Kedro Polis-like Pipelines: Template

This project is an attempt to use Kedro to model Polis-like data pipelines.

This repo builds and runs a set of polis-like pipelines via continuous integration, and publishes a small micro-site that exposes all the pipelines and data.

Usage

You can fork this repo directly, or use it as a template repo for your own Polis conversations.

Existing Pipeline Repos

• Blacksky: Community Guidelines. about
  • pipeline repo | micro-site | data explorer 🔭
  • original polis: report
• Austrian Klimarat. about
  • Energy
    • pipeline repo | micro-site | data explorer 🔭
    • original polis: report
  • Mobility
    • pipeline repo | micro-site | data explorer 🔭
    • original polis: report
• San Juan Islands Land Trust (testing geographic projections)
  • pipeline repo | micro-site | data explorer 🔭
  • original polis: report

Background

Polis is a collective intelligence tool for collecting simple agree/disagree data and from that building maps of the opinion space in which participants reside. This allows sensemaking by surfacing complexity in the groups that agree/disagree together.

Goals

• allow for more visibility into existing Polis pipeline
• support exploration of new parameters and algorithms
• support collaboration on these new pipeline variants
• support generation of standardized data types that new UI can be built around
• modularization of pipeline steps
• help determine best architecture for the standalone red-dwarf algorithm library

Usage

# Build static site
uv run make build
# or: make build

# Run all pipelines
uv run make run-pipelines
# or: make run-pipelines

# Run specific pipelines with parameters
uv run make run-pipelines PIPELINES=bestkmeans PARAMS="polis_id=r29kkytnipymd3exbynkd"
# or: make run-pipelines PIPELINES=bestkmeans PARAMS="polis_id=r29kkytnipymd3exbynkd"

# Start development server
uv run make dev
# or: make dev

# Serve build directory
uv run make serve
# or: make serve

# Show help
make
# or: make help