Mise: Dev tools, env vars, task runner

Original Article ↗ Hacker News Discussion ↗

Overview & Use Cases

  • Many commenters use Mise to unify language runtimes and dev tools (Node, Python, Ruby, Go, Java, Rust, etc.) and to replace multiple tools like asdf, pyenv, nvm, rvm.
  • Common uses: project-local toolchains, automatic virtualenv activation, environment variables per project, and as a general task runner.

Compared to asdf, nvm and other managers

  • Several users treat Mise as a “better asdf”: compatible with .tool-versions, faster, path-based instead of shims, and perceived better security.
  • nvm is widely criticized as slow and brittle; benchmarks show Mise’s shell integration running orders of magnitude faster.
  • Some prefer single-language tools (e.g., fnm for Node) if they only use one ecosystem, but multi-language users favor Mise.

Task Runner vs Make/Just/Other Tools

  • People use Mise tasks to replace Makefiles in Python/JS projects and unify backend/frontend workflows.
  • Compared to just:
    • Mise tasks are more verbose but considered easier to read for newcomers.
    • Extra features: parallelism, file watching, argument parsing, custom completions, and script/file-based tasks.
  • Some still prefer Make for its power and familiarity, or poe/poetry/npm scripts for Python/JS.

Python Ecosystem & Tooling Sprawl

  • Thread extensively debates Python packaging and environment tools (pipenv, poetry, uv, pyenv, direnv, etc.), with frustration over fragmentation and churn.
  • Some praise Mise’s integration with uv and virtualenv handling; others worry about abstraction leaks and subtle bootstrapping failures in multi-language tools.

Nix, Devbox, devenv Comparisons

  • Advocates of Nix (and derivatives like Devbox, devenv) argue it already solves reproducible dev environments and task running.
  • Counterpoints: Nix UX and docs are described as “terrible” or hard to learn; Mise is seen as easier, quicker to onboard, and not tied to Nix’s “split-brain” model.
  • Some see Mise as a pragmatic “90% solution” built on precompiled binaries; others argue that relying on vendor binaries is incomplete or risky.

DX, Integrations, and Limitations

  • Positive notes: fast shell startup, simple mental model, helpful shebang trick, GitHub Actions support, JetBrains plugin.
  • Pain points: JetBrains integration can be finicky; docs are improving but some users still find gaps; Windows/WSL support is noted as “minimal”.
  • Release cadence is very frequent; maintainers of downstream packages find it burdensome, though others appreciate rapid bug fixes.