Hacker News
1Htmx 4.0
- htmx 4.0.0 has been released after 8 months of development, featuring a migration from XMLHttpRequest to fetch() API internally. The major changes include explicit attribute inheritance by default, standardized event names, and removal of localStorage-based history caching. New features include morph swaps and the hx-partial tag, plus numerous extensions for enhanced functionality.
Technology stack combinations
Developers enthusiastically share their preferred stacks combining HTMX with Go/SQLite (HUGS/GOSH), Rust/Axum, Django/Postgres, and other backends. Many praise the simplicity and developer velocity of these combinations, especially for rapid prototyping and AI-assisted development with tools like Claude.HTMX vs SPA frameworks debate
A significant discussion emerges around HTMX's suitability for different project types. Angular developers argue HTMX mixes presentation with business logic and lacks state management, while HTMX advocates counter that it's perfect for traditional websites but not meant for complex SPAs.CEO of HTMX meme
Multiple users jokingly claim to be "CEO of HTMX" throughout the thread, referencing an inside joke from the HTMX community. The actual HTMX creator participates in the meme, and users share links to htmx.ceo where anyone can become a "CEO."
2Our decision on Cursor following its acquisition by SpaceX
- OpenAI announced it will terminate its contract providing AI models to Cursor following SpaceX's acquisition of the developer tool, with service ending November 12, 2026. The decision stems from concerns about SpaceX's compliance with terms of service, citing past contract violations by Elon Musk's companies including Twitter and xAI. OpenAI expressed regret for the impact on developers who rely on their models through Cursor and committed to supporting affected users during the transition.
Model distillation and IP hypocrisy
Users highlight the irony of AI companies banning model distillation while using copyrighted training data themselves. Discussion covers whether distilling models should be fair use like training on books, and how companies enforce ToS when copyright law is unclear.Cursor alternatives and coding workflows
Developers discuss alternatives to Cursor's AI-integrated coding experience, comparing tools like Zed, VS Code with extensions, Claude Code, and Codex. Focus on editor-based review workflows vs terminal-based agentic coding approaches.Business model sustainability
Analysis of Cursor's token reselling model becoming unsustainable due to subsidized subscriptions and provider restrictions. Discussion covers pricing inefficiencies, enterprise costs, and the shift toward first-party models.
3U.S. sanctions against the A/I Collective
- The US Government has designated the host of noblogs.org as a "global terrorist" and imposed sanctions on Italian hosting provider Autistici Inventati. These actions have generated significant discussion with over 350 combined comments across related news posts.
Targeting infrastructure providers as terrorism
Commenters express concern about the unprecedented designation of digital infrastructure providers as terrorists, questioning whether users/developers of I2P, Monero, Signal, and other privacy tools could face similar targeting. Many see this as a dangerous precedent.What Autistici/Inventati actually does
Discussion centers on clarifying the organization's activities - they provide web hosting, email, blogs, and digital services to activists and grassroots movements, with some debate about their vetting process and political alignment.Broader implications of US sanctions power
Users debate the global reach of US financial sanctions through SWIFT banking, the political motivations behind these designations, and concerns about escalating government authoritarianism against privacy tools and anti-fascist movements.
4Inception-style curved map for turn-by-turn directions
- Orbify showcases a web demo featuring 3D rendering technology with WASD movement controls and drag-based camera panning and rotation. The company is seeking pilot projects, collaborations, and investment opportunities, with patent-pending warping technology powered by PlayCanvas Engine.
Prior art and patent concerns
Users identify multiple examples of similar curved map projections predating this concept, including Berg's 2009 "Here and There" poster, Disney's 1997 multiperspective panoramas paper, and existing patents from 2010. Many question the novelty and patentability of the approach.Motion sickness and usability issues
Multiple users report experiencing nausea and motion sickness from the curved projection effect, particularly during turns. The extreme spherical deformation is seen as too distracting for actual driving use, with suggestions to make it more subtle.Navigation functionality concerns
Discussion focuses on practical navigation issues, especially poor visibility around sharp turns where upcoming road information disappears off-screen. Users suggest improvements like predictive camera rotation and better handling of consecutive turns.
5Luanti removed from Google Play due to baseless AI copyright notice
- Luanti's Android app was removed from Google Play due to a baseless DMCA notice from Tracer.AI claiming it infringes Minecraft's copyright. Luanti is an open-source voxel game platform that doesn't contain any proprietary assets. This is the second false takedown from the same company.
DMCA system problems
Users criticize DMCA as fundamentally broken, allowing abuse by large corporations against small creators. Discussion covers Google's non-standard implementation, lack of proper counter-notice processes, jurisdictional issues forcing international creators into US courts, and calls for bonds or penalties to deter frivolous claims.Microsoft's repeated false claims
Microsoft (via Tracer.AI) has filed similar bogus DMCA notices against Luanti before in 2023 and other voxel games. Users see this as bad faith harassment and suggest firing responsible lawyers or pursuing countersuits, noting the AI tool creates false positives without proper human review.Automation without oversight
Discussion of how AI copyright detection tools like Tracer.AI automatically flag content based on visual similarity without understanding context. Users debate whether this represents intentional abuse or flawed automation, with concerns about corporations using cheap AI tools while externalizing costs of false positives.
6Debian votes to allow "responsible use of generative AI"
- The Debian project voted to allow responsible use of generative AI tools for development and documentation. While AI can improve productivity, all contributions must meet the same quality standards regardless of how they were created, and contributors remain fully responsible for reviewing and understanding AI-assisted work before submission.
Accountability and responsibility for AI-generated code
Discussion centers on the principle that developers remain fully responsible for code regardless of AI assistance. The policy requires contributors to understand, explain, and maintain their submissions, with enforcement through banning poor contributors.Impact on code review and open source collaboration
Contributors express concern about AI-generated patches overwhelming reviewers, degrading the mentorship aspect of code review, and forcing projects to close to external contributors due to volume of low-quality submissions.Practical implementation challenges
Developers discuss asymmetry between effort to generate vs. review AI code, management pressure to move faster, and real-world enforcement difficulties when responsible parties leave or can't fix broken production code.
7The internet is kind of a predatory cesspit now
- The internet has evolved from an authentic space built by passionate amateurs into a predatory machine designed to exploit human vulnerabilities for profit. Modern platforms industrialize the extraction of fear, greed, and loneliness, turning users into both consumers and unpaid distributors in endless grift cycles. This system creates addiction by design, where ordinary people promote products they don't understand while remaining trapped in the very conditions the grifts claim to solve.
Internet addiction and its business model
Users discuss how internet addiction has worsened over recent years, with platforms designed to maximize engagement through addictive mechanisms. Solutions include reframing addiction as choice, using delay techniques, and recognizing that knowledge consumption online often lacks real retention or practical value.Growing cynicism and societal decay
Discussion centers on how aging brings increased cynicism and misanthropy, with participants debating whether this is a healthy defense against hostile reality or something requiring active effort to counter through focusing on helpers and maintaining optimism.The old vs. new internet
Commenters contrast the early internet's focus on genuine human connection and community with today's engagement-maximizing algorithms. Many argue the old internet still exists but requires intentional effort to find and use non-corporate alternatives.
8Just the rumour of a bug is enough to find an exploit these days
- The author released a security fix for OCaml's cohttp but discovered automated attackers were already probing for the vulnerability within minutes of opening the public patch. Modern AI agents can now exploit vulnerabilities faster than developers can fix them, with exploitation preceding patches by an average of 7 days in 2024. Traditional security embargoes are becoming ineffective as AI only needs vague hints to independently discover and exploit bugs, forcing open source maintainers to rethink security processes toward continuous rapid fixes and proactive defenses.
AI-accelerated vulnerability discovery
Open source maintainers are overwhelmed by AI-driven security reports. The rclone project saw 40+ security disclosures in one month vs 20 in 10 years. While 75% have merit, the volume is exhausting maintainers who struggle with triage, fixes, and delayed CVE assignments.Management resistance to bug fixing
Despite AI making bug detection/fixing easier, managers prioritize speed over quality. Developers face resistance when trying to fix known issues, with bosses preferring rushed features over security patches. The problem is organizational will, not technical capability.Automated exploit generation from hints
AI can generate working exploits from mere rumors or vague hints about vulnerabilities. This democratizes exploitation beyond traditional skilled researchers, compressing timelines from weeks to minutes and enabling mass exploitation of low-value targets.
9Boot a Virtual iPhone via Apple's Virtualization.framework
- VPhone-CLI is a tool that allows users to boot virtual iPhones on Apple Silicon Macs running macOS 15+ using Apple's Virtualization framework. It supports creating, managing, and running iOS virtual machines with various security bypass levels from basic patches to full jailbreak configurations, requiring SIP/AMFI relaxation and specific dependencies to function properly.
Technical differences from iOS Simulator
Users discuss how this virtual iPhone differs from Xcode's iOS Simulator - this runs actual iOS via virtualization while the simulator compiles iOS frameworks to run natively on macOS, offering different testing capabilities for security research and app development.Regional restrictions and regulatory compliance
Discussion centers on why users must avoid selecting Japan or EU regions during setup due to regulatory checks the VM can't satisfy, including third-party app store access requirements and region-specific features like NFC payments.Security concerns and Apple's response
Users express concerns about running unknown binaries as root and debate whether Apple will eventually break this functionality, drawing parallels to their previous legal battles with Corellium over iOS virtualization.