Quickstart
Pre-1.0 — read this first
Lecture is a working prototype, not a finished product. The public container image and downloads are not published yet, so the commands below are provisional and the image name is a placeholder (registry.example.com/…). Expect names, ports and setup steps to change before 1.0, and keep backups while you experiment. Check the versions page for the current state of publishing.
Run Lecture and its database with Docker Compose. If you can paste six lines and a small YAML file, you have a workspace.
What you need
- Docker Engine with Compose v2 (the
docker composecommand) on Linux, macOS or Windows — details in Requirements. - About 2 GB of RAM free on the host and a few GB of disk.
- The Lecture application image. Until it is published, you will get it from wherever the project shares pre-release builds; the examples use the placeholder name.
- PostgreSQL 17 — the Compose file below starts it for you, you don't need one.
1 · Write the compose file
Create a folder for Lecture (you will return to it for every update) and put a compose.yaml in it:
name: lecture
services:
app:
# Placeholder name — replace with the published image once announced.
image: registry.example.com/lecture/lecture:2026.11.14
restart: unless-stopped
ports:
- "5080:8080" # host port : container port
environment:
# The app reaches the database over the Compose network by service name.
ConnectionStrings__Lecture: "Host=postgres;Database=lecture;Username=lecture;Password=CHANGE_ME"
depends_on:
postgres:
condition: service_healthy
postgres:
image: postgres:17
restart: unless-stopped
environment:
POSTGRES_DB: lecture
POSTGRES_USER: lecture
POSTGRES_PASSWORD: CHANGE_ME
volumes:
- postgres-data:/var/lib/postgresql/data
healthcheck:
test: ["CMD-SHELL", "pg_isready -U lecture -d lecture"]
interval: 5s
timeout: 3s
retries: 20
volumes:
postgres-data:
Two CHANGE_ME spots
Set the same strong password in both places before starting. The data lives in the postgres-data volume: never run docker compose down -v and never docker volume prune on a host that runs Lecture — that is the one command that deletes everything.
2 · Start the stack
docker compose up -d
docker compose ps
Wait until postgres reports healthy; the app container starts after it. If the app exits with a database error, the database is usually still booting — run docker compose up -d again once it is healthy.
3 · Open it in the browser
Go to http://<your-server>:5080 (or http://localhost:5080 on the machine itself). The first load takes noticeably longer than a normal website: Lecture is a WebAssembly app that downloads once and is then cached by the browser — after that it opens fast, even offline. Follow the first-start prompts to create the initial account, then add your study group in a workspace.
4 · Check it works
- Create a note, type a sentence, reload the page — the text is still there.
- Open a second browser (or a phone on the same network) and edit the same note: both sides see the change within a second.
- Disconnect the device from the network, keep typing, reconnect: the header status pill goes Offline → Syncing → Synced and nothing is lost.
Next steps
- Configuration — ports, database connection, image choice, backup volume.
- Updating — the safe path to a new version, and back.
- Troubleshooting — when something doesn't behave.
- Put Lecture behind HTTPS (your usual reverse proxy) before real users connect; make sure the proxy forwards WebSocket upgrades and
X-Forwarded-*headers, or sync will not work.