Super Monkey Ball ported to a website
Summary
Super Monkey Ball 1 web version offers three core play modes—Standard, Challenge, and Story—each with multiple difficulty tiers. Standard mode provides Beginner, Advanced, and Expert levels, plus “Extra” variants (Beginner Extra, Advanced Extra, Expert Extra) and a Master stage. Challenge mode mirrors this structure and adds Master Extra and Challenge stages. Story mode includes Story World and Story Stage selections. Audio controls feature independent volume sliders: Music at 50 %, SFX at 30 %, and Announcer at 30 %. Gameplay controls are mapped to keyboard: WASD or arrow keys tilt the ball, “R” resets the stage, “N” skips the stage; a connected controller is also supported. The site notes that bug reports are unnecessary, implying known issues are already being addressed.
Read full article →
Community Discussion
The comments express overall enthusiasm for the new version, noting its appealing look and feel while also missing elements from the original, such as the “woop woop” sound and the monkey in the ball. Users perceive the controls as slightly more sensitive and request additional features like phone gyro support and implementation details. A few report technical glitches on mobile browsers, particularly on a Pixel 7, but the majority view the work as solid and impressive, balancing praise with constructive feedback.
Prism
Community Discussion
Comments show a mixed reaction to the new AI‑enhanced scientific‑writing platform. Many express skepticism, fearing that easy AI generation will flood journals with low‑quality, hallucinated content and increase the burden on already stretched reviewers, while also worrying about data collection and restrictive licensing. Some appreciate the collaborative features and see potential for faster, non‑substantive edits, yet note that similar capabilities already exist through Overleaf, local toolchains, or other AI services. A few view the OpenAI acquisition as positive for stability, but overall sentiment leans toward doubt about the tool’s necessity and its impact on scientific rigor.
430k-year-old well-preserved wooden tools are the oldest ever found
Community Discussion
The comments express widespread surprise at the exceptionally early dating of wooden tools, noting personal gaps in knowledge and appreciation for new archaeological evidence. Readers highlight the challenges of submerged sites, the longstanding use of materials like obsidian, and the durability of wood in construction, while also questioning the identification of the artifacts and criticizing media phrasing that may imply professional error. Humor and speculative remarks appear alongside calls for cautious interpretation, reflecting a mix of curiosity, skepticism, and interest in broader implications for human evolution.
Rust’s Standard Library on the GPU
Summary
VectorWare has built a framework that lets GPU kernels call Rust’s standard library (`std`) instead of being limited to `#![no_std]`. The system implements many `std` APIs (e.g., `println!`, file I/O, time functions) via a “hostcall” mechanism: the GPU issues structured requests that are serviced either on the host CPU or directly on the device when hardware support exists (e.g., CUDA global timer for `Instant`). Hostcalls are routed through a libc‑style façade to minimize changes to the existing library, while preserving the usual Rust API surface for developers.
Key technical details:
- Uses double‑buffering, atomic operations, and CUDA streams to avoid blocking and ensure memory consistency.
- Protocol is vendor‑agnostic; current implementation targets Linux hosts with NVIDIA GPUs (CUDA), but could be extended to AMD/HIP or Vulkan.
- Enables advanced features such as device‑side caching of hostcall results and virtualized filesystem views.
- Aligns with emerging GPU capabilities (GPUDirect Storage/RDMA) and the convergence of CPU/GPU architectures.
VectorWare plans to open‑source the code, work on upstreaming to Rust, and explore whether the abstraction layer should remain libc‑based or become a native GPU‑aware `std` extension.
Read full article →
Community Discussion
The remarks express skepticism that the title overstates the capability, noting the standard library runs on the CPU via remote calls rather than directly on the GPU. They seek clarification on how data exchange between CPU and GPU is managed without excessive overhead and question whether typical I/O code would execute entirely on the CPU. There is interest in broader goals of writing ordinary Rust code that transparently targets SIMD or GPU hardware, with references to related projects such as rust‑gpu, CubeCL, and Rayon, while probing the feasibility of running more complex programs on the GPU.
A few random notes from Claude coding quite a bit last few weeks
Summary
The page consists of a generic header (“Title: [no‑title]”) followed by a separator line and an error notice stating that something went wrong and advising a retry. Another separator introduces an “Images and Visual Content” section, which contains a single entry: “Image 1” with an alt‑text label of a warning symbol (⚠️). No additional text, data, or functional elements are present. The structure is limited to the title, error message, section heading, and the placeholder for one image with a warning icon as its descriptive text.
Read full article →
Community Discussion
Comments express a mixed view of LLM‑assisted coding. Many note the agents’ tireless, high‑stamina work and the productivity gains they can deliver, especially for front‑end or incremental tasks, while also highlighting the need for new workflows, education, and continued IDE‑based review. A recurring concern is skill atrophy and complacency, with users fearing dependence on opaque models, loss of manual coding ability, and potential societal effects from gatekeeper control. Opinions diverge on whether the technology will amplify top engineers’ output or create a split between “builders” and traditional coders, and adoption is seen as uneven pending cultural and tooling changes.
Time Station Emulator
Summary
GitHub repository **kangtastic/timestation** provides software to synchronize radio‑controlled (“atomic”) clocks and watches using a smartphone or tablet. The project is open‑source under the MIT license. Build and test pipelines are shown via GitHub Actions workflow status and Codecov coverage badges. Implementation combines native code in **C** with web components built using **Lit**, typed with **TypeScript**, and compiled to **WebAssembly** for performance. The repository includes standard project assets such as a logo and licensing information. No additional documentation or usage instructions are present in the scraped content.
Read full article →
Community Discussion
The comments express interest in the concept of generating low‑power radio signals from audio hardware, noting past personal experiments and theoretical possibilities. Enthusiasm is balanced by doubts about practical effectiveness, especially given weak signals, limited demo evidence, and device restrictions such as iPhone browser constraints. Several remarks appreciate the novelty and soothing aspects of received time‑signal tones, while others voice concern over potential interference with clocks and question the claim of universal phone compatibility.
Lennart Poettering, Christian Brauner founded a new company
Summary
Amutable is a foundation focused on providing verifiable integrity for Linux workloads across all environments. Its core mission emphasizes ensuring that Linux-based applications and services can be trusted through measurable integrity guarantees. The organization highlights a team of notable contributors and experts, including Christian Brauner, Chris Kühl, Lennart Poettering, David Strauss, Aleksa Sarai, Daan De Meyer, Joaquim Rocha, Kai Lüke, Michael Vogt, Rodrigo Campos Catelin, and Zbyszek Jędrzejewski‑Szmek. These individuals are presented via image alt text, suggesting their involvement or endorsement of the project. The site’s primary message centers on delivering consistent, verifiable security for Linux workloads worldwide.
Read full article →
Community Discussion
Comments show strong concern that remote attestation and cryptographic integrity could erode user privacy, enable DRM‑like restrictions, and increase corporate control over Linux systems, especially given systemd’s history. Many question the business model, funding sources, and potential mandatory integration, while calling for optionality, transparency logs, and reproducible builds. A minority express interest in legitimate security uses such as server verification and appreciate the focus on immutable, verifiable systems. Overall, skepticism dominates, with calls for safeguards, openness, and clear purpose before adoption.
Doing the thing is doing the thing
Summary
The post presents a concise manifesto on execution, emphasizing that any form of preparation, discussion, or anticipation—such as thinking, dreaming, visualizing success, waiting for readiness, talking, explaining, debating online, announcing intentions, consuming related media, studying tutorials, reading others’ experiences, planning systems, purchasing tools, reorganizing workspaces, feeling guilt, staying “busy,” postponing, or documenting the process—is not equivalent to actually performing the task. It asserts that only tangible action, even if flawed, hesitant, partial, or resulting in failure, qualifies as “doing the thing.” The author concludes with a brief personal note indicating they are currently exploring full‑time and contract opportunities, and a self‑reminder to return to work.
Read full article →
Community Discussion
Comments largely endorse the idea that early, imperfect execution beats prolonged planning, emphasizing that shipping a flawed version provides concrete user feedback, uncovers real edge cases, and accelerates learning. Many contrast this with teams that waste time on extensive design docs, noting higher risk of pivots or failure. Several voices caution that minimal preparation remains essential for safety, quality, and coordination in larger projects, especially in regulated or complex domains. Overall, the consensus favors a bias toward action while recognizing a modest role for disciplined planning.
Try text scaling support in Chrome Canary
Community Discussion
The comment critiques a proposal to scale text independently of layout, arguing that text size is tightly coupled with page structure and that such scaling creates numerous testing permutations that are impractical for developers and can degrade usability. It favors responsive design and browser‑controlled zoom, noting existing solutions on desktop and Android’s non‑linear scaling, and expresses frustration that mobile browsers and sites often restrict user‑preferred scaling. Concerns about added fingerprinting bits, implementation via meta tags, and the broader impact on accessibility are also highlighted.
Xfwl4 – The Roadmap for a Xfce Wayland Compositor
Summary
The Xfce team announced a new Wayland compositor, xfwl4, funded by community donations and led by core developer Brian Tarrice. xfwl4 will be written from scratch in Rust using the smithay library, not a fork of the existing xfwm4 C code. The rewrite avoids X11‑specific architecture, eliminates risk to xfwm4, and leverages Rust’s safety to reduce memory bugs. smithay was chosen over wlroots because it implements most Wayland protocol extensions, offers low‑level customization, has strong documentation, and avoids the difficulty of creating Rust bindings for wlroots. Project goals include achieving feature parity with xfwm4, reusing xfwm4 configuration dialogs and xfconf settings, adding session‑startup changes (the compositor becomes the session root), supporting the xdg‑session‑management protocol, and providing XWayland compatibility. The Xfce CI container build system will be updated to compile Rust via Meson. Development has begun, with a first release expected around mid‑year; further details are available in the issue tracker and source repository, and discussions continue on the #xfce‑dev Matrix channel.
Read full article →
Community Discussion
Comments show mixed but generally hopeful sentiment toward an XFCE Wayland implementation in Rust. Many express optimism that modern libraries can deliver feature‑parity, good performance on low‑end hardware, and attract new contributors, while also emphasizing the need for speed, low memory use, customizability, theme support, and stability comparable to the X11 version. A notable portion remains skeptical, fearing loss of simplicity, potential bugs, and reduced X11 maintenance, and some request specific functionalities such as xscreensaver integration and reliable window shading. Overall, users desire a lightweight, reliable desktop that retains XFCE’s core qualities while embracing Wayland.