---
tags: fun
slideOptions:
center: false
# spotlight:
# enabled: true
theme: white
transition: 'fade'
# parallaxBackgroundImage: 'https://img.rawpixel.com/s3fs-private/rawpixel_images/website_content/rm21-background-teddy-100.jpg?w=800&dpr=1&fit=default&crop=default&q=65&vib=3&con=3&usm=15&bg=F4F4F3&ixlib=js-2.2.1&s=360cc82c5cdd556320d0d93d32721b62'
---

<style type="text/css" scoped>
* { text-align: left; }
.reveal h1 { font-size: 1.5em }
.slides { width : auto }
</style>

# Tezos Governance

A working, real-life, example of on-chain governance.

----

# Tezos

Tezos is a Proof-of-Stake, self-amending blockchain with on-chain governance and a strong emphasis on formal verification:
- Proof-of-Stake: no wasted compute/energy
- Self-Amending: can seamlessly change its code
- On-Chain Governance: changes are decided by stakers
- Formal Verification: automated safety

----

# Self-Amendment Examples

- Various performance optimizations and QoL improvements
- Inflation Funding: Creating money on amendments
- Tickets: Adding linear resources to Tezos (Diem like)
- Sapling: Adding private transactions to Tezos (Z-Cash like)

---

# Table of Content

1. Technological Layer
- Soft Forks vs Hard Forks
- What is a Blockchain?
- Voting Schedule

2. Social Layer
- Not everything is technological!
- Coordinating Core Teams
- Coordinating _everyone_

3. Future of Tezos Governance
- Commons Market
- Futarchy everywhere!
- Full Integration

---

# Technological Layer

1. Soft Forks vs Hard Forks

2. What is a Blockchain?

3. Voting Schedule

----

# Soft Forks vs Hard Forks

- Soft Fork = backward compatible change. Anyone can make the change, and still be included in the network. Example: Restrincting set of operations that can be emitted.
- Hard Fork = breaking change. Making the change excludes one from the network. Example: Adding a new operation, or changing the state.

Problem: Most interesting changes are breaking changes. Adding ZK transactions, moving from PoW to PoS, etc.
Solution: Self-amendments through on-chain governance.

----

# What is a Blockchain? 1/2

In Tezos, the swappable part of the blockchain is _the code of the protocol_. Idea is that this code contains most defining features of Tezos:
- Consensus
- Cryptoeconomics
- Smart-contract virtual machine
- Currency management (transfer & inflation)

----

# What is a Blockchain? 2/2

Code is data! As such, the blockchain (defined as _the code of the protocol_) is just data like any other.
It can be hashed, stored onchain, and mentioned by the protocol itself. And that's exactly what we do: we post the next competing protocols on chain, and we vote on them.

----

# Voting Schedule

Each period lasts 2 weeks.

1. Proposal: Core teams submit the protocol proposals. Bakers vote on the one they prefer the most.
2. Exploration: Bakers learn more about the proposals. Q&As, etc. They then revote on it until a super-majority (~80%) is reached.
3. Testing: A testnet is spun.
4. Promotion: Bakers vote a final time to validate the protocol, if nothing went bad during the Testing period.
5. Adoption: Nothing happens for two weeks, but everyone knows that the protocol will be integrated. This is meant to leave time for tool developers to do the final modifications to integrate everything.

This is the result of many iterations!

---

# Technological Layer: Conclusion

The technological problem is not trivial.
It took a lot of development effort to solve, but we got there! Others still are not. (_mention others_)

However, as we will see, we still have major hurdles to solve.

---

# Social Layer

1. Not everything is technological!

2. Coordinating Core Teams

3. Coordinating _everyone_

----

# Not everything is technlogical!

- Decentralization: of stake, core, tools, funding
- Strategy in a rapidly changing world
- Remote coordination across the world with a number of actors > Dunbar number
- Processes to maintain an evolving codebase that integrates the state-of-the-art of many fields

----

# Coordinating Core Teams 1/3

Core is the most critical part, contains the Protocol and the Shell. But many people work on this:
- Nomadic Labs: Historical core team of Tezos, ~50 people. Based in Paris. Worked there for 1 year.
- Marigold: New team, ~10 people. 100% remote with people all around the world. Founded on November 2020.
- Other smaller teams, ~10 people.

----

# Coordinating Core Teams 2/3

Lots of cultural differences:
- NL office based vs Marigold fully remote
- NL academic vs Marigold technical
- NL French culture vs Marigold world culture
- NL process oriented vs Marigold results oriented
- OCaml x OPAM vs ReasonML x esy

----

# Coordinating Core Teams 3/3

Anecdotes:
- People moving across teams = exploiting comparative advantage
- Diversity = covering each other's blindspots
- Cooperation **and** competition. Consensus first

----

# Coordinating _everyone_ 1/2

Communication costs are quadratic, but a blockchain necessarily has a lot of different actors:
- Core teams
- Tools
- Stakers
- DApp builders
- Users
- Foundation

----

# Coordinating _everyone_ 2/2

**Lots** of small technical teams:
- Information is siloed
- Managers are usually tech lead, rather than actual managers

Suboptimal tools:
- Slack & Tezos Agora. Central places for conversations.
- TZIPs. In the vein of EIPs.
- Regular meetings between key actors.

A lot of boring work. Taking all suggestions!

---

# Social Layer: Conclusion

- I am not good at this!
- Trade-offs between short-term and long-term (_example of programming langs_)
- Place where we can improve things the most

---

# Future of Tezos Governance

1. Commons Market

2. Governance everywhere!

3. Full Integration

----

# Commons Market

- Prediction Markets to rank existing proposals
- Inflation Funding to fund development team so that they capture the value of what they bring
- Commons Market! Now, incentive to increase the value of the token as much you can, as you can capture a part of it

----

# Futarchy Everywhere

- Prediction Markets/Contests to rank grant requests
- Stake-weighted votes to identify priorities
- Perpetuitez to get a fully on-chain indicator of trust in the chain

----

# Full Integration

- Goal: Knowing when anything breaks on protocol changes
- Solution: Complete integration from the protocol to the dapp, as well as tools
- Already exists in OCaml and Coq! OPAM CI and Coq scripts on breaking changes

---

# Future of Tezos Governance: Conclusion

Lot of crazy things left to do.
All of them have tremendous potential, in various directions.

If you are interested in working on this, feel free to reach out at [e-mail] :)