Astral's announcement: https://astral.sh/blog/openai I've really enjoyed Astral's output over the last few years. ruff, uv, and especially ty have been transformative for my experience writing...
Astral has built some of the most widely used open source Python tools, helping developers move faster with modern tooling like uv, Ruff, and ty. These tools power millions of developer workflows and have become part of the foundation of modern Python development. As part of our developer-first philosophy, after closing OpenAI plans to support Astral’s open source products. By bringing Astral’s tooling and engineering expertise to OpenAI, we will accelerate our work on Codex and expand what AI can do across the software development lifecycle.
I've really enjoyed Astral's output over the last few years. ruff, uv, and especially ty have been transformative for my experience writing Python. Astral hit the scene as my frustration was growing towards Python's ecosystem, where the best developer tooling were used as moats for major corporations (pylance + debugpy in vscode, pycharm for Jetbrains, etc...). Astral seemed to offer really compelling alternatives, that ran really fast. And it was a bonus to recognize so many names from the rust and python blogosphere in the contributor list for those project.
I guess I was naive in my hope that Astral would be successful enough with hosted offerings (e.g. pyx) to avoid needing to get acquired.
I don't know that I'd go as far as to say that they've lost my trust. But I am disappointed that their priorities will undoubtedly shift away from those of the developer community towards those of OpenAI. Those two aren't guaranteed to be aligned, so it feels like we're sliding back to a world where the best python tooling is owned by major corporations.
Ultimately these packages are MIT licensed. It's true that in practice, very few open source tools actually get forked and maintained by different developers if the original ones stop, but that is...
Ultimately these packages are MIT licensed.
It's true that in practice, very few open source tools actually get forked and maintained by different developers if the original ones stop, but that is also symptomatic of the underlying problem that everyone wants OSS tools, but no one wants to pay for them, in development hours or money. In which case, it's inevitable that "big companies" rule all the tooling.
Who else is going to if no one but companies wants to foot the bill?
I share your trepidation about OpenAI here, but since Python remains the programming language used in machine learning for now and thus is almost definitely heavily used at OpenAI by their own...
I share your trepidation about OpenAI here, but since Python remains the programming language used in machine learning for now and thus is almost definitely heavily used at OpenAI by their own developers, I'm not completely discarding the possibility that they maintain these tools in ways that are useful to the Python community more generally.
Worst case, uv is hardly the first attempt at giving Python sane package management, and I suspect it wouldn't have been the last even without this OpenAI purchase. I'm confident that if OpenAI does ruin these tools, new things with the same functionality will arise to replace them.
For whatever it's worth, I think that's the steady state. I actually tend to gravitate to tools owned by big companies; not because I like it, but because then I get to choose my corporate...
...sliding back to a world where the best python tooling is owned by major corporations.
For whatever it's worth, I think that's the steady state. I actually tend to gravitate to tools owned by big companies; not because I like it, but because then I get to choose my corporate overlord. Tools made by startups will almost inevitably get acquired, you just don't know by whom until it happens.
Astral's announcement: https://astral.sh/blog/openai
I've really enjoyed Astral's output over the last few years.
ruff,uv, and especiallytyhave been transformative for my experience writing Python. Astral hit the scene as my frustration was growing towards Python's ecosystem, where the best developer tooling were used as moats for major corporations (pylance + debugpy in vscode, pycharm for Jetbrains, etc...). Astral seemed to offer really compelling alternatives, that ran really fast. And it was a bonus to recognize so many names from the rust and python blogosphere in the contributor list for those project.I guess I was naive in my hope that Astral would be successful enough with hosted offerings (e.g. pyx) to avoid needing to get acquired.
I don't know that I'd go as far as to say that they've lost my trust. But I am disappointed that their priorities will undoubtedly shift away from those of the developer community towards those of OpenAI. Those two aren't guaranteed to be aligned, so it feels like we're sliding back to a world where the best python tooling is owned by major corporations.
Ultimately these packages are MIT licensed.
It's true that in practice, very few open source tools actually get forked and maintained by different developers if the original ones stop, but that is also symptomatic of the underlying problem that everyone wants OSS tools, but no one wants to pay for them, in development hours or money. In which case, it's inevitable that "big companies" rule all the tooling.
Who else is going to if no one but companies wants to foot the bill?
Perhaps maintaining forks will become cheaper than it has been, now that we have coding agents.
I share your trepidation about OpenAI here, but since Python remains the programming language used in machine learning for now and thus is almost definitely heavily used at OpenAI by their own developers, I'm not completely discarding the possibility that they maintain these tools in ways that are useful to the Python community more generally.
Worst case, uv is hardly the first attempt at giving Python sane package management, and I suspect it wouldn't have been the last even without this OpenAI purchase. I'm confident that if OpenAI does ruin these tools, new things with the same functionality will arise to replace them.
For whatever it's worth, I think that's the steady state. I actually tend to gravitate to tools owned by big companies; not because I like it, but because then I get to choose my corporate overlord. Tools made by startups will almost inevitably get acquired, you just don't know by whom until it happens.