Public Accountability Chain

A public database tracking accountibility from various sources and APIs. Optionally submits BEACON hashes to FUND Mainchain.

Development - Quick Start

Requires Node JS >= v14.4.x

Docker, Docker Compose and make are the recommended minimum requirements for development.

1. Copy .env.example to .env and edit. Copy .env to src/ui/.env.local

2. Initialise the build environment

make init-dev

Or without make:

docker-compose -f docker-compose.builder.yml run --rm install

3. Run the development environment:

make dev

Or, without make:

docker-compose up

Once up, the front-end UI can be accessed via http://localhost:3000

Changes can now be made, and will be hot loaded in the Docker environment.

Note: the Docker composition internally runs the start-dev-stack script, which only runs the UI and Data Daemon. It is intended for developers who wish to add data sources, or modify the UI.

To bring the composition down, hit Ctrl+C, then run:

make dev-down

Or, without make:

docker-compose down

Note: The Dockerised development environment does not at this time submit hashes to BEACON on Mainchain. In order to test BEACON hash submission, FUND DevNet must be running, and PAC run outside of Docker.

Components

src/daemon

There are currently 4 Daemons which can be run independently:

1. Data Daemon - gathers data from APIs and data sources, and standardises it into a common data structure before inserting into the MongoDB collection. If an indident report exists, the daemon will check from cross references and update the record with links to the duplicate data as required.
2. BEACON Daemon - if BEACON variables have been configured in .env, this daemon will submit the data hashes in batches to the Unification Mainchain. The hashes submitted are generated from the standardised data.
3. Merkle Tree Daemon - generates a Merkle tree from the hashes that have been submitted to Unification Mainchain. The root hash is also submitted to Mainchain.
4. Backup Daemon - creates a full backup of all collections in the MongoDb database. This backup is saved to IPFS offering an immutible copy of the entire database to be available at all times.

Daemons can be configured to run with the desired frequency using a tool such as cron

For example:

crontab -e

and setting

0 * * * * /path/to/node /path/to/lib/daemon data  >> /path/to/log_file.log 2>&1

will run the Data Daemon every hour, and log the results to a file

src/ui

A Next.js (React) front-end that reads data from the MongoDB database.

Developing/testing outside Docker environment

A MongoDB service is required, either running locally or via the cloud.

1. Run npm install
2. Copy example.env to .env and modify MongoDB settings etc. as required.
3. Copy .env to src/ui/.env.local
4. Run node src/db_utils/create_db.js to create DB, collections and indexes
5. Run npm run start-dev-stack

Note: the start-dev-stack script will only run the UI and Data Daemon. It is intended for developers who wish to add data sources, or modify the UI.

Running in Production

TODO

Static data attribution

The the-counted-2015.csv & the-counted-2016.csv data contained in the data/static directory obtained from The Guardian and included here for convenience.

Data Sources

The data used in src/daemon/apis is currently sourced from the following. More will be added over time.

• Police Brutality
• fatalencounters.org
• The Guardian
• Killed By Police
• Mapping Police Violence
• US Police Shootings
• Washington Post