Poetry vs. venv: Python Devs Are At War Over Dependency Hell's Modern Cure
The practical struggle in Python development revolves around dependency management, pushing developers toward modern tools like Poetry, uv, and pixi to escape version conflicts.
The user base is sharply divided. Somegeeks like "somegeek" argue the core issue is the "general instability of the Python ecosystem," suggesting external failure. Others counter, adopting the blame culture: "it’s me not python" (flandish). Meanwhile, pros champion Poetry/uv, comparing them to Rust's cargo via TOML files, while "Incblob" notes that even these tools fail when library bugs force painful downgrades.
The weight of opinion shows a clear desire for standardization. The consensus is that current dependency methods are inadequate. The fault line exists between blaming the inherently unstable ecosystem and accepting that adopting advanced, restrictive tools is the only path forward.
Key Points
Modern tools like Poetry, uv, or pixi are necessary to handle Python dependency management complexity.
Multiple users pointed out the inadequacy of traditional methods like relying solely on venv.
The Python ecosystem itself is fundamentally unstable and difficult to maintain.
"somegeek" stated the core pain point is the general instability of the ecosystem, making hard-to-update versions a constant struggle.
The dependency headache might stem from user workflow rather than the language itself.
"flandish" offered the viewpoint: "it’s me not python."
For large projects, dependency resolution challenges persist even with advanced tooling.
"Incblob" warned that the challenge remains when library bugs force painful downgrades, regardless of the tool used.
There is interest in alternatives that blend conda compatibility with pyproject.toml.
"rutrum" recommended pixi as an alternative for users comfortable with conda workflows.
Concurrency improvements are blocked by slow adoption in critical C-extensions.
"fubarx" identified waiting for libraries like _cryptography_ and _psycopg_ to implement GIL-free support as a major blocker for true parallelism.
Source Discussions (3)
This report was synthesized from the following Lemmy discussions, ranked by community score.