’s Law, something I’m thinking and researching and writing about myself.
Tying it back to your fiduciary duty question, Angela Walch is a great example of another legal scholar who has done important work on this. Coin Center has done really foundational work on this as well. If you choose Szabo’s Law, you say, “Ok, we’re Bitcoin. We’re extralegal or ‘alegal.’ We have no desire to become a legal entity in a meaningful way in any jurisdiction.” Then it’s a moot point because such a system knows nothing and cares nothing about fiduciary duty or any of those duties.
On the flip side, if you believe that we need to legalize blockchain in some fashion, in particular organizations and people working on code in a particular context, then I think you have a very interesting question. We need to begin looking at a lot of the risks faced by the core developers of these platforms. That’s something else that Angela Walch has talked about and that I felt very acutely when I was working on Ethereum, in particular. It’s an interesting question that bears a lot more discussion.
It would be remiss of me not to ask about something that you’ve posited that I find really interesting, a framing I haven’t seen in another context. You make a distinction between “autonocrats” and “anthropocrats.” If you wouldn’t mind, would you briefly explain what those 2 words mean and their relevance in the governance context?
In the broadest terms, we have autonocrats and anthropocrats. An autonocrat is a person who wants to install a maximum agency in autonomous processes and decision-making agents. Blockchain is a very obvious example of this, but it applies in general to AI as well. As I’ve participated over the past year in a few initiatives to work on standardizing “crypto-law” or blockchain governance and looking at best practices, I’ve been very inspired multiple times by work that’s been done in places like the EU with respect to regulation around AI systems. It is actually an avenue of inquiry that has gotten quite a bit of attention over the past few years. There’s a lot we can learn from that space.
This is happening. As Andreas M. Antonopoulos likes to say, Bitcoin can’t be uninvented. A lot of legal scholars are pretty terrified of things like DAOs and autonomous blockchain systems. I think the most dystopian version of this is assassination markets—the idea that a bounty could be placed on a platform like Bitcoin or Ethereum to the tune of God knows how much money. It gets Black Mirror-esque very quickly. We call Ethereum (and for that matter, Bitcoin and a lot of other platforms) “unstoppable networks of unstoppable applications”—and that’s by design.
The argument in favor of “autonocracy” is “the march of technology” rationale. Technology is getting better and better, and we want to embrace it. It can do a lot to improve the lives of many people and increase efficiency and decrease the costs of spinning up organizations and running experiments and things like that. For that reason, we should be unafraid to embrace this technology. Every decision that we hand to a piece of code or autonomous system is a decision that we are freeing ourselves from having to make. A good example would be autopilot on an airplane. It frees the pilots to focus on other more important things, operating “at the top of their license.” They still handle takeoff and landing for a reason, because those are the hardest things that airplanes have to do.
On the other side, we have the anthropocrats. That’s a person who wants to minimize the amount of agency that human beings give up to autonomous systems. Again, this includes blockchains, AI, flight control systems, anything like that. The argument in favor of “anthropocracy” as opposed to the autonocrat point of view is that autonomous systems get quite dystopian very quickly. Skynet [from the Terminator franchise] is not far away. We are not sophisticated enough; we don’t understand these systems well enough. In particular, we don’t understand the most sophisticated machine learning systems. There’s a really famous line: “We shape our tools, and thereafter our tools shape us.”
We build these autonomous systems, and the risk is that those systems then shape our world. And I think there’s no question that will happen. It’s just about making sure that we do this in the right way and that we have the right escape hatches and we don’t sacrifice our agency in such a way that we can’t reclaim it later. This is a very deep philosophical question that has probably come up many times in the history of technology. So this is my attempt to make it legible.
But I think it’s going to be one of the defining debates of our generation, and the reason I use terms that are reminiscent of politics is that I think that this will become a political question pretty quickly. There will probably be a big divide here, and it will be interesting to see how this plays out.