Problem
The dev containers in docker-compose.yml all run as root. On macOS this is fine because Docker Desktop's VM transparently handles UID translation for bind mounts. However, on Linux and Windows/WSL2, files created by root inside the container are owned by root on the host filesystem, causing permission issues for developers.
Proposed Solution
Add a DOCKER_USER variable to the dev services in docker-compose.yml that defaults to empty (preserving current root behavior for macOS and CI), but can be set in .env by Linux/WSL2 developers:
services:
experimenter:
user: "${DOCKER_USER:-}"
# .env (Linux/WSL2 only)
DOCKER_USER=1000:1000
This is a zero-impact change for macOS developers and CI — an empty user: field is ignored by Docker Compose.
┆Issue is synchronized with this Jira Task