RChain Blog

A capital idea

by Lucius Gregory Meredith – originally posted on Medium Oct 23, 2020

When we ask ourselves “What is capital?” different proposals come forward from the community and even from within our own minds. Capital is energy. Capital is information. Capital is a set of beliefs. i want to encourage both a broader and yet deeper and more focused line of thinking. Think about voting. Think about games. Think about the expression of sentiment (likes, emojis, etc) right here on Medium or on social media platforms like Facebook or Twitter. All of these are coordination patterns for coordinating human behavior that — remarkably — share with capital a common set of features. Now, abstract these features and make them programmable and decentralized. Then you will have a public utility for coordination. This is a view of of both the origin and future of capital i want to hone in on.

As a precursor to this line of thinking, i suggest Yuval Harare’s Sapiens. He makes the key point: coordination is Homo Sapiens super power. Further, he observes that capital is one of the tools for focusing and amplifying coordination. What he doesn’t say directly, but is easy to work out is that in moving from an industrial implementation to a digital one, the physical inequities intrinsic to a physical monetary record (inherited from an architecture of centralized minting backed by armed enforcement of distribution rules and norms) were transferred from the physical to the digital and then accelerated. This acceleration arises because the number of iterations of the distribution rules (e.g., rules for lending and interest, rules for inheritance and tax, etc) and the trends these rules amplify are orders of magnitude faster in the digital incarnation. In some sense, that’s the whole point of using networked computers. The bulk of the physical baggage is gone, and so transactions can proceed at many times the speed of physical exchange of paper and coin for goods and services. However, the system supporting the digitization of the physical system programmed in all the assumptions of the physical system without a lot of questioning or rethinking. So, the tilted landscape got a lot more tilted. Essentially, the rich got richer a lot faster. This is a matter of public record.

Thus, this first iteration of digital capital broke because it accelerated toward a distribution where capital became stuck in the hands of a few. This outcome was entirely predictable in hindsight. Without a deliberate and serious commitment to maintain some critical number of coordination models reachable from the current distribution, the likelihood of this particular distribution and its inertia is a virtual certainty. Again, this is an objective assessment, if you take capital as a coordination technology. That is, if the aim is to maximize the range of possible coordination models reachable from a given distribution of capital, then distributions (and the rules that produce them) that minimize the number of coordination models likely or even possible are failure modes.

The good news is that when we acknowledge that the first iteration of digital capital is broken we can bring to bear sufficient societal pressure to make a new iteration. Without uttering words like blockchain we can see that because this is a digital revolution the new iteration must be programmable. Again, if the aim is to maximize the range of possible coordination models, then programmable capital is objectively an improvement because programmability is the quintessential means to maximize the number of coordination models. Programmability means we easily can tweak the rules producing new distributions; and programmability is the means by which we can envision and create effective design patterns that unify capital-based transactions, voting, expression of sentiment, and a host of other coordination mechanisms.

Why do we want to expand the number of coordination models? We need this expansion of coordination models because we have to coordinate globally in a way we never have before. We are in the midst of the 6th mass extinction event. By many estimates more than 70% of insects (by biomass) are gone. When you think about the lynchpin role insects play in the ecosystem, you get an inkling of how bad the situation really is. So, to deal with the situation we have brought on ourselves, we have to coordinate like we never have before.

Having established the need for programmability, let’s turn our attention to decentralization. This is where blockchain makes its appearance. It is certainly possible to generate centralized programmable capital solutions. However, this keeps in place trust assumptions that create the accelerated inequities. In short, these trust assumptions will effectively always result in the rich getting richer faster. Decentralization removes these assumptions, making it possible to explore a wider range of models. To be sure, models where the rich get rich fast are gravitational wells in the space of all coordination models, but they are supermassive black holes in centralized systems. If we want to enable hyperspace jumps toward more equitable coordination models, we have to decentralize the digital capital infrastructure. Blockchain is the best technology for this job, if it can be made to scale.

Enter RChain. RChain brings together several key ideas. First among these is that most transactions — and here we mean both data and financial transactions — are isolated. Certainly, when Alice is buying grilled tofu from a street vendor in Shanghai that transaction is almost always isolated from Roberta buying an empanada from a street vendor in Peru. It’s also the case that the edits i am making to this article are almost always isolated from edits you are making to your most recent social media post. Because these transactions are isolated they can proceed concurrently. That’s how these systems scale today. RChain achieves fine-grained concurrency because of rholang + rspace. The former (rholang) is the smart-contracting language that allows a programmable digital capital infrastructure with fine-grained concurrency built into the programming model. The latter (rspace) is the storage and execution mechanism that realizes the concurrency on modern hardware/software substrate. However, building in fine-grained concurrency that allows for all of these isolated transactions to run at the same time is a necessary but not sufficient condition to scale.

Another key component is the consensus technology. This is the cornerstone of decentralization. An economically secured, leaderless, distributed consensus algorithm is a means by which we remove the trust assumptions inherent in a centralized digital infrastructure. Essentially, it means the infrastructure becomes genuinely public. In principle, anyone can run a node in the network, just like anyone, in principle, can run a tcp server, or an http server. While Bitcoin and Ethereum’s proof-of-work is an excellent teaching device for demonstrating what an economically secured, leaderless, distributed consensus algorithm is and what it can be used for, it is very much like http 1.0. Nobody uses it today because it doesn’t scale. Its primary function, in hindsight, was to show people what you could do with the Internet, not provide a realistic implementation, at scale.

To achieve a blockchain scalable enough to realize a decentralized, programmable digital capital infrastructure sufficient to replace and reboot the incumbent system we have to replace proof-of-work. RChain does this with a version of proof-of-stake that is not only provably safe (it doesn’t get to inconsistent states), but also provably live (it doesn’t get stuck), and provably fair (everyone gets a shot at the mic). What is crucial about RChain’s version of proof-of-stake is that we move from a blockchain to a block graph (a DAG or directed acyclic graph for those in the know). It’s no longer a linear sequence of blocks. It’s a graph of blocks with blocks of transactions enjoying multiple parents and multiple children. This means that nodes can be concurrently proposing and validating blocks.

The two features, fine-grained concurrency, and block level concurrency means that the network enjoys both vertical and horizontal scaling. As we add processing power to a given node in the network, that node scales up, able to process more transactions; and, as we add more nodes to the network, the network scales up and is able to process more transactions as a whole. In recent tests we are seeing near linear scaling. This means the limiting factor on the network is the cost of deploying more hardware.

How does all this manifest in a public coordination infrastructure that has the potential to enable Homo Sapiens to coordinate its way out of the 6th mass extinction, to find its way from a destructive culture to a regenerative one? Let’s look at RChain’s recent experience with self-governance. RChain is a Washington State Cooperative with bylaws crafted from the successful REI’s cooperative bylaws. Those bylaws require an annual general election in which the members of the cooperative vote in a new board, and weigh in on any items of business. If RChain offers a genuine programmable digital capital infrastructure, then constructing an on-chain voting mechanism should be as easy today as it was to set up a web page back in the beginning of the world wide web. In fact, it’s easier.

Building on the basic features of rholang, Dan Connolly (himself one of the original developers of the WWW) and Jim Whitescarver (a pioneer of digitally amplified organizational intelligence) and many other community members built RVote2020 in a few short person hours. In fact, the hours spent in actual development was far smaller than the hours in agreeing and formalizing the requirements — which is as it should be. And this week i and the rest of the Cooperative membership had the chance to voice our will in RChain Cooperative’s annual general election using RVote2020. It’s the first time i’ve voted on-chain.

Many things can be said about RVote2020. It’s cryptographically secure. Anyone who is not a registered voter attempting to spam this system will simply lose their REV tokens. Everyone is guaranteed to see the same ballot. Registered voters can vote from anywhere in the world where they have an Internet connection. In a time where the president of the US and his attorney general are actively attempting to undermine the population’s confidence in their electoral system, and the system is under cyber attack from foreign powers, a system like RVote2020 has a certain appeal.

For me, however, the most encouraging feature of RVote2020 is that it was developed in a decentralized manner by the community. Naturally, there were drivers and key players, as there are in any human endeavor, but it was a community effort to create a tool for the expression of the community’s will. Beyond that, RVote2020 can by used by any member of the Cooperative to create a referendum, get an item of business before the board, or show community support for a proposal. In fact, there are just two inputs to RVote2020, a list of registered voters, and a ballot. That means that any community, not just RChain can use this system. For example, i am going to propose its use for the Seattle Circle in their next board election.

In summary, RChain has been and remains devoted to delivering the next iteration of capital-based coordination as a public utility. We are pursuing this goal because humanity is at a point where we must level-up our coordination super power or give way to the next contestant.