Colony.io Show and telll

On the last week of the month we share our projects! Gus invited Colony.io founders Jack and Auryn to show us a demo and talk about how Colony is different than other DAO platforms. Much discussion ensued! Fantastic notes by Michal.

video link: https://lbry.tv/@dGovChats:a/dgov20May2020Colony-io:8

dGov Chat: Guest Colony.io, May 27 2020

Today’s moderator/facilitator is Gus, who invited the guest from Colony.io, Jack du Rose

Auryn the floating head from the Colony team is also here to help explain.

“An Operating System for Organizations”

  • Kind of like a “DAO platform”
  • But DAO are still kind of ill- or poorly-defined, and changing

Other “platforms for DAOs” are there too, but they mainly focus on “use of funds” or “voting” applications

// Colony tries to avoid voting

Theory of the firm

  • At some point it’s more efficient for companies to have employees than just use the market

  • It happens because of transaction (or coordination) costs

    • Find supply, coordinate supply & demand, do remittances, create agreements, enforce contracts, etc.

“Earning meant getting a job, and work needed management”

Today, software increasingly coordinates supply w/ the market”

DAOs

  • Have to coordinate everything remotely
  • This coordination should be more efficient than traditional means
  • Voting is not a great coordination mechanism, so DAOs have limited uses

Colony

COLONY // Decentralized, self-organising companies where decision-making power derives from high-quality work

  • I.e. the more good work you did in the past, the more say you have
  • Hierarchy as an emergent property (based on engagement, and quality of work)
  • ANTs not Empires
  • Complex Adaptive Systems -> Simple rules lead to emergence of collective behavior

Key principles

  • “Permissiveness” - it’s ok unless it’s not
    • Works for Wikipedia - you can make edit, until you can
  • Believability-weighted decision making
    • Like with Bridgewater associates
    • The more good past work you did in the domain, the more believable you are
    • Weight decision NOT by amount of tokens you have, but by quality of your past contributions
  • Avoid voting
    • Use only as last resort

There’s a live application, updated regularly

Built for developers too

It’s a platform

NO token

  • Colony never did token sale
  • Comes down to legal concerns… (for ICOs)
  • Most tokens have had little to no utility or need for existence

  1. If you can do it with Ethereum - think twice about doing your own token

  2. If people go away value flatlines? - It’s bad

If you’re selling a token…

COLONY TOKEN MODEL

  • Transaction fees on payments
    • Now effectively zero
  • reputation mining
  • Not a security, by design

Will likely do a sale, in the future:

  • Introducing a new mechanism

The Colony Model

  • Why colony doesn’t want to make a decision by voting?
    • Example of: Should Ann be reimbursed for a coffee from last week?
      • In companies it doesn’t work like that
      • People get responsibilities and budget
      • That drives efficiency
      • Voting on everything would make…
        • Stuff awfully slow
        • You do that infrequently
  • Goal of Colony - make companies more agile in the real world

Three paths:

  • Changing something
    • E.g. a new feature,
  • Two - “happy paths”
    • Cancel proposal
    • No objection
  • Third path
    • When someone has objection
      • Effectively creating a “motion in opposition”

Fold = concede the proposal, lose part of the stake (e.g. 10%)

Call => Voting

Loser / Winner get some stake returned

  • Stake * (90% * [% of votes received])

Based on results it’s either pass or fail

Escalation

  • to a higher level in the organization, with more “reputation”

Q&A notes:

  • Reputation decays over time
    • To reward more recent contributions
  • “How are topics organized”?
    • It’s decided by each colony
    • Sample hierarchy example below
      • If you earn reputation in colonyDapp, you’ll also be getting some reputation in Dev (small share) and Root (even smaller share)
    • There’s ability to allocate skills to expenditures

“What prevents Colony from raising protocol fee thus increasing my org transaction costs?’’

  • That’d be mediated by the “Meta-Colony”

Milestone vs. upfront-system

  • With Colony you can check and vote on the work as it goes vs.
  • Autark vs. Aragon lawsuit

EXAMPLE

Future mechanism => Funding blocks, to automate them

Q: “Conflict between teams, e.g. UX vs. Dev”

  • It’d go to the top layer where everyone has a bit of say, based on past work

Q: “Decay rate of reputation?”

  • Fixed for now, but it’s on the roadmap to differentiate

Q: How much upper-layer reputation do you get?

  • Based on total reputation in the system, not a predefined ratio

Q: Can we understand a bit about the budget allocation? It seemed like there is a super-user who gets to allocate the budget to the different departments?

  • There are some differences between now and white paper
    • Core assumption of modularity
  • Permissions were tied to personas, but now can be delegated
    • E.g. to smart contracts

Q: “Budget box”? Can you tell more about it?

  • Kind of a passion-project for now, more than a key milestone
  • Target for end of the year
  • “A way of providing a continuous supply to tokens to people building on Colony”
    • Pairwise voting
    • Based on KPIs and additional engagement generated
  • Creates more of continuity for funding, i.e.
    • Expectation of more regular funding, vs. lump-sum, one-off grants,
  • Takes is fairly simple inputs (pair-wise comparisons, e.g. I like this more than that), but
    • Produces fairly complex outputs
    • Algorithm aggregates seemingly random walks into a joint probability and distribution model for all the features and work
    • Still, low intensity cognitive task

Q: How did you fund development if you didn’t do an ICO yet?

  • Private sale + bootstrapping

Q: “Coin machine? Is it available?”

  • No, not available yet
  • It’s WIP
  • Solidity is done, under testing
  • Need to clarify legal + improve UI
  • Hope to get it out by the end of the year
    • And test it with CLNY
    • Goal - get the token out there

Q: “What does reputation look like? Esp. domain-wise”

  • For now, you see the root-domain reputation

Q: “Policy-making for the initial CLNY”

  • E.g. initial distribution, inflation rate, etc.
    • Target - 10% “inflation” of INITIAL supply (asymptotically to zero in infinity)
    • There are also mechanism to burn CLNY
    • Initial supply - 1 Billion tokens

Q: “Reddit & Colony”

  • Interesting way to see each community as a potential colony
  • Pioneering experiment with “donuts” token
    • A chance to see and observe results

Q: “It’s a lot and confusing… How can I get started in this field?”

  • You need more people
    • To get more people and apply the knowledge they need to understand it first
      • E.g. repairs of a building
  • People who are most interested in those, they’re domain experts, e.g. in banking, transportation, etc. — they speak different language
    • “They want to use it, (…) but they don’t understand it”
    • Is there a plan for some educational materials?
      • Q&A, videos, etc.
      • For people in banks, etc.

“It’s a huge challenge communicating — complex ideas, new ideas [and which require] a lot of implicit knowledge”

  • Key questions:
    • Is it a potential market?
      • Colony is probably less of a tech for existing enterprises
      • Blockchain tech creates a lot of obstacles for enterprise - e.g. privacy, systems integration
  • Where could it be used?
    • E.g. games
    • Generating real value, self-managing,

Q: “Is Colony live on any testnets? or only on mainnet”

  • It’s on mainnet and goerli. but the app only accesses the mainnet deployment
  • aragon offers their dapp on rinkeby and it’s a great way to get started without spending any eth