Funding
  • About
  • Process
    • Funding process
      • Process information
      • Process rate of change
      • Process rankings of importance
      • Delegating process decisions
      • Process complexity
    • Funding process differences
  • Approaches
    • Token funding vs network funding
      • Circular economy ideas
      • Web2 protocols & software library examples
      • Ecosystem funding factors for consideration
      • Token funding
      • Network funding
      • Funding the network, protocol and application layers
    • Contribution approaches
      • Contribution factors for consideration
      • Fully combined submission
      • Idea & contributor combined submissions
      • Independent submissions
      • Funding process data handling
    • Decision approaches
      • Decision factors for consideration
      • Fully combined decision
      • Idea & contributor combined decision
      • Independent decisions
      • Passive idea selection
      • Delegated idea selection
      • Contributor vs voter idea selection
    • Incentive approaches
      • Unsuitable incentive approaches
      • Incentive factors for consideration
      • Priority milestone incentives
      • Priority time incentives
      • Idea milestone incentives
      • Idea time incentives
      • Contributor time incentives
      • Idea cost approaches
        • Idea cost factors for consideration
        • Costs attached to contributors
        • Costs attached to ideas
        • Independent idea costs process
      • Proposal submission incentives
      • Maintenance of information incentives
    • Contribution verification approaches
      • Contribution verification factors for consideration
      • Task milestone based contribution logs
      • Project time based contribution logs
      • Individual time based contribution logs
    • Impact measurement approaches
      • Ecosystem areas for impact measurement
      • Contributors areas for impact measurement
      • Impact measuring factors for consideration
      • Priority impact measurement
      • Idea impact measurement
      • Contributor impact measurement
  • Knowledge
    • Knowledge process
  • Priorities
    • Priority process
  • Ideas
    • Idea process
    • Idea ownership approaches
      • Idea ownership factors for consideration
      • Owned ideas with fixed leadership
      • Owned ideas with distributed decisions
      • Shared ideas with elected leadership
      • Shared ideas with emergent leadership
    • Idea execution considerations
      • Idea examples
  • Contributors
    • Contributor process
    • Self sovereign identity opportunities
  • Contributions
    • Contribution verification
      • Current landscape
      • Recording & measuring contribution efforts
      • Individual monthly contribution logs
  • Outcome influence
    • Voter preferences & opinions
    • Voter outcome influence
  • Incentives
    • Proposal submission incentive approaches
    • Contribution incentive approaches
  • Impact
    • Creating impactful outcomes
  • 🔗Links
    • Treasuries
    • Income
    • Governance
Powered by GitBook
On this page
  1. Approaches
  2. Decision approaches

Idea & contributor combined decision

Idea & contributor combined decision for selecting ideas and contributors

PreviousFully combined decisionNextIndependent decisions

Last updated 10 months ago

Overview

An idea and contributor combined decision means that selection of ideas and contributors are combined into the same decision. Contributors would submit an idea along with who is contributing to that idea at the same time, voters would then decide between these ideas. The priority selection process would happen separately.

High voter participation time required (Score - 2)

Voters are able to participate in the priority selection decision and ideas and contribution selection decision separately. This means they could decide to participate in just one of those votes instead of both. This reduces the amount of context and information required to participate in the funding process as they can decide where and how they participate. Some voters may only want to participate in voting on the priority setting process which would mean they don’t need to review all of the ideas and contributor combined proposals.

High voter decision complexity (Score - 2)

The number of potential ideas can often have the largest variety of options for voters to choose from. When combining this with the contributor selection the voter complexity is increased as voters now must both understand and compare the tradeoffs with each combination of ideas and contributors with all of the others. This is instead of the more simple separate decision of which ideas are the most promising and then separately which contributors are the most suited.

Moderate voter expressiveness (Score - 3)

Voters would be able to select the exact priorities that they believe are the most important for the ecosystem. Voters would need to pick from the idea and contributor combinations presented to them meaning they would not be able to easily influence which ideas and contributors are selected individually. As the number of ideas and contributors would represent the larger part of the decision complexity this can limit the flexibility and options available to voters. Community members who wanted to contribute ideas but not execute them, and contributors who wanted to execute ideas but didn’t have an existing idea, would both currently have a higher difficulty in participating with this approach. This complexity could then result in further limiting the voters choices as some of these potential ideas and contributors might not get submitted for consideration.

Very high funding outcome influence (Score - 5, Multiply by importance of 4/5 or 0.8, Final Score - 4)

The voter's decision will influence exactly what priorities are preferred and exactly how contribution efforts are being directed through the selection of proposals that define the idea and contributors involved.

Moderate voter decision change complexity (Score - 3)

Voters can reuse the same system and process to vote on any of the priorities set and then vote again in the future to change those priorities using the same process. For updating which are the most promising ideas or contributors an ecosystem could reuse the same process and just update the idea or contributors as needed but this then adds bloat to the process due to the repeated information. Alternatively the ecosystem could introduce independent processes for updating the ideas or contributors however this then adds in more complexity by increasing the total amount of processes being used in the ecosystem to manage potential future changes.

Total score = 14 / 24