E-ID Hands-on workshop

These jupyter-notebooks are for the hands-on workshop organized by the C4DT on the 29th of October 2024. This project showcases usage of E-ID through different technologies. It goes through clarifying examples of some of the most important concepts in E-ID today including:

• Selective Disclosure
• Unlinkability
• Predicate Proofs using Zero-Knowledge Proofs

The excercises focus on one of the most prominent cryptographic schemes in the E-ID space today BBS+.
Learn more about BBS+ signature scheme here.

You can find the slides here:

• Overview of eID landscape
• Hands-on workshop on E-ID cryptography
• Summary of the privacy challenges

Installation

This project depends on Python, and Node.js. It has been tested with Python 3.11 and node 18 and 22. There are instructions to install it globally on your computer, or use Devbox for a simpler installation.

Globally

Make sure that you have node and python installed in the correct versions:

python --version
# Python 3.11.10
node --version
# v22.8.0

To install the library dependencies and start jupyter-lab, execute the following commands. If you're using zsh on Mac, you can ignore the errors from the # comments.

# Create virtual environment and activate it.
# If this returns an error, be sure that `python --version` is 3.11.
python -m venv .venv
. .venv/bin/activate

# Install python dependencies
pip install -r requirements.txt

# Install the npm dependencies
npm install

# Install Typescript Kernel for jupyter notebook
npx tslab install

# Run jupyter-lab
PATH=$PATH:node_modules/.bin jupyter-lab

Alternatively, use Devbox

We support installation using Devbox which facilitates environment management.
To install devbox, run the following command for Linux, MacOS, WSL2:

curl -fsSL https://get.jetify.com/devbox | bash

Once devbox is install, you can run the following command to launch jupyter-lab:

devbox run jupyter

This will take care of the correct python and nodejs version, install the packages, and run the jupyter-lab. You might need to add --pure to the command: devbox run jupyter --pure, to make sure no other binaries from your computer interfere.

If the page doesn't open automatically on your computer, look at the output of the devbox run jupyter command:

    
    To access the server, open this file in a browser:
        file:///Users/something/Library/Jupyter/runtime/jpserver-12345-open.html
    Or copy and paste one of these URLs:
        http://localhost:8889/lab?token=thisisnothex480de91eb865b15f43f6fe973b95eb7a64dc
        http://127.0.0.1:8889/lab?token=thisisnothex480de91eb865b15f43f6fe973b95eb7a64dc

Of course your token will be different, so don't use these URLs!

Exercises

The following exercises are available:

• Exercise 1 - Simple Signing
  • Section 1 - Basic E-ID example using RSA cryptographic scheme
  • Section 2 - Selective Disclosure using hashing
  • Discussion - Security of this scheme, and introduction to unlinkability
  • Coding exercise - protect the hashes
• Exercise 2 - Unlinkability with BBS+
  • Section 1 - Credential - Setting up a corresponding JSON Schema
  • Section 2 - Issuer - Setting up and creating a BBS+ signature
  • Section 3 - Holder - Creating a proof
  • Section 4 - Verifier - Verifying the proof is valid
  • Discussion - Unlinkability achieved?
  • Coding exercise - Simulate an issuer-holder-verifier setup
• Exercise 3 - Zero Knowledge Proof
  • Section 1 - Issuer - Setting up and signing a credential
  • Section 2 - Creating a range proof using LegoGroth16
  • Section 3 - Creating a range proof using Bulletproofs++
  • Discussion - Are we anonymous and unlinkable now?
  • Coding exercise - Using the ZKPs
  • Hard Coding Exercise - Revocation test: add an accumlator using a negative membership check
• Exercise 4 - ZKP Measurements
  • Exercise 4a - ZKP Measurements calculations
  • Exercise 4b - Plotting of results

Solutions

There are now solutions to the challenges and questions in the exercises. You can find them in the exercise-x-solution files.

Feedback

Happy to receive feedback at c4dt-dev@epfl.ch.

LICENSE

This is licensed under CC BY-SA 4.0 and MIT license.