Community Vision & Adoption Aspirations

This document reflects on the current state of dependency management in Python and the philosophy guiding vnma.

1. The Context: The Virtualenv Paradigm

For over two decades, Python's community standard for application isolation has been the virtual environment (venv / virtualenv). By creating a directory containing a copy or symlink of the Python executable alongside an isolated site-packages directory, developers can install libraries without administrative rights or conflicts with system-level packages.

This model solved package isolation for its time, but introduces specific trade-offs:

• Footprint & Redundancy: Replicating Python interpreter binaries, scripts, and base tooling for every single project consumes storage and increases initialization latency.
• Non-Portability: Virtual environments hardcode absolute paths into script headers (#!...) and configuration files (pyvenv.cfg), making them non-relocatable and non-exportable across filesystem locations.
• Tooling Fragmentation: Activating and deactivating environments introduces cognitive friction across terminal multiplexers, shell scripts, CI runners, and IDE configurations.

2. The Vendoring Alternative

Languages such as Go, Rust, and JavaScript (with tools like pnpm, yarn zero-installs, and Go vendoring) have demonstrated the ergonomic advantages of project-local dependency storage:

• Dependencies reside cleanly within the workspace.
• The system or container Python executable is reused directly without mutation.
• Sharing a project requires only sharing source definitions; dependencies are restored deterministically in seconds.

vnma seeks to offer this workflow natively to Python developers. By combining:

1. Standards-compliant wheel distribution extraction,
2. Hermetic path sanitization at runtime,
3. Deterministic DLL registration on modern operating systems, and
4. Explicit, zero-magic configuration files (vnma_config.toml),

vnma aims to demonstrate that application-level vendoring in Python is not only feasible, but can be robust, predictable, and clean.

3. Aspirations for Python Ecosystem Adoption

While virtual environments will remain the default for many workflows, our hope is that vnma contributes a viable, lightweight alternative for:

• CLI Tools & Desktop Applications: Where self-contained, relocatable distributions simplify packaging.
• Game Development & Embedded Engines: (e.g., Pygame, Godot-Python, Blender scripting) where binary dependencies need direct, deterministic loading.
• Microservices & Serverless: Where cold-start latency and minimal container layers provide tangible performance benefits.

Feedback, community experimentation, and real-world testing from Python developers are welcome as vnma progresses through Beta.