Fatbobman

Hacker News

JavaScript Weekly

Kent C. Dodds

Laravel News

<h2>Must know</h2>
<ul>
<li><strong>Systemd 260 introduces AI coding guidelines and controversy</strong>: The latest Systemd release officially drops support for legacy System V init scripts and, more controversially, includes an <a href="http://AGENTS.md" target="_blank" rel="noopener nofollow">AGENTS.md</a> file to guide AI coding agents. After a submodule commit was partially written with Claude’s help, the project was placed on the ‘OpenSlopware’ list, leading to new mandatory disclosures for AI-generated code. Systemd now requires human review for all AI-assisted commits and has bumped the minimum kernel requirement to 5.10+. <a href="https://app.daily.dev/posts/ynJvFeH96" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GNOME 50 goes Wayland only and lands in Ubuntu 26.04</strong>: GNOME 50, codenamed ‘Tokyo’, has officially launched, marking the complete removal of X11 support in favor of a Wayland-only architecture. The update graduates fractional scaling and variable refresh rate (VRR) to stable status and significantly improves NVIDIA GPU performance via enhanced frame scheduling. It is scheduled to be the default desktop environment for the upcoming Ubuntu 26.04 LTS and Fedora 44 releases. <a href="https://app.daily.dev/posts/Xg0Exf55t" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wasmer Edge.js brings sandboxed Node.js to WebAssembly</strong>: Wasmer has released Edge.js, a JavaScript runtime that allows unmodified Node.js workloads to run inside a WebAssembly sandbox using WASIX. Unlike Deno, it aims for 100% Node.js v24 API compatibility by sandboxing system calls rather than the JS engine itself, though it currently sees a 5–30% performance hit. The project notably used OpenAI Codex to accelerate its development from years to weeks. <a href="https://app.daily.dev/feed-by-ids?id=qHzldup0J&amp;id=FUA9hYuPV" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>DBStan for Laravel analyzes database schemas for performance and integrity issues</strong>: A new package for Laravel 9–12 that detects missing indexes, normalization problems, and foreign key inconsistencies via the CLI. <a href="https://app.daily.dev/posts/K5sWdO7ie" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>JDK 26 arrives with G1 GC optimizations and HTTP/3 support</strong>: The non-LTS release includes JEP 522 for reduced GC overhead and JEP 517 for native QUIC support. <a href="https://app.daily.dev/feed-by-ids?id=26rrhGWJk&amp;id=QPwbRCIfz" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Lens for Laravel integrates WCAG accessibility auditing into local development</strong>: This dev-only tool uses Axe-core to map accessibility violations directly to Blade template line numbers with optional AI fix suggestions. <a href="https://app.daily.dev/posts/qRXXbQg9S" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Zeroboot achieves sub-millisecond VM forking for AI sandboxes</strong>: Using Firecracker and KVM, this project forks pre-snapshotted VMs in 0.8ms, significantly faster than E2B or Daytona. <a href="https://app.daily.dev/posts/0KZTYS81s" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Fedora Asahi Remix 43 brings Linux to M1 and M2 Macs</strong>: The new release supports 120Hz refresh rates on MacBook Pros and introduces microphone support for M2 Pro/Max models. <a href="https://app.daily.dev/posts/585WUMr9y" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Bun v1.3.11 adds native OS-level cron job scheduling</strong>: The update introduces Bun.cron for cross-platform task scheduling and fixes over 100 bugs, including Windows ARM64 shims. <a href="https://app.daily.dev/posts/DuG54Q8tB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Perfex CRM hit by unauthenticated RCE via insecure deserialization</strong>: A critical flaw in versions &lt;=3.4.0 allows attackers to bypass XSS filters using hex-escaped null bytes to execute arbitrary PHP code. <a href="https://app.daily.dev/posts/ksvDreRwI" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>The superintelligence pivot as a distraction for Copilot failure</strong>: Microsoft is moving Mustafa Suleyman to a ‘superintelligence’ initiative while Jacob Andreou takes over Copilot, which has reportedly struggled to make any market impact despite being shoved into every corner of Windows. It’s the classic tech shuffle: when your current AI product isn’t moving the needle, just rename the department to something more sci-fi and hope nobody notices the billions in Reality Labs-style losses. Apparently, ‘superintelligence’ is the new ‘metaverse’ for burying underwhelming KPIs. <a href="https://app.daily.dev/posts/jTMZOA2Fr" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Banning AI on Fridays to solve the hallucination problem</strong>: A Gartner analyst is seriously suggesting that companies should ban Microsoft Copilot on Friday afternoons because ‘tired workers’ are too lazy to check for AI errors. Instead of addressing the fact that these models are confidently wrong by design, the solution is apparently to treat LLMs like heavy machinery that you shouldn’t operate while drowsy. If your enterprise tool is so unreliable that it requires a ‘no-fly zone’ based on the day of the week, maybe the tool is the problem, not the calendar. <a href="https://app.daily.dev/posts/6OISLpOWY" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>The $200,000 hackathon to find the AGI goalposts</strong>: Google DeepMind is launching a Kaggle hackathon to ‘measure progress toward AGI’ by asking the community to build benchmarks for things the models currently suck at, like metacognition and social logic. They’re calling it ‘meatbags v machines’ to drum up hype, but it’s really just a way to get cheap labor to define the metrics that will inevitably be used to claim AGI is ‘just six months away’ for the next decade. It’s not a competition; it’s a request for someone to please explain why the chatbots still can’t do basic reasoning. <a href="https://app.daily.dev/posts/NhVe5iupm" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
