Files
agentic-govbot/templates/do-ocracy.md
Nathan Schneider bda868cb45 Implement LLM-driven governance architecture with structured memory
This commit completes the transition to a pure LLM-driven agentic
governance system with no hard-coded governance logic.

Core Architecture Changes:
- Add structured memory system (memory.py) for tracking governance processes
- Add LLM tools (tools.py) for deterministic operations (math, dates, random)
- Add audit trail system (audit.py) for human-readable decision explanations
- Add LLM-driven agent (agent_refactored.py) that interprets constitution

Documentation:
- Add ARCHITECTURE.md describing process-centric design
- Add ARCHITECTURE_EXAMPLE.md with complete workflow walkthrough
- Update README.md to reflect current LLM-driven architecture
- Simplify constitution.md to benevolent dictator model for testing

Templates:
- Add 8 governance templates (petition, consensus, do-ocracy, jury, etc.)
- Add 8 dispute resolution templates
- All templates work with generic process-based architecture

Key Design Principles:
- "Process" is central abstraction (not "proposal")
- No hard-coded process types or thresholds
- LLM interprets constitution to understand governance rules
- Tools ensure correctness for calculations
- Complete auditability with reasoning and citations

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-08 14:24:23 -07:00

262 lines
8.2 KiB
Markdown

# Do-ocracy Constitution
This constitution establishes governance based on initiative and action: those who step up to do work have authority over how they do it.
## Article 1: Core Principle
### Section 1.1: The Do-ocracy Principle
"If you want to see something happen, make it happen."
- Members who take initiative have authority over their work
- Contributors decide how to accomplish their projects
- Action and contribution create legitimacy
- Community trusts those who do the work
### Section 1.2: Autonomy and Responsibility
Those who do the work:
- Have freedom to choose their approach
- Are responsible for the outcomes
- Should communicate what they're doing
- Remain accountable to community values
### Section 1.3: Low Barriers to Action
This constitution aims to:
- Minimize bureaucratic overhead
- Encourage experimentation and initiative
- Trust members to make good decisions
- Let results speak for themselves
## Article 2: Membership
### Section 2.1: Who Can Do?
All community members can take initiative:
- New members equally welcome to contribute
- No formal permission needed for most actions
- Contribution is the primary membership activity
- Bot tracks member participation and contributions
### Section 2.2: Member Rights
All members have the right to:
- Propose and execute projects
- Make decisions within their work
- Request resources for their initiatives
- Collaborate with other members
- Discuss and advocate for approaches
### Section 2.3: Joining the Community
New members join by:
- Expressing interest and introducing themselves
- Starting to participate and contribute
- Following the code of conduct
- Learning community norms through doing
## Article 3: Taking Initiative
### Section 3.1: Starting Something New
To launch an initiative:
1. Announce what you plan to do
2. Explain the goal and approach
3. Mention @govbot to log your initiative
4. Start working
5. Keep community informed of progress
### Section 3.2: Authority Through Action
When you take initiative:
- You have authority over your project
- You decide on implementation details
- You set your own timeline and milestones
- Others can offer input but you make the calls
- Bot supports your work with necessary capabilities
### Section 3.3: Collaboration Welcome
While doers have authority:
- Others can offer to help
- Collaboration is encouraged
- Contributors share decision-making
- Lead doer coordinates unless group decides otherwise
## Article 4: Limits and Boundaries
### Section 4.1: When You Need Permission
Some actions require community input via lobbying (Article 5):
- Spending significant community funds
- Making major platform configuration changes
- Establishing new official policies
- Actions that restrict other members
- Changes affecting everyone's experience
### Section 4.2: Reversible Actions Preferred
Do-ocracy works best when:
- Actions can be undone if problematic
- Changes can be iterated on
- Experiments are low-risk
- Bot maintains reversibility logs
### Section 4.3: Respect for Others' Work
Don't interfere with others' initiatives:
- Don't undo someone's work without discussion
- Don't take over projects without invitation
- Offer to help rather than criticize
- If you disagree, do it differently yourself
## Article 5: Lobbying and Input
### Section 5.1: Announcing Intentions
Before major initiatives:
- Post your plans to the community
- Explain what you want to do and why
- Mention @govbot to create discussion thread
- Give others time to respond (typically 2-3 days)
### Section 5.2: Listening to Input
During lobbying period:
- Community members offer feedback
- You consider the input seriously
- You may adjust plans based on concerns
- You make the final call as the doer
- Bot facilitates discussion
### Section 5.3: Proceeding Despite Concerns
You can proceed even if some disagree:
- If you've heard the concerns
- If you're willing to be responsible
- If you'll reverse if it doesn't work out
- But if strong opposition, consider pausing for more discussion
## Article 6: When Things Go Wrong
### Section 6.1: Reversing Actions
If an action causes problems:
- Community discusses the issue
- Original doer can reverse their action
- If unavailable, others can reverse with discussion
- Bot tracks reversals and reasons
### Section 6.2: Learning from Mistakes
Mistakes are learning opportunities:
- No punishment for good-faith efforts
- Discuss what went wrong
- Adjust approach for next time
- Community supports experimentation
### Section 6.3: Addressing Problematic Patterns
If a member repeatedly:
- Ignores community input
- Causes harm through their actions
- Acts in bad faith
- Violates community values
Then stronger intervention needed (see Article 7).
## Article 7: Accountability and Boundaries
### Section 7.1: Code of Conduct
All actions must align with:
- Community code of conduct
- Shared values and principles
- Legal requirements
- Platform terms of service
### Section 7.2: Community Intervention
The community can stop or reverse actions that:
- Violate code of conduct
- Harm members or community
- Exceed reasonable authority
- Demonstrate bad faith
Process:
1. Member raises concern about action
2. Community discusses (minimum 2 days)
3. If rough consensus to stop/reverse, bot implements
4. Original doer can explain and defend their action
### Section 7.3: Removing Harmful Members
For serious or repeated violations:
- Community discusses removal (minimum 5 days)
- Member in question can participate in discussion
- If rough consensus, member removed
- Bot implements removal decision
## Article 8: Resources and Infrastructure
### Section 8.1: Access to Tools
Members should have access to:
- Platform capabilities for their initiatives
- Bot functionality to support their work
- Information and documentation
- Collaboration tools
### Section 8.2: Resource Allocation
For limited resources:
- Small amounts: just take what you need
- Larger amounts: lobby the community first
- Track usage so others know what's available
- Bot helps track resource allocation
### Section 8.3: Infrastructure Changes
Platform configuration changes:
- Routine maintenance: do-ocracy applies
- Major changes: lobby community first
- Security updates: act immediately, announce after
- Bot logs all infrastructure changes
## Article 9: Conflict Resolution
### Section 9.1: Direct Communication
For conflicts between doers:
- Talk directly first
- Explain your perspectives
- Try to find mutually acceptable solution
- Assume good faith
### Section 9.2: Community Mediation
If direct discussion doesn't resolve:
- Bring to community for input
- Each side explains their view
- Community helps find middle ground
- Bot facilitates mediation process
### Section 9.3: Rough Consensus
For persistent conflicts:
- Community gauges rough consensus
- Not unanimous but general agreement
- Direction with most support prevails
- Minority can still do it differently if not disruptive
## Article 10: Constitutional Matters
### Section 10.1: Constitutional Interpretation
When constitutional meaning is unclear:
- Member requests interpretation from @govbot
- Bot provides interpretation with reasoning
- Community discusses if interpretation is contentious
- Rough consensus on interpretation guides future cases
### Section 10.2: Constitutional Amendments
To amend this constitution:
1. Member proposes amendment
2. Community discusses (7 days minimum)
3. If rough consensus supports change, amendment adopted
4. Bot updates constitution
### Section 10.3: Core Principles
Amendments should preserve:
- Low barriers to action
- Authority through contribution
- Individual initiative and autonomy
- Accountability to community
- Trust in members
---
## Implementation Notes
This constitution creates action-oriented governance:
1. **Empowerment**: Members can act without permission
2. **Efficiency**: Low overhead for getting things done
3. **Flexibility**: Doers choose their approach
4. **Accountability**: Community can intervene when needed
5. **Trust**: Assumes good faith and responsibility
The bot should support member initiatives, track what's happening, facilitate lobbying when needed, and help the community intervene only when actions become problematic. Err on the side of letting people do rather than blocking action.