Uv is the best thing to happen to the Python ecosystem in a decade • Hacker News Board

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#891

Post by franktankbank » Thu, Oct 30, 2025, 2:45 PM UTC

teiferer wrote: post #686

ederamen wrote: post #517

Uv is so good. I'm a curmudgeon about adopting new tooling, and tried uv with a lot of skepticism, but it was just better in every way. And even if it wasn't so polished and reliable, the raw speed makes it hard to go back to any other tool. Uv combined with type hints reaching critical mass in the Python ecosystem, and how solid PyLance is in VSCode, feels so good it has made me consider investing in Python as my pr…

> But then I remember that Python is dog slow compared to other languages with comparable ergonomics and first-class support for static typing, and...idk it's a tough sell. Case in point: uv itself is not written in Python. It's a Rust tool. It always amazes me when people work on an ecosystem for a language but then don't buy enough into that to actually use it to do the work. Avoidance of dogfooding is a big red fl…

Bash is written in what?

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#892

Post by dalf » Thu, Oct 30, 2025, 2:51 PM UTC

This blog post was really helpful to understand the ecosystem: https://alpopkes.com/posts/python/packaging_tools/

It mentions uv at the end and rye at first (which use uv internally).

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#893

Post by ModernMech » Thu, Oct 30, 2025, 2:51 PM UTC

darkamaul wrote: post #770

sanskarix wrote: post #675

[dead]

I think better developer tools tend to compound over time, they raise expectations and push the whole ecosystem forward. IMO, uv is quickly becoming one of the best reasons to start a new project in Python. It’s fast, and brings a level of polish and performance that makes Python feel modern again.

I think uv is one of the best examples of why not to start a new project in Python. I mean, depending on the project really, Python has its place. But if your project is anything like uv, uv proves you can't actually write an app like uv in Python. So for as great as uv is, its proximity to Python is a constant reminder of its shortcomings.

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#894

Post by oofbey » Thu, Oct 30, 2025, 2:51 PM UTC

JodieBenitez wrote: post #885

jonnycomputer wrote: post #873

I have to say that I am extremely reluctant to switch over to yet another python management system (packaging, environment, python version). Every few years someone says: this is it. Switch to poetry! Okay, I did. And, at least for some academic packages, psychopy I'm looking at you, it was a friggin disaster. so. will uv install psychopy (say version 3.2.4)?

Well... I did pip -> poetry -> rye -> uv in less than a year, so I can understand the fatigue. But the fact is that uv is well above the rest and was well worth the effort (and the wait).

I followed much of this path as well, but with conda (and mamba) in the mix. They all had fairly obvious flaws that uv doesn’t AFAICT.

Poetry was the worst for me. It doesn’t even try to manage the Python distribution, so it’s only a partial solution. It was so slow our CICD would timeout and fail. And I watched the maintainers actively refuse to fix super annoying bugs for YEARS, blaming others for the problem.

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#895

Post by the__alchemist » Thu, Oct 30, 2025, 2:55 PM UTC

prpl wrote: post #495

gostsamo wrote: post #79

Earlier quoted context omitted.

As far as I get it, conda is still around because uv is focused on python while conda handles things written in other languages. Unless uv gets much more universal than expected, conda is here to stay.

conda (and its derivatives that are also “conda” now), and conda-forge specifically, are the best ways to install things that will work across operating systems, architectures, and languages - without having to resort to compiling everything. Want to make sure a software stack works well on a Cray with MPI+cuda+MKL, macOS, and ARM linux, with both C++ and Python libraries? It’s possible with conda-forge.

You can do all of the above with Wheels.

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#896

Post by blactuary » Thu, Oct 30, 2025, 2:55 PM UTC

kardos wrote: post #92

dekhn wrote: post #58

I hadn't paid any attention to rust before uv, but since starting to use uv, I've switched a lot of my performance-sensitive code dev to rust (with interfaces to python). These sorts of improvements really do improve my quality of life significantly. My hope is that conda goes away completely. I run an ML cluster and we have multi-gigabyte conda directories and researchers who can't reproduce anything because just to…

It would be nice indeed if there was a good solution to multi-gigabyte conda directories. Conda has been reproducible in my experience with pinned dependencies in the environment YAML... slow to build, sure, but reproducible.

uv does it by caching versions of packages so they can be shared across projects/environments. So you still have to store those multi-gig directories but you don't have so much duplication. If conda could do something similar that would be great

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#897

Post by maleldil » Thu, Oct 30, 2025, 2:55 PM UTC

lijok wrote: post #854

hexbin010 wrote: post #844

Earlier quoted context omitted.

That's just marketing. Only time will tell. I'll be very happy to be wrong

The time has come. They’re in closed beta with pyx, their first product they’re charging for.

That assumes that pyx (or whatever else they come up with) will be enough to sustain the company.

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#898

Post by insane_dreamer » Thu, Oct 30, 2025, 2:56 PM UTC

miki123211 wrote: post #647

dekhn wrote: post #58

I hadn't paid any attention to rust before uv, but since starting to use uv, I've switched a lot of my performance-sensitive code dev to rust (with interfaces to python). These sorts of improvements really do improve my quality of life significantly. My hope is that conda goes away completely. I run an ML cluster and we have multi-gigabyte conda directories and researchers who can't reproduce anything because just to…

As a person who has successfully used uv for ml workloads, I'm curious what makes you still stay with Conda.

cuda-nvcc was a blocker for us but it looks like a stable version is coming to pypi

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#899

Post by jelder » Thu, Oct 30, 2025, 2:56 PM UTC

> On GitHub Actions, we’re planning to use uv to quickly build a Python environment and run our unit tests. In production, uv already manages Python for all of our servers.

Does that mean they aren't running unit tests _at all_ in CI yet, or they just use a totally different, newer system in production than they do for CI? Either way, brave of them to admit that in public.

Re: Uv is the best thing to happen to the Python ecosystem in a decade

#900

Post by blactuary » Thu, Oct 30, 2025, 2:57 PM UTC

miki123211 wrote: post #647

dekhn wrote: post #58

I hadn't paid any attention to rust before uv, but since starting to use uv, I've switched a lot of my performance-sensitive code dev to rust (with interfaces to python). These sorts of improvements really do improve my quality of life significantly. My hope is that conda goes away completely. I run an ML cluster and we have multi-gigabyte conda directories and researchers who can't reproduce anything because just to…

As a person who has successfully used uv for ml workloads, I'm curious what makes you still stay with Conda.

I stay with it because last time I tried uv it was still directory/project focused vs. environment. With conda it doesn't matter where I am, I can activate any of my environments and run code