The concept of systems thinking has always been fascinating to me. While growing up, I started noticing how rigid every system appeared to be – it seemed like every system (e.g. school, politics) didn’t have enough flexibility to really make a difference. There is a proverb that has always stuck with me: “If you give a man a fish, he eats for a day. If you teach a man to fish, he eats for a lifetime”.

I think this little proverb encapsulates really well the essence of systems thinking. After all, what is systems thinking? Donella Meadows is my favorite author from the field and this post will be very much inspired by her posthumous book (2008) entitled “Thinking in Systems: A Primer”. On that book, she wrote the following:

A system is a set of things (…) interconnected in such a way that they produce their own pattern or behavior over time.

This simple definition sets the tone for this article. If the generated pattern is not sustainable over time, the system is flawed by nature. Therefore, for a system thinker, it is crucial to think about long-term consequences that come up as a result of the relationship between the systems’ elements. Moreover, a well-designed system should have internal mechanisms to deal with external (and internal) forces.

Currently, our society is made up of systems embedded into other systems embedded into other systems and so on, making the overall system extremely rigid. This is one of the main problems in the field - and in society itself - and one of the reasons why cryptoeconomics is so important to change this paradigm.

At its essence, any system is made up of three things: its elements, the interconnections between them and a purpose (or function). Let’s take a closer look at how systems really work:

Elements, interconnections and purpose

The elements of a system are the easiest part to notice. They are usually tangible things, clearly identifiable by an external entity. If you consider, for example, a football team, the elements of that team are the players, the coach, medical staff, the president, etc. Usually, changing an element won’t cause profound impacts on a system – unless that element is too important in such a way that it changes the purpose of the entire system, or it changes the interconnections between other elements in a meaningful way.

While the elements of a system are usually easy to see, the interconnections between them might be subtler. Following the same analogy, it is not easy to understand all relationships between all the players and the coach and other staff members. It’s quite complex actually - that’s why we call them complex systems! Even more importantly, it’s harder to tell how much impact does a certain relationship might have for the general well-being of the system. We live in a nonlinear world, which makes interconnections hard to understand. Only through experimentation we can see what works and what does not work. A change in the interconnections (e.g. a bad relationship between two elements becomes good) may provoke a change in the overall system behavior.

The purpose of the system is the goal that it aims to achieve. The best way to understand the system’s purpose is to watch for a while to see how the system behaves. More important than the actions themselves, is the intentions behind them. Let’s go back to the football team – let’s imagine it lost the last 20 games. Can we accurately say that the purpose of our team was to lose? Not really – we would have to take into consideration the team’s behavior during those 20 games. Did they run in every game? Were they improving? Were they just bad players (elements) who gave everything but still managed to get bad results? The key takeaway is this – purposes are inferred from behavior, not from stated goals. The behavior of a system is its performance over time (i.e. its growth, stagnation, decline, randomness or evolution – basically, its patterns). System behavior reveals itself as a series of events that happen over a certain period of time.

Here’s two great quotes from Meadows regarding the system’s purpose and behavior dichotomy:

If a government proclaims its interest in protecting the environment but allocates little money or effort toward that goal, environmental protection is not, in fact, the government’s purpose.

Systems thinking goes back and forth constantly between structure (diagrams of stocks, flows, and feedback) and behavior (time graphs).

A holistic approach to systems, considering its structure and patterns over time, is the best way to design, model and simulate a sustainable system.

Well-designed systems

When systems work well, there is a harmony in each sub-section of the system. It’s beautiful to see – structure and behavior are perfectly aligned. The systems’ elements and their interconnections serve the system’s initial purpose. For a system to work well, these four characteristics must be, somehow, present: resilience, self-organization, hierarchy and security. Let’s break them down step-by-step:

  • Resilience is the system’s ability to persist in a mutable environment. A resilient system is a flexible system, able to adapt to different scenarios and to recover from interferences to its regular behavior.

  • Self-organization is the ability of a system to make its own structure more complex. For a quick analogy with the blockchain world, quantum computers are eventually coming and can be potentially harmful to cryptocurrencies. However, a self-organized system would be able to come up with solutions (e.g. zk-STARKs) against this sort of threats.

  • Hierarchy is best defined in the dictionary: “a system in which people or things are arranged according to their importance”. This might be a tricky one. A good hierarchical system is designed in such a way that makes the system more modular. Just like in software engineering, modularity can be highly beneficial and increase specialization. However, hierarchies are prone to suffer from abuse of power from the higher layers, which may compromise the whole integrity of the system. This is why decentralized governance is so important – to reduce the problems and inefficiencies caused by malfunctioning hierarchies.

  • Regarding security, good mechanism design requires building systems that resist collusion even by the elements running their own system.

Moreover, an important function common to almost every system is to ensure that the system itself is sustainable and preserves over time.

If the behavior of a system is consistent in the long-term, it’s probably because that system has a mechanism creating that persistent behavior. In these cases, we can say that the mechanism operates through a feedback loop. When that behavior is considered to be positive, stable and aligns with the system’s purpose, we call it a balancing feedback loop – if a system gets to this point – without compromising on resilience, self-organization, hierarchy and security – I would consider it to be a well-designed sustainable system.

Mechanism design and visual modeling

Mechanism design is a fascinating field because you start with a pre-defined goal (e.g. protect the Ethereum ecosystem against malicious actors), and then you create the incentives and penalties (e.g. PoS) to make sure that your goal is achieved. The main difficulty with mechanism design is creating the conditions that incentivize the system’s elements to behave in a certain way – we can see this, for example, in the difficulty of creating incentives to build public goods and increase the work in R&D.

Studies of complex systems are designed to explore what would happen in that system if a number of driving factors unfold in certain ways – therefore, it is encouraged for all system/mechanism designers to start their work with a spreadsheet in order to explore possible alternatives, and ask “what-if” questions (while iterating over time). Asking the right questions is critical to predict the model’s behavior.

Systems thinkers use graphs of system behavior to understand trends over time, rather than focusing attention on individual events. Moreover, they use behavior-over-time graphs to learn whether the system is approaching a goal or a limit, and if so, how quickly. The pattern of the graph is important, and so are the inflexion points. This is why tools like cadCAD fit in so well with mechanism design and should be more and more used when designing new mechanisms for the web3.

Conclusion

As we’ve seen throughout the article, every system thinker or mechanism designer should be concerned in building sustainable and defensive systems that can prosper in the longer-term and that can cope well with outside forces (e.g. collusion) – we live in the real world, a place where agents can cooperate, take unpredictable behaviors and, therefore, good system design requires contemplating all sorts of possibilities.

I believe systems thinking will be an increasingly important field throughout the next decade. With web3 and the metaverse emerging, this is the perfect time to experiment and build new systems. Every mechanism designer should take into consideration the fundamental characteristics that make up a good, sustainable system.

The new paradigm of web3 is developing itself at a fast speed and innovation doesn’t slow down in the crypto area – decentralized finance (DeFi) already has many useful protocols and banks are becoming increasingly obsolete; NFTs are here to stay and to empower artists and creative people; crypto play-to-earn games are becoming increasingly more important (e.g. Dark Forest was one of the most funded projects in the Gitcoin GR11 and the amount of web3 games being developed right now is… insane).

The only thing missing in the puzzle is decentralized governance, which requires moving beyond coin voting - coin voting is not credible-neutral. Currently, there are some limitations, but luckily we have two very important tools that can (and will) be extremely important in the development of new solutions and mechanisms:

  • Cryptography (especially zero-knowledge proofs): Cryptography will play a massive role in upcoming DeGov solutions. I believe that ZKPs are an extremely powerful technology, and they are as big as blockchains. I also believe ZKPs are very underrated at the moment and zk-technology will be essential for mechanism designers. ZKPs combine the benefits of privacy within a system, while making sure that people follow the rules of that same system (e.g. MACI, CLR.fund, anti-spam solutions (!), voting systems with Aragon, etc);

  • cadCAD: cadCAD is an open-source Python library that assists in the process of designing, testing and validating complex systems through modeling and simulation. There are many interesting use cases already happening, including this open-source Ethereum economic model simulator (more detailed explanation here). In my opinion, it is an amazing tool that really fits in with the whole “systems thinking” approach mentioned throughout this post. In this new context, with blockchain technology, which is so multidisciplinary and has so many new variables, I believe tools like this will be extremely valuable.

A combination of the right tools with the right incentives (and coordination mechanisms) will allow us to overcome our current limitations in cryptoeconomics regarding decentralized governance and move towards a self-sustainable system that will empower all of us. Mechanisms like RPGF are a great starting point, but I’m confident we will be able to come up with more and more ideas to fund public goods and other types of important assets.

Cryptocurrency is all about experimentation, and we have the right tools to design, build, iterate, validate and expand on possible solutions – which may very well be hybrid solutions combining both financial and non-financial elements. It’s the decade for mechanism design, and with the development of zk-technology - combined with the help of cadCAD - I am very confident in the mechanisms that we are going to be able to come up with towards DeGov! It’s all about incentives, security and coordination.

More than ever, being multidisciplinary is extremely important. In the first place, knowing how to code should be mandatory in all schools – it is as important as reading or writing. Secondly, the best DeGov solutions will probably combine elements of finance and non-finance – therefore, a good understanding of finance and DeFi protocols is also relevant. Game Theory is another field critical to understand (and create) strategies and incentives. Additionally, a good understanding of zk-technology (especially non-interactive ZKPs) will also enter the equation. At last, if you learn cadCAD, you will be able to design, validate, optimize and understand new protocols and mechanisms. Knowing the basics of each area is a great starting point for having the full vision of where we currently stand and where we can go from here.

The intersection points of multiple fields are the points that will make a difference and allow us to go beyond coin voting governance in a sustainable way. Once we find these points, we will be taking important steps towards creating a decentralized credible-neutral governance system. The future is bright, and I expect to see - in the next decade - a new system that will, eventually, replace our current form of governance and reinvent capitalism and democracy – all of this will happen while having fun in the process, as we like to do in the Ethereum ecosystem.