Hacker News Daily — 2026-07-11 Trending
Scraped at 22:21, July 11, 2026 (PDT)
(1) What xAI's Grok Build CLI Actually Sends to xAI
The Grok Build CLI's data flows are exposed, revealing what information is sent to the xAI service. The post flags potential privacy and data governance implications and suggests how to limit or audit what’s transmitted.
揭示 Grok Build CLI 向 xAI 服务器发送的数据类型，引发隐私和数据治理方面的关注，并给出限制与审计的建议。
(2) Why are US consumers so angry? It's not just high prices
The piece argues that US consumer anger runs deeper than prices, citing wage stagnation, service frictions, and market concentration. It paints a broader picture of consumer sentiment in a changing economy.
分析美国消费者愤怒的根源，除了高价外还包括工资停滞、服务摩擦和市场集中等因素，呈现经济环境变化下的情绪图景。
(3) Mesh LLM: distributed AI computing on iroh
Mesh LLM demonstrates distributed AI computing across a mesh of nodes using iroh, reducing reliance on centralized hardware. The approach highlights how decentralized coordination can scale model inference and resilience in AI workloads.
Mesh LLM 展示了在 iroh 上通过网状节点进行分布式 AI 计算，降低对中心化硬件的依赖。去中心化协作在推理扩展和 AI 工作负载的弹性方面展现潜力。
(4) We Know Simple Fluids Can Flow. Turns Out, Some Can Fracture
New research shows that some simple fluids can fracture under extreme conditions, not just flow. This challenges conventional rheology and could affect modeling in soft matter, geophysics, and industrial processes.
研究表明，一些简单流体在极端条件下也会断裂，而不仅仅是流动。这挑战了传统的流变学，并可能影响软物质、地质与工业过程的建模。
(5) Under federal rule, colleges must leave grads better off or lose financial aid
A federal rule requires colleges to demonstrate that their graduates fare better financially or risk loss of federal financial aid eligibility. The shift signals tighter accountability for higher education outcomes.
联邦规定要求高校证明毕业生在经济上更有改善，否則将可能失去联邦资助资格。这一转变标志着对高校结果的更严格问责。
(6) Show HN: Ant – A JavaScript runtime and ecosystem
Ant introduces a new JavaScript runtime and its own ecosystem, offering a fresh option for running JS outside the browser. It emphasizes a streamlined, integrated toolchain that could influence server-side JavaScript workflows if it gains traction.
Ant 提供一个新的 JavaScript 运行时及其生态系统，为浏览器之外的 JS 提供新选项。若获得广泛采用，可能影响服务器端 JS 的工作流与工具链。
(7) Weightlifting beats running for blood sugar control, researchers find (2025)
A 2025 study finds weightlifting outperforms running for blood sugar control in certain populations, with implications for exercise prescriptions in diabetes management.
一项 2025 年的研究发现，在特定人群中举重对血糖控制的效果优于跑步，这对糖尿病管理的运动处方有影响。
(8) Long Covid May Physically Damage the Nerves That Control the Stomach
Long Covid may cause physical damage to nerves that regulate stomach function, offering a possible mechanism for persistent GI symptoms. The finding adds to the growing evidence of autonomic involvement in long Covid.
长新冠可能损伤调控胃功能的神经，为长期胃肠症状提供可能机制，进一步支持长新冠与自主神经受累相关的证据。
(9) Doctors die. It's not like the rest of us, but it should be (2016)
An essay about doctors' mortality, questioning how the medical profession contends with death and the cultural expectations around it.
本文探讨医生死亡的现实及其在医疗文化中的地位，提出应正视这一议题并思考相关的职业与情感影响。
(10) Apple sues OpenAI, accuses ex-employees of stealing trade secrets
A high-profile lawsuit pits Apple against OpenAI over alleged misappropriation of trade secrets by former OpenAI staff. The case centers on proprietary AI techniques and internal data siphoned during a transition, signaling growing tension as big tech fights over AI talent. The outcome could influence how firms protect confidential knowledge during acquisitions and departures.
一起备受关注的诉讼指控苹果指控OpenAI及其前员工窃取商业机密，聚焦于AI技术与内部数据的可能外泄。此案凸显大型科技公司在人才流动与知识保护方面的博弈，将影响企业在并购和员工离职时的机密信息防护策略。
(11) Female US rower completes historic solo journey from California to Hawaii
Profiles a female rower completing a solo cross-ocean voyage; highlights endurance, training, and the growing visibility of women in extreme exploration.
报道这位女性实现了单人横跨太平洋的纪录，展现极限耐力、训练与女性在极限探索领域的崭露头角。
(12) A pure scheme web programming tool
A minimal, pure Scheme toolchain for web development, avoiding other languages. It showcases how functional programming can handle frontend and backend tasks with a small, coherent toolset.
一个纯 Scheme 的网页编程工具，尝试用最小的工具链完成前后端开发。强调函数式编程在网页开发中的表达力和简洁性。
(13) Prefer strict tables in SQLite
Promotes using STRICT tables in SQLite to enforce data types and constraints, reducing silent data quality issues. Includes guidance on enabling strict tables and migrating schemas.
倡导在 SQLite 中使用 STRICT 表以强制数据类型和约束，降低隐性数据质量问题。提供启用严格表和迁移模式的实用建议。
(14) A dock that wakes up reliably
A USB-C/Thunderbolt dock that reliably wakes a host from sleep, addressing a common UX pain point. The write-up outlines hardware choices, wake signaling, and implications for laptop docking ecosystems.
一个扩展坞，能在休眠后可靠唤醒主机，解决常见的唤醒体验问题。文中讨论了硬件选择、唤醒信号及其对笔记本扩展生态的影响。
(15) Nvidia, CoreWeave, and Nebius: Inside the Circular Financing of the GPU Boom
Explains circular financing used by GPU-focused players to recycle capital across rounds, accelerating equipment deployment for AI workloads; discusses players' roles and financial tactics.
揭示围绕 GPU 行业的循环融资如何在多轮融资中回流资本，加速 AI 工作负载的显卡部署。分析参与方的角色和财务策略。
(16) Martha Lillard, last US polio patient using iron lung, dies at 78 in Oklahoma
Martha Lillard, the last U.S. polio patient using an iron lung, dies at 78 in Oklahoma, marking the end of an era in medical history. The obituary reflects on polio, medical technology, and societal changes.
美国最后一位使用铁肺的脊髓灰质炎患者玛莎·利拉德在俄克拉荷马州去世，享年78岁，标志着一个医疗历史时代的结束。对脊髓灰质炎、医疗技术及社会变迁的反思。
(17) RISCBoy is an open-source portable games console, designed from scratch
RISCBoy is an open-source, portable game console built from scratch, likely leveraging a RISC-V core. It showcases a DIY hardware ethos and offers a community-driven alternative to proprietary handhelds.
RISCBoy 是一个开源、便携的游戏机，完全从零设计，可能基于 RISC-V 架构。它体现了动手做硬件的精神，以及对专有掌上设备的社区驱动替代方案的探索。
(18) AI 2040 and the cult of intelligence
Examines how the AI 2040 horizon fuels cult-like hype around machine intelligence. It argues for a pragmatic, measurement-based approach to progress and broader governance that includes non-technical voices.
本文剖析关于 AI 2040 的热潮如何催生‘智能崇拜’，对以工程为中心的乐观叙事提出批判。呼吁以务实、可衡量的进展标准引导 AI 发展，并引入更广泛的治理与多元声音。
(19) A Erlang style pure Scheme Webserver and further
An Erlang-style pure Scheme web server demonstrates actor-like concurrency in a minimal Scheme environment, with follow-on improvements. It’s a playful take on merging concurrency models with a functional Lisp dialect.
介绍一种 Erlang 风格的纯 Scheme Web 服务器，展示在纯 Scheme 环境中实现类似 actor 的并发，以及后续改进。
(20) Leaded gas was a known poison the day it was invented (2016)
Reviews historical evidence that leaded gasoline was recognized as poisonous from the outset, challenging rosy narratives about the technology. Discusses policy implications for tech risk governance.
回顾证据，表明铅化汽油在发明之初就被视为有毒，这与乐观叙事相悖。探讨技术风险治理与政策制定的启示。
(21) We scaled PgBouncer to 4x throughput
Shows how to push PgBouncer-based connection pooling to four times the throughput for PostgreSQL workloads, with practical tuning tips and pooling-mode considerations. Demonstrates integration with a managed Postgres stack to reduce connection overhead and latency.
展示如何将 PgBouncer 基于连接池的吞吐量提升至四倍，给出实际的参数调整和池模式选择要点。展示在 ClickHouse 管理的 Postgres 场景中降低连接开销、提升延迟表现的做法。
(22) Modern decor may be straining people's brains
Cites studies suggesting modern decor increases cognitive load and reduces focus, with implications for workplace design and user interfaces.
研究显示某些现代装饰风格会增加认知负荷、降低专注度，对工作场所设计和用户界面有启示。
(23) How to hide from killer drones
Raises ethical and policy questions about civilian responses to drone threats, noting the risks and legal constraints. Highlights the need for robust drone governance and defense strategies.
提出关于民用对抗武装无人机的伦理、法律与可行性问题，强调相关风险与监管难点。强调需要加强无人机治理与防御策略。
(24) UPI: Anatomy of a Payment Transaction
Outlines how UPI payments flow from payer to merchant, detailing routing, settlement, and security checks; implications for throughput, fee structures, and interoperability.
解析 UPI 支付的完整交易流程，包括路由、结算与安全校验，揭示其对吞吐量、费用和互操作性的影响。
(25) QuadRF can spot drones and see WiFi through my wall
QuadRF demonstrates that RF sensing can reveal both drone activity and WiFi signals through walls. This kind of passive observation highlights security and privacy implications for home networks and aerial devices, even when devices aren't in visible sight. It also underscores practical limits like range, clutter, and legal boundaries.
QuadRF 展示了射频探测可以在墙后揭示无人机活动和 WiFi 信号。这类被动观测引发对家庭网络和无人机安全与隐私的担忧，同时也暴露出覆盖范围、信号干扰和法律边界等局限。
(26) GPT-5.6
A new GPT-5.6 model appears, signaling continued iteration on the GPT-5 family. Expect refinements in reasoning, safety, and efficiency, reflecting the ongoing cadence of large language model updates.
GPT-5.6 版正逐步进入，表示 GPT-5 系列的持续迭代。预计在推理、安全与效率方面有改进，体现大语言模型更新的持续节奏。
(27) New York City to ban deceptive subscription practices
New York City plans to ban deceptive subscription practices, including auto-renewal traps and hidden charges. The move aims to improve consumer protection and force clearer disclosures from subscription-based services, potentially reshaping how onboarding and renewal terms are presented online. This could raise the bar for startups and platforms relying on subscription models.
纽约市拟禁止欺骗性订阅行为，包括自动续订陷阱与隐藏费用，提升消费保护。此举将促使订阅型服务在用户注册与续订条款上提供更清晰的披露，或对采用订阅模式的企业带来影响。
(28) EU Parliament greenlights Chat Control 1.0
This marks a major push to enable cross-border scanning of encrypted messages in pursuit of child safety and crime prevention. Critics warn that the measure expands surveillance at the expense of privacy and free expression, while Breyer argues that children will lose out.
这标志着推动跨境扫描加密通讯以提升儿童安全和打击犯罪的重大举措。批评者担心此举侵蚀隐私与言论自由，而Breyer则认为这会让儿童吃亏。
(29) Show HN: Learn by rebuilding Redis, Git, a database from scratch
A learning-by-rebuilding project; shares experiments and code for reconstructing core systems, offering practical insights into data structures and OS concepts.
一个以重建核心组件来学习的练习项目，包含 Redis、Git、数据库等实现的代码与笔记，帮助理解底层数据结构与系统设计。
(30) GPT-5.6 Sol Ultra produces proof of the Cycle Double Cover Conjecture [pdf]
GPT-5.6 Sol Ultra has produced a proof of the Cycle Double Cover Conjecture, with a PDF provided. If verified, this would mark a notable milestone in AI-assisted formal proofs and could accelerate future mathematical work, but the proof will be rigorously scrutinized by the community. The episode highlights AI's potential to aid deep mathematics while underscoring the need for careful validation.
GPT-5.6 Sol Ultra 似乎给出了循环双覆盖猜想的证明，并附有 PDF。若经同行验证，这一成果可能标志着 AI 辅助形式化证明迈出重要一步，或加速后续的数学研究，但证明将接受严格审查。此事凸显 AI 在高深数学领域的潜力，同时也强调需要谨慎验证。
(31) Ghost Font: A font that humans can read but AI cannot
A font designed to be legible to humans but difficult for AI systems to read, offering a tactic to protect privacy and combat automated data capture. It raises questions about practicality, accessibility, and the broader arms race between human readability and machine recognition.
这款字体让人类易读但对 AI 不易识别，旨在提升隐私保护并抵抗自动化数据采集。它也提出了人类可读性与机器识别之间的权衡，以及相关的现实可用性与伦理问题。
(32) Einstein's relativity rules chemical bonds in heavy elements, new research shows
Relativistic effects become significant in very heavy elements, reshaping electron orbitals and bond strengths. The new work shows that Einstein's relativity plays a decisive role in how chemical bonds form in superheavy elements, with implications for modeling their chemistry and discovering new materials. This challenges non-relativistic intuition that underpins much of chemistry for lighter elements.
在极重元素中，相对论效应显著改变电子轨道、键能与分子几何。最新研究表明，爱因斯坦的相对论在超重元素的化学键形成中发挥关键作用，这可能改变对其化学性质的建模与新材料的设计。
(33) The early History of the Singular Value Decomposition (1993) [pdf]
Tracing the origins of the singular value decomposition and its early formulations, showing how the idea underpins modern data compression, PCA, and machine learning. The PDF offers historical context for a fundamental tool in linear algebra.
追溯奇异值分解的早期发展及其在现代数据分析中的影响，展示该工具在降维、数据压缩和机器学习中的核心作用的历史背景。
(34) ZeroFS vs. Amazon S3 Files
ZeroFS contrasts its filesystem approach with Amazon S3's object storage, discussing trade-offs in semantics, performance, and usability.
ZeroFS 与 Amazon S3 的文件模型对比，讨论语义、性能与易用性等权衡。文章强调在某些用例中，ZeroFS 提供更接近本地文件系统的语义和更顺畅的开发体验。
(35) Reverse centaurs are the answer to the AI paradox (2025)
Advocates hybrid human-AI collaboration where humans and AI work as 'reverse centaurs', arguing this reduces overreliance on AI and leverages human intuition.
主张以“反向半人马”的方式把人和 AI 结合起来协作，认为这能降低对 AI 盲目信赖、更好地发挥人类直觉与技能。
(36) Networking and the Internet, from First Principles
A primer that distills networking down to its core ideas, from physical links to application protocols. It highlights key abstractions and design choices that enable scalable, interoperable communications—useful mental models for developers.
这是一份以第一性原理解析网络与互联网的入门阅读，从物理链路到应用层协议，强调核心抽象和可扩展的通信设计。为开发者提供有用的思维模型，帮助理解网络的可用性与互操作性。
Discusses governance and control of autonomous AI agents; who bears responsibility for their decisions, what safety rails exist, and how to audit their actions. Puts focus on accountability, safety, and system-wide risk management.
探讨自治 AI 代理的治理与控制：谁应对代理的决策负责、有哪些安全机制、以及如何对其行为进行审计。强调问责制、系统性风险管理与透明度的重要性。
(38) Show HN: 18 Words
Show HN: 18 Words introduces a minimal writing tool that limits posts to 18 words, prioritizing brevity and speed. It exemplifies a design philosophy where constraints drive clarity and reduce noise.
Show HN: 18 Words 展示了一个极简写作工具，帖子被限制为18个单词，强调简洁与快速分享。此举体现了通过约束提升表达清晰度、降低噪声的设计理念。
Great tools fade into the background by design, reducing cognitive load and letting people focus on what matters. The piece cautions that invisibility should not hide value from teams and argues for clear outcomes and good integration.
优秀的工具被设计为在工作流中隐形，降低认知负担，让人专注于核心任务。文章提醒不要让它们的价值被忽视，并强调要有清晰的成效与良好的集成。
(40) Show HN: Orbit – AR satellite tracker, watch 15k+ objects
Presents Orbit, an AR satellite tracker that visualizes 15k+ objects in real time; demonstrates how AR can augment space awareness and educate users.
介绍 Orbit，这款 AR 卫星追踪器可在现实世界中实时可视 15k+ 颗对象，展示了 AR 在提升空间认知方面的潜力。
(41) An update on residential proxies and the scraper situation
Residential proxies remain a core tool for large-scale scraping, but policy moves and anti-abuse measures are tightening the space. The update points to changes in pricing, reliability, and access that could disrupt data-driven workflows. For developers and researchers, it signals the need for compliant scraping strategies and robust data sourcing.
住宅代理仍是大规模抓取的核心工具，但相关政策和反滥用措施正在收紧。更新指出定价、可靠性与获取途径的变化，可能影响需要大量网页数据的工作流。对开发者和研究者而言，意味着需要采取合规的抓取策略与更稳健的数据获取方案。
(42) The vintage beauty of Soviet control rooms (2018)
These rooms showcase vast, modular consoles, glowing CRT displays, and tactile controls that reflect Soviet-era computing and operational culture. The imagery highlights a design language where humans and machines coexisted in the control room, a contrast to today’s lean dashboards. The piece is a reminder that industrial aesthetics can influence contemporary UI/UX design.
这些照片展示了广阔的模组化控制台、发光的 CRT 显示和大量触控旋钮，映射出苏联时代的计算与运维文化。其设计语言强调人机共处的工作场景，与今日的简洁仪表板形成对照。该组影像提醒人们，工业美学仍能影响现代 UI/UX 设计。
(43) The Chinese Voice Actor Forced to Prove He's Human
The Chinese voice actor finds himself compelled to prove he is human amid rising AI-generated voices, highlighting issues around worker rights and authenticity in media.
在人工智能合成声音日益普及之际，一名中国配音演员被迫证明自己是人类，凸显了对工人权益与媒体真实性的关注。
(44) Your code is fast – if you're lucky
Performance is often brittle and depends on more than the code itself—compilers, runtimes, and hardware can shift timings. Profiling hot paths, avoiding micro-optimizations, and testing under realistic workloads reveal true speed.
代码快不快往往不仅取决于实现，还取决于编译器、运行时和硬件环境。文章建议以真实工作负载进行基准测试，关注热路径并避免过度追求微观优化。
(45) FCC approves test of space mirror to light night sky
The approval spotlights a clash between ambitious space-illumination experiments and regulatory concerns. A space mirror test could illuminate the night sky but risks light pollution and interferes with astronomy, illustrating the trade-offs of large-scale optics projects.
FCC 批准对太空镜进行夜空测试，计划在天空中投射光线，但引发天文学家与环保组织的抗议。此事凸显宏大太空视觉项目与对天空观测影响之间的矛盾。
(46) Speculations Concerning the First Ultraintelligent Machine (1965) [pdf]
I. J. Good's 1965 speculation on an ultraintelligent machine anticipates rapid self-improvement and the intelligence explosion that dominates AI safety discussions today. The piece underscores early concerns about control and the sweep of societal impact that such systems could unleash.
1965 年，I. J. Good 提出超智能机器可能通过自我改进迅速提升智能，预示了后来 AI 安全领域的“智力爆炸”议题。文本也反映了对控制与社会影响的早期担忧。
The Late Bronze Age collapse refers to a complex, multi-state disruption around 1200 BCE. The post offers a concise introduction to prevailing theories—from drought and famine to invasions and trade breakdown. It matters because it informs how interconnected civilizations can fail, a topic with resonance for modern supply chains and risk management.
晚青铜时代崩溃指的是公元前约1200年的多国共同性崩溃，理论包括干旱、饥荒、入侵与贸易网络破裂等。该简介帮助理解为何高度互联的文明体系会崩溃，这对当今的供应链与风险管理仍具启示意义。
(48) SpaceX wants to launch 100k more Starlink satellites for 100x the bandwidth
SpaceX aims to deploy up to 100,000 new Starlink satellites to boost total bandwidth by around 100x. The plan would dramatically reshape global connectivity, but faces colossal challenges in propulsion, spectrum, orbital debris, and regulatory approvals. If realized, it could push internet access to remote regions while stressing space logistics and climate concerns.
SpaceX 计划再发射十万颗 Starlink 卫星，带宽有望实现约100倍的提升。此举可能彻底改变全球连接格局，但在推进、频谱、轨道碎片与监管方面仍面临巨大挑战。若实现，将使偏远地区上网更易，但也对太空运营与环境提出新压力。
(49) Show HN: Getting GLM 5.2 running on my slow computer
A developer documents getting GLM 5.2 to run on a slow machine, sharing practical steps, constraints, and tweaks. It shows how lightweight inference paths and community tooling can unlock usable performance on modest hardware.
开发者记录在慢速计算机上运行 GLM 5.2 的过程，提供具体步骤、限制与调整方法。展示了通过轻量化推理路径和社区工具在中等硬件上也能获得可用性能的可能。
(50) Why it's so difficult to produce American-made medical gloves
Explains manufacturing, labor, and policy constraints that complicate domestic production of American-made medical gloves; discusses supply chain resilience.
分析妨碍国内生产医用手套的制造、劳工与政策难点，探讨供应链韧性与国家自主性。
(51) The tech of 'Terminator 2' – an oral history (2017)
An oral history revisits the technical feats behind Terminator 2, highlighting practical effects, model work, and the early balance between analog and digital tricks. It shows how teams solved filmic challenges with ingenuity under tight budgets and time pressures. The piece illuminates a pivotal moment in visual effects history.
这部《终结者2》技术口述史回顾了特效背后的硬件与软件成就，强调实拍特效、模型工作以及早期的数字与模拟技术的结合。作者揭示了团队在预算和时间压力下如何凭借创造力解决难题，塑造了影像特效的历史转折点。
(52) Show HN: Reame – a CPU inference server that gets faster as it runs
Showcases Reame, a CPU inference server whose performance improves as it runs, likely through caching or adaptive optimizations; discusses practical ML inference on CPU.
展示 Reame，这个 CPU 推理服务器随着运行时间提升性能，可能通过缓存与自适应优化实现；讨论在 CPU 上进行高效推理的实用性。
(53) BLISS
Historical overview of BLISS, a pioneering language used in early Unix-like systems; notes on syntax, capabilities, and influence on later languages and OS research.
回顾 BLISS 这门开创性的系统编程语言在早期 Unix-/OS 研究中的应用，介绍其语法特性与对后续语言与系统设计的影响。
(54) Tropical forests facing increasing risks of exposure to critical temp thresholds
Reports growing risk that tropical forests cross critical temperature thresholds, threatening carbon storage and biodiversity; underscores need for climate adaptation.
指出热带森林越来越容易暴露在关键温度阈值之下，威胁碳储存和生物多样性，强调应对气候变化的紧迫性。
(55) How the terrorist group Boko Haram uses frontier AI
A CASP report documents how Boko Haram leverages frontier AI to plan and execute operations, illustrating the evolving threat of AI-enabled extremism. The findings underscore the need for monitoring, safeguards, and countermeasures to prevent misuse. It also highlights the broader risk landscape as AI technologies spread.
一份 CASP 报告揭示博科圣地利用前沿 AI 进行作战计划与行动，体现了 AI 助力极端主义的新演变。研究强调需要加强监控、建立防护措施并制定对策以防止滥用，同时也反映出随着 AI 技术普及，风险格局正在扩大。
(56) Ask HN: Why isn't Google indexing information about the AT Protocol?
Questions indexing of AT Protocol; touches on documentation and search discoverability for new protocols; implications for developer onboarding.
讨论 AT Protocol 的信息为何在搜索中缺乏可发现性，涉及文档分发、索引策略与新协议的开发者上手难题。
(57) Snails' teeth beats spider silk as nature's strongest material (2015)
A 2015 study found that snail teeth are among the strongest natural materials, outmatching spider silk due to their mineralized structure. This reshapes assumptions about how biology builds tough materials and inspires bio-inspired engineering approaches. The finding continues to influence materials science discussions today.
2015年的研究发现，蜗牛牙齿因矿化结构而具备极高强度，甚至超过蜘蛛丝。这一发现改变了对自然材料强度的直觉，并为生物启发式材料设计提供新思路，成为材料科学讨论的经典案例。
(58) Show HN: Earth Game – An offline CLI for turning life goals into quests
Introduces Earth Game, an offline CLI that translates life goals into quests, enabling lightweight gamification without online dependencies.
展示 Earth Game，这款离线命令行工具把生活目标转化成任务/挑战，适合在没有网络的情况下进行自我驱动的游戏化管理。
(59) Successful companies go blind
As companies scale, they can lose sight of fundamentals like customer needs and product quality. The article argues for deliberate checks—maintaining a 'healthy paranoia' about risk, investing in reliability, and keeping feedback loops intact.
企业在快速成长后容易忽视基本面，如客户需求和产品质量。文章提倡保持对风险的健康警觉、加强可靠性建设，并维护有效的反馈循环。
(60) After 7 years in production, Scarf has reluctantly moved away from Haskell
After seven years in production, the Scarf project has reluctantly moved away from Haskell, citing reasons like ecosystem maturity, performance, or maintenance constraints. The decision reflects the real-world tradeoffs teams face when choosing a language for long-running systems software.
经过七年的生产实践，Scarf 项目被迫放弃在生产环境中使用 Haskell，原因包括生态系统成熟度、性能与维护成本等权衡。这反映出长期运行的系统在语言选择上的现实考量。
(61) War Atlas: An interactive cartography of every named war in human history
War Atlas offers an interactive map of named wars across history, letting users explore when and where conflicts happened and how large they were. The visualization helps researchers, educators, and curious readers grasp the scale and geography of conflict over time. It’s a compelling data-driven lens on world history.
War Atlas 提供跨历史的命名战争互动地图，便于用户查看战争发生的时间、地点及规模。该可视化工具对研究者、教育工作者与好奇读者理解冲突的时空分布具有帮助，提供了以数据驱动的历史视角。
(62) In Emacs, everything looks like a service
An essay arguing for modular, service-like abstractions in Emacs customization to improve composability and testability. It discusses trade-offs between simplicity and modularity and offers practical guidance.
作者主张在 Emacs 的扩展中采用类似服务的模块化抽象，以提升组合性和测试性。讨论了简化与模块化之间的取舍，并给出实际建议。
(63) A love letter to flashcards
A Love Letter to Flashcards praises flashcards as a learning tool, emphasizing spaced repetition and concise prompts. It discusses design choices that make flashcards effective and enduring in modern study workflows.
这篇文章以情感笔触赞扬抽认卡在学习中的作用，强调分散性重复和简短提示的重要性。还讨论了让抽认卡在现代学习流程中高效持久的设计要点。
(64) AI 2040: Plan A
AI 2040 presents Plan A—a long-range vision for AI governance, safety, and societal impact through 2040, contrasting optimistic and cautionary scenarios.
AI 2040 的 Plan A 提出面向2040年的长期愿景，涵盖治理、安全和社会影响等方面，展现乐观与审慎情景的对比。
(65) Punk, or why I don't stream anymore
A personal reflection on streaming culture, burnout, or policy concerns that led the author to stop streaming. It touches on privacy, community dynamics, and sustainability of creator work.
作者就流媒体文化、职业疲劳或平台政策等问题进行反思，解释为何不再进行直播。文中涉及隐私、社区关系以及创作者工作的可持续性等议题。
(66) Computation as a universal and fundamental concept
A discussion framing computation as a universal, fundamental idea that underpins sciences, math, and engineering. It promotes computational thinking as a cross-disciplinary literacy.
讨论将计算视为跨科普与工程等领域的普遍、基础的概念，倡导计算思维作为跨学科的基础素养。
(67) GhostLock, a stack-UAF that has existed in ALL Linux distributions for 15 years
A long-lived stack-use-after-free vulnerability dubbed GhostLock affects many Linux distributions, enabling potential exploitation across multiple kernels and users. The writeup details the vulnerability's root cause, its persistence, and suggested mitigations.
被称为 GhostLock 的栈上使用后释放漏洞在多数 Linux 发行版中长期存在，可能跨内核与用户造成利用。文中分析了漏洞的根本原因、持续存在的原因以及缓解方案。
(68) What's the best way to do authentication in modern applications
Modern apps should rely on standards-based authentication flows like OAuth 2.0 / OIDC with PKCE for public clients. Use short-lived access tokens paired with refresh tokens, and consider passwordless options such as WebAuthn to reduce phishing. Store tokens securely (HttpOnly cookies) and enforce MFA and CSRF protections to harden security.
现代应用应依赖基于标准的认证流程，如针对公客户端的 OAuth 2.0 / OIDC + PKCE，并使用短期访问令牌搭配刷新令牌。可以考虑无密码选项如 WebAuthn 以降低钓鱼风险，并将令牌安全存储（HttpOnly cookies），配合 MFA 与 CSRF 防护加强安全性。
(69) AI Can't Recreate the Thrust Game (But It Can Help You Understand It)
Illustrates why a faithful AI recreation of Thrust is tricky due to physics and timing, while showing how AI-powered tooling can illuminate legacy software architecture and decisions.
说明因物理引擎与时序等原因，AI 很难完全还原 Thrust；但 AI 辅助的工具能揭示早期软件架构和设计取舍。
(70) The Annotated JEPA
Annotated JEPA provides commentary to help readers understand the original text, with clarifications and context for modern audiences.
JEPA 的注释版为读者提供额外的注释与背景，帮助现代读者理解原文内容。
(71) Cpp2Rust: Translates C++ to safe Rust automatically
Cpp2Rust automatically translates C++ code to Rust with safety guarantees, easing migration of legacy codebases. While it promises safer memory management, developers should scrutinize correctness and performance post-migration and plan for integration challenges. It opens a path to safer reimplementations of old code.
Cpp2Rust 能自动将 C++ 代码转换为具有安全性保证的 Rust，帮助遗留代码的迁移。虽然承诺更安全的内存管理，但迁移后仍需对正确性和性能进行审查，并考虑集成中的挑战。为旧代码提供了更安全的重实现路径。
(72) Book: RISC-V System-on-Chip Design
This book surveys how to design Systems-on-Chip around the RISC-V ISA, covering architecture, verification, toolchains, and ecosystem considerations. It helps engineers reason about open ISA advantages and practical design trade-offs for modern SoCs.
本书系统介绍在 RISC-V 指令集上的系统芯片设计，包括架构、验证、工具链与生态等方面，帮助工程师理解开源 ISA 在现代硬件设计中的应用与取舍。
(73) Show HN: Reverse-engineering web apps into agent tools
Show HN describes turning existing web apps into agent tools, enabling automation and composition, while weighing security and reliability implications.
Show HN 展示了将现有网页应用转化为代理工具的思路，便于自动化与组合使用，但也需权衡安全性与可靠性问题。
(74) Show HN: Wyrm – Solve algebra by touch, built on an open-source soundness engine
Wyrm lets users solve algebraic problems through touch-based interaction, built on an open-source soundness engine. The project demonstrates novel human-computer interaction for math education and accessibility, potentially enabling new ways to teach and explore algebraic concepts. It also provides a glimpse into how formal-method tooling can empower creative coding.
Wyrm 让用户通过触觉交互来求解代数题，底层依赖开源的 soundness 引擎。该项目展示了用于数学教育和无障碍的新型人机交互方式，或为教学和探索代数概念带来新路径，同时体现了形式化工具在创意编码中的潜在应用。
(75) Billions of Sketches Reveal Hidden Cultural Variation in Human Concepts
Researchers analyze billions of sketches to show that concept representations vary across cultures, challenging universal assumptions in AI and cognitive science. The findings imply that AI systems trained on a single cultural dataset may misinterpret concepts in diverse user groups.
研究者对数十亿幅素描进行分析，揭示人类概念在不同文化中存在显著差异，质疑AI与认知科学中的普遍性假设。结果意味着以单一文化数据集训练的AI系统在不同群体中的概念理解可能存在偏差。
(76) Apple Silicon Exec Explains Mac Mini AI Demand and On-Device Future
An Apple Silicon executive explains that AI workloads are driving demand for the Mac mini and that Apple is prioritizing on-device AI to improve privacy and latency. The note suggests the Mac mini could serve as a low-cost testbed for developers and as a stepping stone to a broader on-device AI strategy across Apple devices.
苹果芯片高管表示，AI 工作负载推动 Mac mini 的需求增长，苹果将优先发展设备端推理以提升隐私与响应速度。Mac mini 可能成为开发者的低成本测试平台，并作为苹果在全系设备上推进设备端 AI 策略的一个起点。
(77) Lost city discovered beneath Egypt's desert with ancient church
Archaeologists have uncovered a previously unknown city buried beneath the Egyptian desert, along with an ancient church. The find sheds light on urban life and early Christian presence in the region, illustrating how desert geology can preserve antiquities for study. It also hints at trade routes and cultural exchange in ancient Egypt.
考古学家在埃及沙漠下发现一座未知的失落城市，以及一座古老教堂。此发现揭示了该地区早期城市生活与基督教早期足迹，说明沙漠地质对文物保存的重要性，同时也指向古代埃及的贸易网络与文化交流。
(78) Alternate clock designs and time systems
The piece surveys unconventional clock designs and timekeeping ideas, weighing pros and cons of decimal time or alternative calendars. It shows how time design shapes software, culture, and human routines.
文章概览了非传统的时钟设计与时间系统，评估十进制时间等替代日历的利弊，展示时间设计如何影响软件、文化和日常生活。
(79) An iroh powered smart fan
An iroh-powered smart fan demonstrates how compact compute modules can manage sensor input, motor control, and network connectivity in a small form factor. The post showcases use cases for iroh-powered devices and discusses performance, power efficiency, and integration with smart home ecosystems. This highlights how modular hardware platforms enable smarter, affordable devices.
一个由 iroh 驱动的智能风扇展示了小型计算模块如何在紧凑机身内处理传感数据、执行风扇控制并连接网络。文章探讨了 iroh 驱动设备的应用场景、性能、能耗，以及与智能家居生态的集成前景，揭示了模块化硬件带来的机遇。
(80) Otary – Image and Geometry Python Library Now Has Tutorials
Otary's image and geometry library just released tutorials that walk users through installation, basic workflows, and typical use cases. The materials lower the barrier to experimenting with advanced image-processing and geometric computations in Python.
Otary 的图像与几何处理库推出教程，涵盖安装、入门和常见工作流，降低学习门槛。有助于科研和教学中使用该库进行图像与几何计算。
(81) Show HN: Reviving my 2001 college band with AI
A developer uses AI to revive a college band from 2001, sharing the creative process and newly generated tracks. The project demonstrates AI's ability to reinterpret past sounds and produce fresh material, inviting discussion on authorship and tooling. It’s a personal, tech-forward take on musical resurrection.
作者利用 AI 复活了自己2001年的大学乐队，分享了创作过程和新生成的音乐作品。这一尝试展示了 AI 重塑往日声音并创作新素材的能力，同时引发关于作者权与工具使用的讨论。
(82) Combustion engine web-based simulator
A web-based simulator lets learners explore combustion engine behavior directly in the browser, enabling hands-on thermodynamics experiments without hardware. It’s a useful educational tool for students and hobbyists to test concepts and ideas quickly. The project lowers barriers to understanding engine dynamics.
一个基于网页的模拟器允许学习者直接在浏览器中研究燃烧发动机的行为，免去硬件需求，便于做热力学的动手实验。对学生和爱好者来说，这是一个快速测试概念与想法的有用教育工具。
(83) Biff.graph: structure your Clojure codebase as a queryable graph
Biff.graph lets you model a Clojure codebase as a graph that can be queried, enabling structural analysis and refactoring insights. It helps developers inspect dependencies, call graphs, and relationships across namespaces.
Biff.graph 允许把 Clojure 代码库建模为一个可查询的图谱，便于进行结构分析和重构洞察。它帮助开发者查看依赖关系、调用图以及各命名空间之间的关系。
(84) Inference Optimization for MiMo v2.5: Pushing Hybrid SWA Efficiency to the Limit
MiMo v2.5 showcases inference optimization by combining SWA (Stochastic Weight Averaging) with hybrid strategies to squeeze efficiency for on-device AI. The update highlights gains in speed and energy efficiency for mobile models, enabling stronger performance within constrained hardware. This matters for mobile developers pushing AI features in edge devices and cameras.
MiMo v2.5 展示了将 SWA（随机权重平均）与混合策略结合的推理优化，在受限硬件上提升了速度与能耗效率。更新强调了移动模型在边缘设备上的性能提升，使在手机、相机等场景落地的 AI 功能更具竞争力。
(85) Silent speech with ultrasound
Researchers use ultrasound to capture articulatory movements or to stimulate vocal tract, enabling silent speech communication. The approach could empower people with speech impairments and enable quieter communication in noisy environments, but faces challenges in accuracy, privacy, and real-time practicality.
研究团队利用超声波捕捉舌、口腔等发音器官运动，或用于刺激声道，从而实现无声语音通信。这项技术对语言障碍者和嘈杂环境中的安静交流具有吸引力，但在准确性、隐私与实时性方面仍面临挑战。
(86) Google Search lets creators know more about their reach
Google Search Console's reach insights give creators visibility into where their content appears and how audiences find it across search. The data helps optimize content strategy and understand the real impact of search visibility.
Google 搜索控制台的新数据让创作者看到内容在搜索中的覆盖情况与受众来源。通过这些数据，创作者可以更好地优化内容策略并评估搜索可见性的实际影响。
(87) An agent in 100 lines of Lisp
A compact Lisp-based agent implemented in about 100 lines, illustrating how a small interpreter and simple loops can drive goal-oriented behavior. It highlights Lisp’s expressiveness for quick AI experiments.
用约 100 行 Lisp 实现的智能体，展示了一个小型解释器和简单循环如何驱动目标导向的行为，凸显 Lisp 在快速 AI 实验中的表达力。
(88) Digital Deli, 1984 book by early PC hackers and enthusiasts
Represents a snapshot of 1980s PC culture, revealing DIY hacking ethos, toolsets, and community norms that shaped personal computing.
记录 1984 年代早期 PC 黑客与爱好者的文化风貌、工具与社区规范，揭示个人计算的发展脉络。
(89) I Did Not Kill Stanley Lieber: How to Draw (With 9front)
A quirky exploration of drawing with 9front, a Plan 9-inspired OS, through a Stanley Lieber reference. The piece uses retro tooling to show how constraints can inspire creative programming experiments.
以 9front 为背景的绘画实验，借用斯坦利·李的名字制造幽默参照，展示用 Plan 9 式系统进行图形创作的乐趣。通过受限工具链，体现了约束也能激发创作。
(90) The mathematical secrets of Barcelona's Sagrada Familia
The Sagrada Família showcases sophisticated geometry and proportional systems that guided Gaudí’s design. From curved surfaces to structural strategies, the article highlights how math underpins iconic architecture.
圣家堂的设计大量使用复杂几何与比例体系，指引了高难度曲面与结构策略。文章揭示了数学在标志性建筑中的核心作用。
(91) Ditching Vagrant: VMs with KVM and Virsh on Debian
A practical guide to ditching Vagrant in favor of KVM with virsh on Debian, covering setup, networking, and lifecycle management. Highlights performance gains and closer control over virtual environments.
提供在 Debian 上用 KVM 与 virsh 取代 Vagrant 的实用指南，涵盖安装、网络与生命周期管理。强调性能提升与对虚拟环境的更直接掌控。
(92) Ask HN: Are systems ready for the first negative leap second?
The question asks whether existing systems can handle a negative leap second, a hypothetical situation that would subtract time from a day. It highlights concerns about timekeeping, clock synchronization, and time APIs, inviting engineers to share preparedness, mitigations, and past experiences. A reminder of the fragility of digital timekeeping.
这个提问探讨若出现首个负闰秒，现有系统是否能够承受在一天中“减去时间”的情景，涉及时钟同步、时间 API 与时间管理的鲁棒性。邀请工程师分享应对策略、缓解措施和相关经验，提醒数字时间管理的脆弱性。
(93) The mask that compiles to nothing: how HotSpots JIT learned to reason about bits
HotSpot JIT can use bit-level reasoning to optimize away masking work, turning what looks like meaningful code into lean machine instructions. This reveals how compiler optimizations quietly boost performance for heavy JVM workloads.
HotSpot JIT 如何通过位级推理把看似有意义的掩码运算优化成零成本甚至被省略的指令。揭示编译器优化在高性能 JVM 工作负载中的作用与影响。
(94) Show HN: Frugon – Find which LLM calls a cheaper model could handle (local, MIT)
Frugon is a local-first tool that analyzes prompts to determine whether a cheaper model could handle the LLM call, helping teams cut inference costs without sacrificing quality. Built with MIT-licensed tooling, it helps teams optimize inference spend in real-world workloads.
Frugon 是一款本地优先的工具，能够分析提示并判断是否可由更便宜的模型处理某些 LLM 调用，从而降低推理成本。该工具使用本地托管、MIT 许可的组件，帮助团队在实际工作负载中优化成本。
(95) Amber the programming language compiled to Bash/Ksh/Zsh
Describes a language that compiles to shell scripts, enabling rapid automation and scripting; discusses trade-offs like portability and readability.
介绍一种可编译为 Bash/Ksh/Zsh 脚本的编程语言，强调在自动化与快速原型中的应用，同时也讨论可移植性和可读性等权衡。
(96) Lost and Found
Lost and Found centers on recovering misplaced items and reuniting them with their owners.
Lost and Found 专注于帮助找回丢失物品并将其归还给物主。
A contemplative look at Hiroshima's iconic mushroom cloud and what it reveals about history and memory.
对广岛蘑菇云的沉思，探讨其历史意义与记忆的启示。
(98) Preemption is GC for memory reordering (2019)
Preemption is GC for memory ordering by interrupting execution and breaking pathological reorderings. By forcing context switches and isolating critical sections, preemption reduces the need for heavy memory barriers and simplifies reasoning about concurrency.
抢占被视为内存重排序的一种垃圾回收机制，通过中断执行来打断可能引起的异常重排序。通过强制切换和隔离临界区，抢占可以降低对昂贵内存屏障的需求，同时简化对并发的推理。
(99) The footgun of right-to-left decorative characters
Decorative characters that run RTL can inadvertently affect layout, text rendering, and security-sensitive contexts, creating subtle UX bugs and potential spoofing surfaces. Awareness helps avoid confusing interfaces and security pitfalls.
RTL 的装饰字符可能无意中影响布局、文本渲染与安全上下文，带来微妙的用户体验 bug 与潜在的钓鱼表面。文中总结了在标识、界面标签或 URL 中使用此类字符时的具体坑点，提醒开发者避免界面混乱和安全隐患。
(100) Documentation is still in your Mum's filing cabinet
Organizing documentation for humans and AI requires clear structure, consistent tone, and searchable indexing. It advocates moving away from hidden, appendix-like docs toward centralized, navigable content with machine-readable metadata so assistants can summarize, answer questions, and generate docs.
为人和 AI 组织文档需要清晰的结构、一致的语气和可检索的索引。它倡导将文档从隐匿的附录转向集中化、可导航的内容，并配以可机器读的元数据，便于助手摘要、回答问题和生成文档。