📣 mkdocstrings is now supported in Zensical!

We just shipped Zensical 0.0.11 with initial mkdocstrings support. Features like cross-refs and backlinks are still in progress, but API docs already look great with Zensical.

Give it a spin:
zensical.org/docs/setup/e...

Teaching a Billion People to Code: How JupyterLite Is Scaling the Impossible QuantStack, an “almost accidental startup,” is building a serverless distro of JupyterLab for Jupyter’s global adoption.

@quantstack.bsky.social, an “almost accidental startup,” is building a serverless distro of @jupyter.org for Jupyter’s global adoption.

Feat. @sylvaincorlay.bsky.social

YouTube video by JupyterCon Lightning Talk: What's New in Jupyter Frontends - Jeremy Tuloup, QuantStack & Rosio Reyes, Anaconda

Want to catch up on the latest JupyterLab and Jupyter Notebook updates?

This year at JupyterCon 2025 in San Diego, Rosio Reyes and I presented two years of development highlights from the Jupyter Frontends sub-project.

And the recording is now available 🎉

www.youtube.com/watch?v=xiPJ...

This is a problem because of the old adage:
"Debugging is twice as hard as writing the code in the first place."

Complex solutions have a much higher maintenance cost which will limit your ability to support the code base. AI just supercharges the ability to generate tech-debt if... you allow it.

There isn't an AI development session I have where Claude doesn't say "Yes, you're absolutely right! It will be much simpler and more elegant to do it that way!"

Given tests, they can almost always come up with *a* solution to a problem but, like any junior developer, they solve problems by writing more and more code - adding more and more (unnecessary) complexity.

I would phrase that as "can produce excellent code with expert guidance".

My experience is that they're still terrible at designing complex systems.

It does, but they're also impossible guarantees to keep, because bugs.

Therefore everyone claiming to follow SemVer is doing so on a best-efforts basis. What constitutes best-efforts varies wildly between projects, so I agree it's worthwhile for projects to clarify what SemVer means to them.

The only things I take away from SemVer are:

1. Major version bumps will have breaking changes, which may or may not affect you.

2. Patch version bumps *should* not break your code, but sh*t happens, so be prepared regardless.

People get too hung up on SemVer "guarantees".

There are no guarantees. It's just a best efforts basis to convey *some* information to the users of a package.

JupyterLab 4.5 and Notebook 7.5 are available! JupyterLab 4.5 has been released! This new minor release of JupyterLab includes 51 new features and enhancements, 81 bug fixes, 44…

JupyterLab 4.5 and Jupyter Notebook 7.5 are here! 🎉

Highlights 🎁

- Enhanced notebook scrolling behavior
- Native audio and video support
- New Terminal search
- Debugger, Notebook and File Browser improvements

Check out the blog post to learn more!

blog.jupyter.org/jupyterlab-4...

Lots of interesting LLM releases last week. My fav was actually Olmo 3 (I love the Olmo series due to their full open-sourceness and transparency).
If you are interested in reading through the architecture details, I coded it from scratch here: github.com/rasbt/LLMs-f...

A conda package from GIT SOURCE? How?! 🤯 We might have been hinting at "pixi-build", but now we're unveiling what makes it work: pixi build backends. A build backend basically produces a "recipe" from language-specific metadata, and pixi build orchestrates the build.

An idiotic, anti-science decision which will make us all poorer.

Extremely disappointing from the Albanese government.

All you need to do is include a `[tool.pixi.workspace]` table in your `pyproject.toml` and you can use `pixi`.

The only time I fall back to using `uv` is for thirdparty libraries that don't have a `[tool.pixi.workspace]` defined.

To use `pixi` you do need to add (a tiny amount of) extra config.

I'm pretty sure that to solve an environment with wheels requires actually downloading the wheels as not enough metadata is available from PyPI. The packaging experts may have to weigh in on the correctness of that though.

You touched on the "contains all compiled dependecies" in the wheel package which increases the environment size whcih is an important consideration for storage costs, and container pull latency, but they also duplicate all compiled symbols *in memory* so they require more memory to run as well.

Using Pixi as a System Package Manager with Shortcuts and Completions Pixi Global can be used for much more than just downloading and exposing CLI tools. In this blog post, we demonstrate two capabilities of pixi global which are core to making it a featureful and power...

And another one on the amazingly powerful `pixi global` feature:
prefix.dev/blog/using-p...

Conda ≠ PyPI: Why Conda Is More Than a Package Manager | conda.org Part 1 of the 'Conda Is Not PyPI' series—why conda is a multi-language user-space distribution, not just a Python package manager.

There's an excellent series of blog posts on `conda` packages starting with:
conda.org/blog/conda-i...

I use `pixi` to manage PyPI projects. You don't need to duplicate your dependencies if you're ok with using wheels.

`pixi` will automatically map your existing `[project.dependencies]` array to `[tool.pixi.pypi-dependencies]` and will also map `[optional-dependencies]` to `pixi` features.

`uv` is great if all you want is a faster `pip`, but `conda` packages are so much more powerful and versatile and `pixi` really leverages that whilst providing ~everything that `uv` does.... because it includes the `uv` codebase for managing wheels.

The enshittification of GitHub continues apace...

WeatherNext 2: Our most advanced weather forecasting model The new AI model delivers more efficient, more accurate and higher-resolution global weather predictions.

GDM WeatherNext 2

8x faster than v1, it can compute extreme situations and game out scenarios in one minute flat on a single TPU (as opposed to hours of supercomputer time for traditional algorithms)

will be available in all of Google’s weather apps

blog.google/technology/g...

Release v6.0.0 · vega/altair Release v6.0.0 The Vega-Altair team is pleased to announce the release of version 6.0.0. Firstly, we are grateful for the many returning contributors (@franzhaas, @dangotbanned, @dsmedia, @joelostb...

The Vega-Altair team is pleased to announce the release of version 6.0.0. Check out the release at github.com/vega/altair/...

ESOC Report: Implementing Pixi Extensions Hey there! I'm Swastik. I completed a 3-month internship at prefix.dev under the ESoC'25 (European Summer of Code) program, and this blog is all about my internship experience.

Our ESOC (the European version of GSOC) student Swastik Patel blogged about his experience with prefix this summer. Thanks for pushing `pixi` and `pixi-extensions` forward, Swastik 🚀

prefix.dev/blog/esoc-i...

Practical Power: Reproducibility, Automation, and Layering with Conda | conda.org Part 3 of the 'Conda Is Not PyPI' series: how conda enables reproducibility, automation, layered workflows, and rolling distribution.

Practical Power: Reproducibility, Automation, and Layering with Conda

Part 3 of the 3-part series is live! 🚀 Beyond theory into engineering practice: provenance, lockfiles, rolling distribution, and real-world workflows.

#conda #packaging #python #reproducibility
conda.org/blog/conda-p...

[EuroPython 2025 — Terrace 2B on 2025-07-18]🎤 *Unlocking Hidden Power of Conda with Pixi by Ruben Arts, Julian Hofer* 🔗 https://ep2025.europython.eu/sessi... Unlocking Hidden Power of Conda with Pixi — Ruben Arts, Julian Hofer

Missed us at @europython? Here is the recording:

www.youtube.com/watch?v=HOq...

Conda in the Packaging Spectrum: From pip to Docker to Nix | conda.org Part 2 of the 'Conda Is Not PyPI' series, placing conda among pip, Docker, and Nix and explaining its middle-path design.

Conda in the Packaging Spectrum: From pip to Docker to Nix

If conda is a distribution, where does it fit alongside pip, Docker, and Nix? Part 2 of the 3-part series! 🚀

#conda #packaging #python

conda.org/blog/conda-p...

The new dashboard feed.

🚀 Exciting news! The SciPy 2025 Proceedings are officially published:
👉 proceedings.scipy.org/2025

Huge thanks to the Proceedings Committee, @curvenote.com, Jim Weiss, all the authors, and reviewers who made this happen. 🙌