What happened to Poetry? did it just not gain enough traction? I was under the impression it was the hot new thing a few years back then uv comes out of nowhere.
I thought the same thing. Just a few years ago, everyone here was proclaiming Poetry was great, the python ecosystem was finally tamed, pip/conda/setuptools was dead, and every project and developer needs to adopt it.
Now it’s just a has-been. The churn in python is incredible.
Yeah. There was a wave of "rebuild it in Rust" companies and my complaint with all of them is that they became largely from-spec reimplementations of existing tools with little imagination of their own. The Rust projects are a fraction of the age of the scripted ones and they already feel ossified.
I'm not just talking about Astral here, I'm talking about the JS equivalents too like Biome and VoidZero which basically took the existing ossified architectures and made them gospel -_-
That's not a good interpretation of Astral and uv.
uv is way more than a rewrite in Rust of Poetry or pip - it has a metric ton of smart design decisions that are independent of the language it was written in, both in terms of what it does and how it speeds things up under the hood.
Not to gatekeep things but if you think uv solved python dependency issues then you probably never had those issues in the first place and pip would have been enough for your use case. Conda on the other hand, with external binary dependencies, now we're talking.
I’m someone who is not a python developer but has to use python tools and run other people’s python code. I have suffered through learning about anaconda, virtualenv, pip, and more. Uv is the first time there’s a tool that just runs the software without requiring me to become a python ecosystem expert
The key issue uv solved wasn't dependencies, it was environments.
I used to have hundreds of venv folders scattered around my machine. These days I use "uv run" or "uvx" or "uv run --with boto3 python" and uv handles all of the bookkeeping for me.
> The key issue uv solved wasn't dependencies, it was environments.
and not specific, niche features.
I don't know of such a command in Poetry, but it could still be solved with dev groups or running Python in the virtual environment and installing the package.
what is stopping the poetry guys to rewrite poetry from the ground up as part of a breaking new release?
I liked Poetry until I went to contribute a bugfix to the codebase and quickly realized it was already an ossified nightmare.
What happened to Poetry? did it just not gain enough traction? I was under the impression it was the hot new thing a few years back then uv comes out of nowhere.
I thought the same thing. Just a few years ago, everyone here was proclaiming Poetry was great, the python ecosystem was finally tamed, pip/conda/setuptools was dead, and every project and developer needs to adopt it.
Now it’s just a has-been. The churn in python is incredible.
It's almost like you're talking about JavaScript. I use Python here and there but wasn't aware of this kind of thrash
The news of this acquisition has made me too scared to admit that I actually have no issues Poetry.
uv is much much faster, but has substantially the same feature set as poetry.
Yeah. There was a wave of "rebuild it in Rust" companies and my complaint with all of them is that they became largely from-spec reimplementations of existing tools with little imagination of their own. The Rust projects are a fraction of the age of the scripted ones and they already feel ossified.
I'm not just talking about Astral here, I'm talking about the JS equivalents too like Biome and VoidZero which basically took the existing ossified architectures and made them gospel -_-
That's not a good interpretation of Astral and uv.
uv is way more than a rewrite in Rust of Poetry or pip - it has a metric ton of smart design decisions that are independent of the language it was written in, both in terms of what it does and how it speeds things up under the hood.
https://nesbitt.io/2025/12/26/how-uv-got-so-fast.html is a useful review of some of the architectural tricks they came up with.
they have bad taste obviously otherwise we'd all be using their project
Yes, when the poetry people purposely added a feature to fail CI with a 1/10 chance because they wanted to depreciate a feature, I depreciated poetry.
Not to gatekeep things but if you think uv solved python dependency issues then you probably never had those issues in the first place and pip would have been enough for your use case. Conda on the other hand, with external binary dependencies, now we're talking.
I’m someone who is not a python developer but has to use python tools and run other people’s python code. I have suffered through learning about anaconda, virtualenv, pip, and more. Uv is the first time there’s a tool that just runs the software without requiring me to become a python ecosystem expert
The key issue uv solved wasn't dependencies, it was environments.
I used to have hundreds of venv folders scattered around my machine. These days I use "uv run" or "uvx" or "uv run --with boto3 python" and uv handles all of the bookkeeping for me.
Poetry had already solved that.
What's the poetry equivalent of this?
The claim was:
> The key issue uv solved wasn't dependencies, it was environments.
and not specific, niche features.
I don't know of such a command in Poetry, but it could still be solved with dev groups or running Python in the virtual environment and installing the package.
That command is an environment command. It runs Python in a temporary environment with boto3 installed - I use that trick all the time.
It was very slow compared to uv
Yes, but it still solves dependencies and environments.
"Solves one problem by introducing another, possibly worse problem"
And you wonder why people didn't adopt it.
Well it is much faster so you can keep iterating installing dependencies until you fix the issue in much less time
Discussion:
https://news.ycombinator.com/item?id=47438723
[flagged]
[dead]