Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
EmbeddingGemma 2: an open, lightweight multimodal embedding model
Google launches EmbeddingGemma 2, a 740M-parameter Apache 2.0 embedding model built on Gemma 4 that puts text, code, images, video and audio in one embedding space, with modular encoders, Matryoshka truncation, an 8K context window and on-device RAM use as low as about 191MB for text-only weights.
3 min · 668 words
Introducing the Personal Agent Consent & Trust Protocol (PACT)
Decagon open-sources PACT, co-developed with Instinct, so businesses can verify which personal AI agent is acting for a customer and what that customer authorized — built on Agent2Agent and OAuth 2.0 — and joins the Personal Agent Protocol working group around Meta’s Muse.
4 min · 910 words
Last rites for Gentoo's Chromium package
LWN covers Gentoo's decision to last-rite its Chromium package: maintenance burden, toolchain and sandbox complexity, and what the end of a distro-packaged Chromium means for Linux users who relied on it.
58 min · 13,293 words
Command injection in Mage AI's git config
Yunus Aydın details an OS command injection in Mage AI's git utility code (shell=True with unsanitised config values), with proof of concept, and recounts the 26 months between his April 2024 report and the June 2026 fix merging.
5 min · 1,109 words
The Slow Formation of Durable SoftwareHow a group of historians took years to imagine Zotero, the research application that would eventually be used by millions
Today, software can be created more or less instantly with AI, for a large user base or for yourself, for any purpose or for no serious purpose at all. As Robin Sloan enticingly puts it, just “ask for what you want.” The genesis of Zotero, the “free, easy-to-use tool to help you collect, organize, annotate, cite, and share research,” used by over 20 million people in dozens of languages and countless disciplines, is the opposite of this instant gratification. It took five…
15 min · 3,481 words
OpenWAM: An Open Framework for Composable World-Action Models
Stanford researchers introduce OpenWAM, an open framework for world-action models in robotics. It adapts Wan2.2-5B on 14.64k hours of robot and human video, pairs it with a 2B action expert, compares video-then-action, action-then-video, joint and decoupled programs, and adds local-context inverse and forward dynamics models that can be composed across independently trained parts.
9 min · 1,990 words
Raspberry Pi Desktop now available for PC and Mac
Simon Long announces an updated Debian Trixie with Raspberry Pi Desktop image for PCs and Macs, including a Raspberry Pi Connect client, and recounts how the x86 version began as a 2016 side project after Eben Upton asked whether the desktop could run on his Mac.
4 min · 1,020 words
Burn 0.22.0: Faster Builds, Easier Extensions, and Smarter Autotuning
Tracel's Burn 0.22 deep learning framework for Rust removes backend type parameters from user code, selecting execution through the device, which made rebuilds up to 15x faster in their tests. It also adds adaptive memory management, smarter autotuning, GPU graph replay, a reusable CubeCL Environment cache, LoRA and QLoRA fine-tuning, ONNX export and expanded remote compute.
12 min · 2,837 words
A Terminal Protocol for Program Status (OSC 7501)
Mitchell Hashimoto introduces OSC 7501, a generic terminal escape sequence that lets programs report whether they are idle, working, blocked on the user, finished or failed, and why. He argues heuristics and non-terminal APIs fall short, especially for agentic inboxes tracking many coding agents, and describes implementations in libghostty and Rex.
5 min · 1,236 words
tapo speaks TPAP: Third-Party Compatibility can stay off
Mihai Dinculescu explains how v0.11.1 of tapo, his unofficial Rust and Python client for TP-Link Tapo devices, now logs in over TP-Link's TPAP protocol so users no longer need the Third-Party Compatibility switch, and covers new H200/H500 camera hub recording downloads plus plug schedules and timers.
10 min · 2,290 words
The Polars team ships Polars 2.0: initial out-of-core spill-to-disk support on by default, core engine and optimizer speedups, first-class SQL that leads DuckDB and DataFusion on TPC-H and TPC-DS style benchmarks, a new Map dtype, and stricter dtype handling, plus a detailed benchmark appendix.
6 min · 1,287 words
Mistral launches a public API preview of Mistral Large 4 ("le Chonk"), a 1-trillion-parameter natively multimodal mixture-of-experts model with 49B active parameters, pitched as the strongest open-weight model from the US or Europe, with weights due at the end of the month and training funded by its €3B Series D.
10 min · 2,203 words
I hired an illustrator to draw my house. Now it's my Home Assistant dashboard
Anton Frolov describes commissioning hand-drawn day and night illustrations of each floor of his house, with every device on its own layer and a shared canvas, then wiring them into a Home Assistant picture-elements dashboard in YAML so lights, fireplace and air conditioners animate on a wall TV the whole family actually uses.
10 min · 2,239 words
Open Source as We Know It Is Dead
James Ross looks back on 15 years in open source and argues that AI is killing it as he has known it: AI-generated pull requests made contributing cheap but maintaining far more expensive, so the projects he loves are closing their doors to outside contributors, and he admits he has no neat fix.
9 min · 1,974 words
mold 3.0.0 is the first release of the high-speed linker after its rewrite from C++ to Rust: a drop-in replacement for 2.42.1 with the same options, targets, and linking performance, bounds-checked reads on corrupted input, a Cargo-based build, and a long list of bug fixes and compatibility improvements across architectures.
5 min · 1,254 words
Property Testing with Agent Swarms
Inanna Malick shows how to have coding agents build property-testing machinery (reference models, generators and assertions) for a repo, then turn the failures into small, reviewable fixes. Pointed at OpenAI's Codex, the public playbook found thirteen distinct correctness bugs, and runs on jj, uv, Prometheus and Babel produced merged upstream fixes.
6 min · 1,325 words
TidesDB now available for MySQL v9, v26
The TidesDB team announces TideSQL-MySQL v2.0.0, which loads the TidesDB LSM-tree storage engine into stock MySQL 9.7 and 26.7 as a plugin rather than a fork. The post covers per-table options via ENGINE_ATTRIBUTE, compression, value separation for large values, durability modes, TTL, encryption, optimistic MVCC, and sysbench comparisons against InnoDB.
7 min · 1,580 words
OpenSSH Ships on Every Mac, Linux Server and Windows PC. Its Creator Trusts No One.
A long-form profile of Theo de Raadt, from his exit from NetBSD and lost DARPA grant to founding OpenBSD and OpenSSH, framed by the 2024 xz backdoor, and how his famously paranoid, combative approach to security came to protect remote access on most of the world's servers.
29 min · 6,772 words
Gleam doesn't compile to Erlang source anymore
The Gleam v1.19.0 release notes explain that the compiler's Erlang backend was rewritten to emit Erlang abstract forms instead of Erlang source, cutting build times and giving accurate source locations, alongside JavaScript decision-tree and list-literal optimisations, TypeScript overloads, language server label support, and better error messages.
12 min · 2,663 words
Introducing Beam: Reflection’s 501B open-weight model
Reflection introduces Beam, its first open-weight model: a sparse Mixture-of-Experts with 501 billion total and 23 billion active parameters, pretrained on 23.8 trillion tokens and trained with large-scale reinforcement learning for coding and agentic work, with Apache 2.0 weights, a technical report, and a model card promised later this month.
12 min · 2,858 words