1/9/90 in Crypto

1/9/90 in Crypto

Distributed autonomous organizations (DAOs) are one of the most vibrant areas of crypto. DAOs are groups of people with the right to vote to change how a crypto company operates.

DAOs pop up for many different reasons: to form a social club that requires a token to join, to buy a copy of the Constitution, to manage open source software. Almost any group of people can organize a DAO to pursue a shared purpose.

Anyone holding a voting token can vote, much like anyone who buys a share of voting stock in a public company can vote during a shareholder meeting.

But DAO culture is unique. It’s no boardroom. The vibe is more Github, less Goldman.

Within DAO forums, token-holders propose ideas. Any member of the DAO can suggest a proposal, which members debate then ratify or strike-down. A proposal may advocate altering the roadmap (evolve from proof-of-work to proof-of-stake); or distributing some funds from the company’s treasury (as a dividend or to pay a vendor for a service); or advocate changes voting rules, among many other things. To get a feel for these types of conversations, visit Uniswap’s DAO.

DeepDAO, a provider of data on DAOs, counts more than 4000 DAOs with more than 0.5m active voters.[1] By examining the behavior of the top 20 most valuable DAOs, we can see some patterns emerge.

No alt text provided for this image


 First, DAO populations follow a power law. ENS (Ethereum Name Service), which sells usernames for people and names for wallets, manages the largest of these DAOs with more than 95,000 active members who have cast 98.6k votes across 15 different proposals.

In contrast, Radicle counts 68 active members for a DAO worth $249m who have cast 76 votes across 5 proposals, according to the DeepDao data.

No alt text provided for this image


Second, there seem to be at least two clusters DAO types by activity rate. Those with high member participation rates and those with very low participation rates. This behavior seems to be scale-free or independent of size.

ENS has more than 100,000 members and nearly 100% of them are active. Aave has a similar size voter base but the opposite level of activity. OlympusDAO and Ribbon break the mold with relatively large voting populations a moderate rate of activity at 25%.

What’s best? It depends on the goals of the DAO. A social club might seek maximum participation. A highly technical Defi protocol may prefer a coterie of experts to shepherd its direction. Some DAOs see many proposals. Decentraland, a 3D virtual world, counts more than one thousand; while others' ballots number in the single digits.

One of the key responsibilities of DAO founders will be determining how they prefer their community to behave, what norms they will seek to encourage. Setting those early will be key to the success of the DAO.

I suspect 1/9/90 heuristic - 1% of the population creates, 9% comments, and the remaining 90% lurk - will emerge as the dominant pattern. This rule seems to govern most social networks today and DAOs are another form of social network - one with hundreds of millions of dollars behind them.

DAOs hold the promise to a new way to organize people with a shared interest in project or purpose. We should expect some of the same behaviors in existing social networks to arise within DAOs, and some entirely new ones enabled by large treasuries and the voting construct that enables self-determination.

[1] DeepDAO is in alpha so there’s likely to be some tightening up of the numbers as the company progresses.

[2] Active rates are noisy because voters can provide their votes to a protocol proxy, who votes on their behalf.



love concept of 1/9/90. but have to agree with David Schoenberg ☁️ what is TAM of DAOs? social media had growth >200% is there similar rate with DAO. not looking if this is a niche a use case

Like
Reply
Porendra Pratap

Bachelor of Commerce - BCom from Nizam College at Hyderabad Public School

2y

👍

Like
Reply

thanks for sharing. how do you see this evolving. decentralized boardrooms, purchase large assets and make them public goods, etc. Naveen Koorakula

Luigi Mallardo

Go-To-Market Leader | 6X hands-on Chief Revenue Officer (CRO) | 4 Exits | Helping B2B SaaS, AI agent and recurring revenue businesses grow healthily and get a good exit as a result

2y

To view or add a comment, sign in

More articles by Tomasz Tunguz

  • Back to Text: How AI Might Reverse Web Design

    Back to Text: How AI Might Reverse Web Design

    AI saved me from cookie banners, travel insurance popups, car rental quotes, & the special frustration of comparing…

    26 Comments
  • 2025 Predictions

    2025 Predictions

    Every year I make a list of predictions & score last year’s predictions. Here are my predictions for 2025.

    7 Comments
  • My Favorite Books of 2024

    My Favorite Books of 2024

    From the pit lanes of Formula One to the secretive world of commodities trading, from the championship poker tables to…

    8 Comments
  • Desktop, Touch, Browser, Now AI? The Next OS in Computing

    Desktop, Touch, Browser, Now AI? The Next OS in Computing

    Remember the first time you touched a computer screen instead of typing commands? We’ve lived through distinct epochs…

    13 Comments
  • The Intelligence Imperative: How AI Went From Feature to Foundation

    The Intelligence Imperative: How AI Went From Feature to Foundation

    If you needed proof that every software company is an AI company, here’s the evidence. 46% of all US venture financings…

    7 Comments
  • Finding the Midnight Snacker : Agentic Marketing

    Finding the Midnight Snacker : Agentic Marketing

    “Feel Good Fun Mix” tops my recommendations on Spotify. Spotify has created over 6,000 such labels by hand.

    2 Comments
  • I Use AI 100 Times Per Hour

    I Use AI 100 Times Per Hour

    How often do you use AI? I tracked my Sunday workday to find out. Between 4:30-9:00 PM (with a dinner break), I…

    5 Comments
  • Top 10 Posts of 2024

    Top 10 Posts of 2024

    These are the top 10 posts of 2024. I like to do a roundup each year as a retrospective of how much the industry has…

    2 Comments
  • Why I Let AI Grade My Writing (Just Like My AP English Teacher)

    Why I Let AI Grade My Writing (Just Like My AP English Teacher)

    Ten years ago, I hired an editor to grade my blog posts as an AP English teacher might. Yesterday, I asked an AI to do…

    7 Comments
  • The Year Algorithms Learn to Act

    The Year Algorithms Learn to Act

    In the bustling tech campuses of 2024, the age of passive AI – systems that merely respond to our queries – is giving…

    5 Comments

Insights from the community

Others also viewed

Explore topics