This Dockerfile installs Jupyter and the IHaskell kernel, and exposes
port 8888.
The fastest way to get up and running is:
docker run -d --name ihaskell -p 127.0.0.1:8888:8888 codebje/ihaskell
You can mount a volume into /books if you'd like external persistence of
your notebooks, and you can require a password by setting the password
environment variable when launching the container.
The notebook server runs as user ID 1000.
Be cautious about letting Docker expose ports without a specific IP address binding, as the default run state of Jupyter has no security and will allow anyone who finds your notebook server to execute arbitrary code inside your container.