Samsung Magician disk utility takes 18 steps and two reboots to uninstall
(chalmovsky.com)
A writer reports that removing Samsung Magician on macOS takes an unusually complex process, including a cleanup script, extensive manual file deletion, and two reboots into Recovery Mode to temporarily disable and re-enable SIP. The post says the built-in cleanup script fails to remove components due to repeated chown errors, leaving multiple Samsung files scattered across system areas such as kernel extensions and staged extensions. The author also criticizes the software’s bundled bloat, including animation assets, an embedded Electron framework, and ads.
Big-Endian Testing with QEMU
(hanshq.net)
The article explains endianness and shows how to test byte-order-dependent code using QEMU’s user-mode emulation by cross-compiling a small C program for big-endian targets like MIPS and IBM z/Architecture.
Marc Andreessen is wrong about introspection
(joanwestenberg.com)
Joan Westenberg argues that Marc Andreessen’s claim that introspection was invented by Freud and the Vienna Circle is historically and philosophically wrong, citing examples from ancient Greek, Roman, Christian, Chinese, and literary sources. The piece also contends that dismissing introspection leads to a thin model of human flourishing—one that can optimize measurable outputs like engagement while missing what people actually find meaningful. Westenberg concludes that Andreessen’s “move forward, go” mindset is direction without a clear account of what people want, which requires examining inner life.
Reverse Engineering Crazy Taxi, Part 2
(wretched.computer)
The post continues a series on reverse engineering Crazy Taxi assets, focusing on how the game’s .shp “shape” files likely store 3D model geometry. Using the small cube0.shp as a “Rosetta Stone,” the author compares the file’s structure to common 3D concepts like vertices, normals, and indexed triangle faces as seen in formats such as Wavefront .obj, then begins inspecting cube0.shp in a hex editor to identify sections and data types. The article outlines a methodical approach—take detailed notes and validate theories with code—while mapping the first steps toward decoding the proprietary .shp format.
Salomi, a research repo on extreme low-bit transformer quantization
(github.com)
AI
Salomi is a GitHub research repo exploring extreme low-bit (near-binary) transformer quantization and inference for GPT-2–class models, with code, experiments, and evaluation tooling. It specifically tests whether strict 1.00 bpp post-hoc binary quantization can match or beat higher quantization baselines and concludes it does not hold up under rigorous evaluation. The repo instead reports more credible results around ~1.2–1.35 bpp using methods such as Hessian-guided vector quantization, mixed precision, and magnitude-recovery, and directs readers to curated assessment and validation documents over older drafts.
Wikidata: WikiProject Ontology/Ontology Course
(wikidata.org)
Wikidata’s WikiProject Ontology posts details for an eight-week “Wikidata Ontology” course, including eligibility, required exercises, and a project aimed at improving and expanding the Wikidata ontology. The second offering is scheduled for May–June 2026 on Thursday afternoons (1–3 pm EDT) with registration open until 27 April. The page also outlines course topics such as ontology fundamentals, editing and querying with SPARQL, inference and constraints, and includes information from the prior 2025 offering.
If you read cursive, the Newberry has a job for you
(lithub.com)
The Newberry Library in Chicago is recruiting volunteer transcribers to decipher handwritten materials in its collections, as part of an ongoing project that began with Civil War letters in 2013. The effort has expanded across the archive, but fewer volunteers are available as cursive is dropped from Common Core and digitization work continues to grow. The article highlights the breadth of what’s at stake—from correspondence of major Chicago writers to firsthand accounts of major events—urging readers to join the Newberry’s “Transcribe” program.
Your sign-up form is a weapon
(bytemash.net)
Bytemash reports that it saw low-volume bot sign-ups using real victims’ email addresses and then immediately triggering password resets to generate multiple unwanted emails (“subscription bombing”). The attack stayed under typical rate limits and relied on forms that allow email entry without verification, aiming to overwhelm inboxes and bury account-security alerts. The site says it mitigated the issue by tightening bot/firewall rules, deploying Cloudflare Turnstile via its auth stack, and changing its email flow to send only a verification email until the address is confirmed.
Show HN: Mkdnsite – Markdown-native web server for humans (HTML) and agents (md)
(github.com)
AI
Mkdnsite is an open-source “Markdown-native” web server that serves a directory or GitHub repo of .md files without a static-site build step. It renders HTML for browsers and uses HTTP content negotiation to return raw Markdown for AI agents (e.g., via Accept: text/markdown), along with an auto-generated /llms.txt and an optional MCP endpoint. The project supports Bun/Node/Deno, runtime editing without redeploy, and includes features like search, theming, math (KaTeX), Mermaid, and syntax highlighting.
Show HN: Semantic atlas of 188 constitutions in 3D (30k articles, embeddings)
(constitutionalmap.ai)
AI
Constitutional Map AI is a web tool that builds a 3D semantic atlas of constitutional law by embedding thousands of constitutional articles from 188 constitutions. It clusters the text into thematic “neighborhoods” and lets users compare countries on a shared semantic space using keyword or semantic search, with metrics like coverage and entropy. The site’s data is sourced from the Constitute Project and the code is open source, with a note that AI clustering or segmentation errors are possible.
This Month in Ladybird – March 2026
(ladybird.org)
Ladybird’s March 2026 update reports major progress across the project, including a rewritten Rust regex engine, performance gains to the JavaScript engine, and persistent bookmarks. The team also implemented parts of Media Source Extensions (VP9/WebM and Opus/audio), added new CSS parsing/support features, and parallelized HTML script parsing. In infrastructure work, it removed the legacy LibJS compiler pipeline and overhauled macOS IPC to use Mach ports, while noting continued improvements to Web Platform Tests.
Why Iran targeted Amazon data centers
(theconversation.com)
The article explains that Iranian Shahed drone strikes on Amazon Web Services data centers in the UAE in March 2026 were a first in deliberately attacking commercial cloud infrastructure, aiming to disrupt both military-relevant systems and wider society. It notes that while data centers increasingly support AI and government services, the attacks may also be about pressuring the UAE and U.S.-Gulf ties rather than signaling a wholly new form of warfare. Overall, it argues that data centers are likely to remain recurring targets as dependence on cloud and AI grows.
What Gödel Discovered (2020)
(stopa.io)
The article explains, in a programmer-friendly way, how Kurt Gödel’s 1931 work showed that any sufficiently powerful formal system for arithmetic (like Russell and Whitehead’s “Principia Mathematica”) cannot be both complete and consistent. It walks through the motivation from earlier foundational efforts and Russell’s paradox, then sketches Gödel’s method of encoding logic and proofs into numbers using Gödel numbering, leading to unprovable statements. The overall point is that mathematics cannot be fully captured by a single, purely mechanical set of rules.
The Anti-Intellectualism of the Silicon Valley Elite
(thenation.com)
AI
The article argues that Silicon Valley’s top figures—citing figures like Peter Thiel and Marc Andreessen—promote an anti-intellectual worldview that treats deep learning as unnecessary, even while profiting from it. It links this stance to attacks on higher education and the humanities, skepticism toward inquiry that could challenge the managerial class, and a broader desire for insulation from accountability. The piece also criticizes how AI and tech “shortcuts” can be used to replace thinking, while the same elite dismisses the people and disciplines that make that knowledge possible.
AbodeLLM – An offline AI assistant for Android devices, based on open models
(github.com)
AI
AbodeLLM is an Android app that runs an offline AI assistant using open-source models such as LLaMA and DeepSeek, with chat processed entirely on-device and no internet required. It supports optional multimodal inputs (vision and audio depending on models), context retention, and an “Expert Mode” for tuning generation and cache/token limits. The project includes installation steps and a list of supported model variants along with minimum hardware requirements.
The Claude Code Leak
(build.ms)
AI
An article argues that the alleged leak of Claude Code’s source code matters less than the broader lessons it highlights: product-market fit and seamless model-to-agent integration outweigh the quality or even the cleanliness of the underlying code. The writer also discusses how the code appears to be “bad” yet still supports a valuable product, why observability and automation may be more important than implementation details, and how the ensuing DMCA and clean-room rewrites reflect ongoing copyright tensions in AI development.
Steam on Linux Use Skyrocketed Above 5% in March
(phoronix.com)
Phoronix reports that Valve’s Steam Survey shows Linux reaching 5.33% market share for March 2026, surpassing 5% and more than doubling Steam’s share on macOS. Windows share fell by 4.28% to 92.33%, with Valve attributing part of the change to corrections to Steam China language data. The survey also suggests roughly a quarter of Linux Steam users run SteamOS, and that AMD CPUs account for just under 70% of Steam on Linux.
Escaping the Ogallala Trap
(worksinprogress.co)
The article argues that widespread use of autonomous vehicles could recreate the “tragedy of the commons” seen with the Ogallala Aquifer: roads will be overused if they are effectively free at the point of use. It reviews how new technologies historically contributed to resource depletion, then proposes road pricing approaches that could become politically durable, such as charging only autonomous vehicles and pairing new fees with new infrastructure. The author says acting before AVs dominate is key to preventing gridlock.
Artemis II's toilet is a moon mission milestone
(scientificamerican.com)
Scientific American reports that NASA’s Artemis II will carry its first fully functional spacecraft toilet, the Universal Waste Management System, designed to work reliably in microgravity. Unlike Apollo’s bag-and-tube approach, the system includes a door, improved stability controls, and hardware that can handle urine and feces together while accommodating male and female astronauts. The technology has been tested on the International Space Station and is intended to inform future Artemis and Mars waste-management systems.
Trinity Large Thinking
(openrouter.ai)
AI
OpenRouter lists Arcee AI’s open-source “Trinity Large Thinking” model and its pricing on the platform, including per-token input/output costs and usage statistics. The page explains how OpenRouter routes requests to multiple providers with fallbacks to improve uptime, and how to enable reasoning output via a request parameter and the returned reasoning_details.
Solar Balconies Take Europe by Storm
(hackaday.com)
The article explains how “balcony solar” (German “balkonkraftwerk”) lets people in apartments or renters generate electricity with small, plug-in solar kits using DC-to-AC inverters. It describes Germany’s growth—now over a million installations—along with updated feed-in limits (800W, up to 2000W peak panels) and how similar regulatory approaches are spreading across parts of Europe and to some US states. While output is modest and mostly used for household loads, the low cost and DIY setup are driving broader adoption.
All Travelers Will Need Clear Carry-On Bags Starting This Summer
(upgradedpoints.com)
The TSA announced a Transparent Screening Initiative that will require passengers to use clear carry-on bags at security checkpoints this summer at several major airports, with warnings rather than penalties during an initial adjustment period. The agency says the move is intended to reduce manual bag checks and improve checkpoint efficiency. The pilot will expand more broadly in 2027, and the article notes retailers and luggage brands are preparing clear-bag options.