CD Projekt Red targets 2028 release for The Witcher 4

CD Projekt Red joint CEO Michał Nowakowski has revealed that The Witcher 4 is targeting a 2028 release window. The Polish studio has been developing the next mainline entry in the Witcher series for several years and has previously shared footage of the game. Before The Witcher 4 arrives, the studio is set to release Songs of the Past, an expansion to The Witcher 3, slated for 2027. Songs of the Past is being showcased at Gamescom this week, including an appearance at the Opening Night Live event hosted by Geoff Keighley.

IPFS Maintainers Winding Down

Article URL: https://ipshipyard.com/blog/2026-the-end-of-ipfs-at-shipyard/ Comments URL: https://news.ycombinator.com/item?id=49421489 Points: 179 # Comments: 75

Trump needs to risk pain to make Iran sanctions bite

How to Build a Modular C++ Static Library with Safe Input Handling

As C++ projects grow, cramming all logic into a single main.cpp file leads to technical debt, slower compilation, and difficult testing. A modular architecture addresses this by separating function declarations from their definitions and compiling utility code into reusable static libraries. The tutorial walks developers through structuring a C++ workspace with isolated headers and implementation files, organized under a CoreUtils namespace to avoid global pollution. Key safety measures include null pointer checks, 64-bit accumulators to prevent integer overflow, and stream-flush recovery to handle invalid user input. The project is compiled using a three-stage pipeline targeting C++17, producing a static archive file that can be linked across multiple projects.

100-Lens Framework Proposed to Evaluate Context-Aware AI Coding Agents

A new framework argues that passing tests is insufficient to determine whether an AI coding agent made the right engineering decision for a given software system. The proposal distinguishes between functional correctness, contextual correctness, and system-level correctness, noting an agent can produce valid, compiling code that still violates current architecture or project constraints. The argument draws on OpenAI's own guidance for its Codex agent, which recommends structured repository-level context files to help agents understand naming conventions, business logic, and dependencies. Existing benchmarks like SWE-bench Verified have faced growing reliability issues, with OpenAI flagging contamination and estimating roughly 30% of tasks in SWE-bench Pro as broken. The author concludes that because context materially affects agent performance, it must also become a formal part of how AI coding agents are evaluated.

$20 mln: New cost of shipping oil through Hormuz

ChatGPT Ad Library

Discussion | Link

Karnataka passes 8 Bills covering property registration, land use and transport

The Karnataka legislature has passed eight new Bills that will bring significant changes to property and land management across the state. The legislation introduces remote e-registration and strengthens digital infrastructure for property transactions. Unapproved rural properties will now be required to meet town planning standards. Parkland may be allocated for public utility purposes under regulated oversight, while additional measures aim to improve commercial rules and upgrade Bengaluru's transport network.

US student builds EV battery model with 2.36% error

Kangana Ranaut lashes out at Rakhi, Taapsee, Sonu Sood and Ramdev over criticism

BJP MP and actor Kangana Ranaut publicly hit back at several celebrities who had made remarks about her, including Rakhi Sawant, Taapsee Pannu, Ashutosh Rana, Sonu Sood, and Baba Ramdev. She described her critics as 'failed, stupid, and disgusting,' accusing them of using abuse to court media attention. Kangana declared 'enough is enough,' signalling she would no longer stay silent in the face of such comments. She also took particular aim at Baba Ramdev, alleging he made derogatory remarks about women while invoking 'Sanatan Dharm.'

Why Form Callbacks Break Portability and How Declarative Trees Fix It

Most form libraries manage state within a single application using callbacks for validation and conditional logic, but these cannot be serialized or shared across systems. When a form must be rendered by multiple clients, generated by a visual builder, or validated independently on a server, its logic must travel with its structure. Encoding behavior as plain strings or embedded expressions introduces parsing risks, versioning issues, and potential security concerns. A closed declarative expression tree — representing conditions and effects as pure JSON data — solves this by making form logic serializable, inspectable, and executable across different runtimes. This approach treats a form as a true cross-system contract rather than application-specific component state.

MS Paint and Photos Found to Embed Hidden GUIDs in Locally Saved Images

Microsoft's MS Paint and Photos apps have been found to invisibly embed unique identifiers (GUIDs) as watermarks in image files, even when those images are created or edited entirely offline. The discovery was made through reverse engineering and documented by a researcher at xusheng.dev. The hidden watermarks persist in locally generated output without any visible indication to the user. This raises potential privacy concerns, as the embedded identifiers could theoretically be used to trace images back to a specific installation or device. The finding drew significant attention on Hacker News, accumulating over 270 points and more than 120 comments.

How Astro Sites Can Handle Form Submissions Without a Backend Server

Astro, a framework popular for static, content-heavy websites, poses a challenge when developers need to handle form submissions without a server-side endpoint. A third-party service called onsubmit.dev allows Astro projects to send form data to an external endpoint, eliminating the need for a dedicated API route or server runtime. This approach keeps the site fully static and deployable to CDNs or platforms like GitHub Pages, while still accepting contact requests, feedback, or registrations. Developers can use plain HTML forms pointing to the external endpoint, meaning the form works even without JavaScript enabled. A key security reminder accompanies the approach: any credentials or API keys included in client-side code or generated HTML are publicly visible and must never be embedded there.

General Intuition in talks to raise funds at $6B valuation backed by Valor, Point72

General Intuition, an AI startup developing foundation models for generalized agents, is in talks to raise a new funding round at a $6 billion pre-money valuation. The round is set to include investments from Valor Ventures, Point72 Ventures, and Seven Seven Six. The company focuses on training AI agents to navigate and understand space and time. The fundraise signals growing investor confidence in foundational AI infrastructure as the startup eyes expansion into robotics.

Harper Beckham honors mother Victoria in heartfelt Italian holiday post

Harper Beckham shared photos from the family's Italian holiday, describing her mother Victoria as her 'whole world' in an emotional post. Victoria responded warmly by reposting the image along with a loving message for her daughter. David and Victoria Beckham continue to travel in Italy with Harper, while sons Romeo and Cruz have returned to the United Kingdom. Eldest son Brooklyn Beckham was notably absent from the holiday, amid ongoing reports of family tensions.

OpenAI: GPT 5.6 Sol price reduction (until at least Nov 21)

Article URL: https://developers.openai.com/api/docs/pricing Comments URL: https://news.ycombinator.com/item?id=49421074 Points: 174 # Comments: 168

'Hindu baiter': CM Himanta slams Rahul Gandhi over Manusmriti row

Silent Regex Bug Let Faulty Children's Alphabet Episodes Pass Automated Review

A developer running an automated pipeline to produce children's animated episodes discovered that a critical validation rule had silently stopped working due to a subtle Python string escape error. A regex meant to verify that narrator examples matched the correct letter — such as confirming 'P is for pillow' — never flagged any issues because \b was interpreted as a backspace character rather than a word-boundary anchor. Since the broken rule returned an empty result just like a passing rule would, the flaw was invisible during normal operation and would have allowed incorrect educational content to be published. The bug was only caught because the developer tested the gate against a set of real episodes that a human reviewer had already flagged with known defects. The incident highlighted how a validation check that silently fails is indistinguishable from one that finds no problems, making pre-deployment testing against known-bad inputs essential.

3D-printed sensor-embedded repairs help monitor and preserve ancient pottery

Researchers at Koc University have developed a new conservation method combining photogrammetry, 3D printing, and embedded sensors to restore damaged ancient pottery. The custom-printed replacement sections can fill in missing parts of artifacts while simultaneously tracking environmental factors such as temperature, humidity, pH levels, and physical stress. The technique was tested on maritime amphorae originating from Türkiye and Italy. The approach aims to give conservators real-time data on changes affecting artifacts, supporting more effective long-term preservation efforts.

How reCAPTCHA Evolved From Distorted Text to Silent AI Risk Scoring

CAPTCHA technology began as a simple text-distortion puzzle designed to distinguish humans from bots, relying on the fact that humans could read garbled characters while machines could not. As machine learning improved, the puzzles grew harder, creating a frustrating experience for legitimate users who were repeatedly asked to identify traffic lights or crosswalks. Google's reCAPTCHA v3 moved away from visible challenges entirely, instead running a background risk-analysis engine that silently observes behavioral signals such as mouse movement, click patterns, and browsing context. The system returns a score between 0.0 and 1.0, where lower values flag more suspicious activity, and website owners can set their own thresholds to decide how to respond. Rather than delivering a binary human-or-bot verdict, modern reCAPTCHA functions as a continuous risk assessment tool that balances security with user experience.