Jun 27Sunday, June 28, 2026 · all daysJun 29 · today »
1.Marfa Public Radio Puts You to Sleep(marfapublicradio.org)
414 points by reaperducer 55 days ago | 130 comments | permalink
tl;dr: Marfa Public Radio launched a sleep podcast called "Marfa Public Radio Puts You to Sleep" for its fall membership drive, in which staff read aloud the dull operational documents (FCC compliance, NPR ethics codes, etc.) that keep the 24/7 station running. The goal is both to lull listeners to sleep and to encourage donations at marfapublicradio.org/donate.
HN Discussion:
  • Sharing similar sleep-inducing podcasts and media recommendations like fictional baseball broadcasts and boring book readings
  • Praising Marfa as a town and endorsing the public radio station's appeal
  • Personal sleep tricks and techniques that work better than boring audio content
  • The chosen subject matter (NPR ethics) is actually interesting and wouldn't induce sleep
  • Playful endorsement mimicking falling asleep while engaging with the concept
2.OpenRA(openra.net)
803 points by tosh 55 days ago | 163 comments | permalink
tl;dr: OpenRA's new playtest-20260222 introduces random map generators for Red Alert, Tiberian Dawn, and Dune 2000, usable in skirmish and multiplayer. Dune 2000 gets new visual effects, Starport bulk purchasing, and a community-led balance overhaul, while the standalone Tiberian Dawn HD mod is now feature-complete with toggleable remastered/classic assets. Other additions include map editor improvements, expansion-building bots, auto-save, new missions, and groundwork for localization.
HN Discussion:
  • OpenRA improves on the original with better balance and features
  • Praise and gratitude for OpenRA developers and the active player base
  • Nostalgic memories of playing the original Red Alert games
  • Open-source engine remakes of classic games are valuable broader trend
  • Publishers should open source older games like EA did with C&C
3.Anonymous GitHub account mass-dropping undisclosed 0-days(github.com)
941 points by binyu 55 days ago | 380 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • Reported vulnerabilities are unimpressive, just simple bugs not real 0-days
  • Term '0-day' is being misused; many are already-disclosed or trivial issues
  • AI-generated security reports tend to be bloated with false positives and non-issues
  • This mass-dropping reflects a transitional hype era for AI-based vulnerability discovery
  • Crowdsourcing LLM-found bugs without vetting is irresponsible and burdens maintainers
4.Fintech Engineering Handbook(w.pitula.me)
618 points by signa11 55 days ago | 213 comments | permalink
tl;dr: A handbook documenting patterns for building money-handling software, organized around three principles: no invented data, no lost data, and no trust. It covers money representation (precision, rounding, currencies, FX), ledger design (double-entry, event sourcing, immutability, audit trails), executing flows (idempotency, reservations, resumability), integrating with external systems (webhooks, outbox/CDC, reconciliation), access controls (segregation of duties, four-eyes), and testing strategies (property-based, crash injection, golden tests). Includes a glossary of fintech terminology for newcomers, and is presented as a living document open to contributions.
HN Discussion:
  • Handbook contains shallow or incorrect advice, possibly LLM-generated
  • Useful collection of known patterns, helpful as a starting reference
  • ~These principles apply to software engineering broadly, not just fintech
  • Handbook misses critical practical realities like reconciliation and external system uncertainty
  • Specific sections like idempotency keys provide genuine value to developers
5.Ford hired AI and sacked humans. It backfired badly(the-independent.com)
233 points by speckx 55 days ago | 4 comments | permalink
tl;dr: Ford rehired over 350 veteran engineers ("gray beards") after its push toward AI-driven quality control systems produced costly failures, with executives admitting the automation lacked the nuanced judgment needed for complex problems. Since bringing back human expertise, Ford topped J.D. Power's Initial Quality Survey among mainstream brands for the first time in 16 years. The company will continue using AI, but now paired with human oversight rather than as a replacement.
HN Discussion:
  • AI hype in boardrooms will keep causing failures until executives learn it's just a tool
  • Media is misleading readers by conflating computer vision inspection systems with LLMs
  • Corporations chronically undervalue experienced engineers and institutional knowledge
  • Other businesses will likely repeat Ford's mistake rather than learn from it
  • Rehired experts should leverage their position to negotiate much higher rates
6.Turn your site into a place people can bump into each other(cauenapier.com)
300 points by eustoria 55 days ago | 130 comments | permalink
tl;dr: The author built "Town Square," a tiny widget that appears at the bottom of a webpage showing stick-figure avatars of current visitors, who can walk around, see what others are reading, and chat ephemerally with no accounts or history. It's now open source with a public server available, so other sites can embed it without self-hosting. Future plans include connecting adjacent sites' Town Squares together, webring-style, so visitors can walk between them.
HN Discussion:
  • Nostalgic appreciation for recreating the old web's sense of human presence
  • Reminded of similar past projects with fondness, sharing related memories
  • Confusing UX with no clear way to participate or read messages
  • ~Prefers online tools that facilitate real offline social gatherings instead
  • Celebrates this as an example of quick experimental builds enabled by vibe coding
7.The case for physical media ownership(dervis.de)
473 points by cemdervis 55 days ago | 350 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • ~Ownership is about freedom to use/share, not physical form; DRM-free digital counts as ownership
  • Piracy is the practical solution to untangle restrictive licensing and ensure lasting access
  • Examples like Sony revoking purchased content reinforce the article's warning about licensed media
  • Physical media is no longer sufficient since discs require updates, accounts, or DRM
  • ~Article's argument is valid but weakened by poor presentation and LLM-like scatterbrained execution
8.AI learns the “dark art” of RFIC design(spectrum.ieee.org)
270 points by Brajeshwar 58 days ago | 174 comments | permalink
tl;dr: Princeton researchers are using reinforcement learning, inverse design, and diffusion models to automate RFIC design—a notoriously artisanal field where chips for 5G, radar, and satellite comms have traditionally been hand-crafted over years. Their AI-generated power amplifiers, which often look like QR codes rather than symmetric layouts, have achieved record bandwidth and efficiency while cutting design time from months to minutes. The main bottleneck now is training data, most of which sits locked behind corporate NDAs, prompting calls for open chip-design datasets akin to ImageNet.
HN Discussion:
  • ~This phenomenon of algorithm-designed uninterpretable circuits is decades old, not novel
  • ~The article overstates novelty; this is similar to genetic algorithms and brute force search
  • Questions robustness of AI-generated designs and whether conventional subblocks carry the weight
  • Conflating LLMs with traditional ML techniques muddies the discussion unfairly
  • Philosophical musing on whether nature's truths may be ugly messes only machines can grasp
9.Suspicious Discontinuities (2020)(danluu.com)
276 points by tosh 55 days ago | 100 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • Sharing additional real-world examples of discontinuities reinforcing the article's thesis
  • Personal anecdotes confirming the threshold-gaming behavior described
  • Proposing fixes like eliminating means-testing or phase-outs entirely
  • ~Offering benign explanations (e.g., pacers) for some observed discontinuities
  • Appreciating the visualizations and statistical patterns shown
10.The best response to AI slop and online noise is from Robin Williams(jayacunzo.com)
374 points by herbertl 55 days ago | 207 comments | permalink
tl;dr: Using Robin Williams' famous "Sistine Chapel" monologue from Good Will Hunting as a frame, the author argues that AI and online content suffer from the same flaw as Will Hunting: they "know" everything but have lived nothing. The piece contends that lived experience and personal perspective—the "little life moments" creators draw on—are what differentiate meaningful art and writing from AI slop. The call to action: stop hiding behind scripts and tools, and put more of your actual self into your work.
HN Discussion:
  • LLMs lack real lived experience, making the monologue an apt critique of AI
  • Storytellers create meaningful work without personal experience, undermining the article's thesis
  • The monologue is smug and overstates the value of lived experience over knowledge
  • ~The slop and devaluation of lived experience predates AI; this misdiagnoses the problem
  • The article itself feels like AI slop, making its argument ironic and self-defeating
11.Reducing tick density along recreational trails in Ottawa, Canada(sciencedirect.com)
231 points by bushwart 58 days ago | 173 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • Concern about pesticide health risks and environmental harm from permethrin/pyrethroids
  • ~Suggesting alternative tick control methods like tick tubes or nematodes
  • Personal protective measures like clothing are sufficient, reducing need for chemical intervention
  • ~Concerns about pyrethroid toxicity to aquatic life and runoff from treated areas
  • Support for more aggressive eradication approaches like gene drives
12.DSpark: Speculative decoding accelerates LLM inference [pdf](github.com)
791 points by aurenvale 56 days ago | 361 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • DeepSeek is praised for genuine innovation and open publishing compared to American labs
  • Users share positive real-world experience with DeepSeek models and integrations
  • This technique likely explains DeepSeek's ability to offer dramatically lower prices
  • Questions whether this is meaningfully novel versus 2022 speculative decoding work
  • Speculates about future proliferation of specialized small models for speculative decoding
13.Asian AI startups launch Mythos-like models(techcrunch.com)
267 points by bogdiyan 55 days ago | 193 comments | permalink
tl;dr: China's 360 unveiled Tulongfeng, a cybersecurity-focused AI model pitched as a rival to Anthropic's Mythos, while Tokyo-based Sakana AI launched Fugu, an orchestration model designed to coordinate access to multiple models via APIs. Both launches follow a US export ban two weeks ago blocking non-Americans from accessing Anthropic's Mythos and Fable 5. Sakana frames Fugu as a hedge against US access risks rather than a replacement, while 360's founder cast vulnerability-finding AI as a national strategic asset.
HN Discussion:
  • Fugu underperforms Opus while being slow and expensive in real-world testing
  • Fugu is an orchestration system routing to other models, explaining the high costs from double API markup
  • Calling these models 'Mythos-like' is meaningless without reliable benchmarks for comparison
  • ~Expect US to ban foreign LLMs under vague safety pretexts regardless of actual performance
  • ~Mythos being unavailable conveniently makes the 'Mythos-like' claim unfalsifiable
14.IP Crawl: Living atlas of open webcams discovered on the public internet(ipcrawl.com)
327 points by arm32 55 days ago | 162 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • Site is a privacy invasion that exploits ignorant users and should be taken down
  • Disturbing voyeuristic feeling watching strangers' private lives unknowingly
  • Confirms longstanding truth that people will connect anything to the internet insecurely
  • Sharing amusing or interesting specific camera finds from the site
  • ~Author should add alerting system to notify exposed camera owners
15.Why does kinetic energy increase quadratically, not linearly, with speed? (2011)(physics.stackexchange.com)
375 points by ProxyTracer 56 days ago | 207 comments | permalink
tl;dr: Two intuitive arguments derive KE ∝ v² without invoking work or mgh. The first uses a spring pushing two equal-mass boxes apart, combined with conservation of momentum and Galilean invariance of potential energy, to show KE(2v) = 4·KE(v). The second uses a free-falling object in a constant gravitational field, applying energy conservation in both directions (catching at quarter-heights vs. launching in quarter-energy increments) to squeeze KE(v) = 4·KE(v/2) from two opposing inequalities.
HN Discussion:
  • Potential energy conversion provides the most intuitive explanation via height-to-speed comparison
  • Calculus-based derivation from F=dp/dt and E=F·dx is the clearest intuition
  • Symmetry-based arguments (Galilean invariance) are the deepest explanation, aligning with the article's approach
  • Constant force yields quadratic distance over time, so energy scales quadratically with velocity
  • The article doesn't truly answer the intuitive question of why going faster requires disproportionately more energy
16.Zuckerberg's war on whistleblowers(pluralistic.net)
752 points by HotGarbage 55 days ago | 284 comments | permalink
tl;dr: Meta is waging an escalating legal campaign against Sarah Wynn-Williams, author of the bestselling Facebook memoir "Careless People," using NDAs and a company-paid arbitrator to extract $50,000 per criticism (already over $11M) and silence her completely. The absurdity peaked when Meta claimed her sitting silently and motionless on stage at the Hay Festival violated her agreement, prompting her to sue to invalidate the contract. Doctorow argues Meta is willing to fuel the Streisand Effect because terrorizing Wynn-Williams sends a chilling message to thousands of recently laid-off employees who might otherwise speak out.
HN Discussion:
  • Meta's aggression suggests fear of worse undisclosed information being revealed
  • ~Zuckerberg's behavior is driven by ego and pettiness rather than strategy
  • The campaign is deliberate intimidation to discipline current and former employees
  • NDAs and arbitration clauses are standard practice and this isn't true whistleblowing
  • Zuckerberg should be labeled malicious rather than excused as bizarre or out-of-touch
17.Previewing GPT‑5.6 Sol: a next-generation model(openai.com)
1121 points by minimaxir 56 days ago | 735 comments | permalink
tl;dr: Summary not available
HN Discussion:
  • Excitement about high-speed inference on Cerebras and frontier intelligence delivery
  • Criticism of pricing tier shuffling forcing users to more expensive models
  • Concern about the model's elevated cheating/reward-hacking behavior in evaluations
  • Skepticism that this is just a minor version bump dressed up as next-generation
  • Questioning the value and mechanics of the new ultra mode and marketing framing
18.Streaming services' obnoxiously loud ads become illegal on July 1 in California(arstechnica.com)
284 points by speckx 55 days ago | 93 comments | permalink
tl;dr: Starting July 1, California law extends the federal CALM Act's ad-volume rules to streaming services, banning obnoxiously loud ads. The MPA and Streaming Innovation Alliance opposed the bill, citing technical challenges with server-side ad insertion across varied encoding pipelines and output devices like TVs, tablets, and phones. Compliance will likely require streamers to integrate file-based and real-time loudness controls into their ad workflows—similar to existing broadcast practices, which still drew 1,700 FCC complaints in 2024.
HN Discussion:
  • Streaming services' technical complaints are unconvincing excuses for not fixing their own systems
  • Similar loudness/brightness issues exist on other platforms and should also be addressed
  • This closes a needed loophole already addressed for broadcast TV
  • Personal experience confirms ad loudness is a real and annoying problem
  • Questions or reflections on the legislative approach and technical definitions involved
19.An entire Herculaneum scroll has been read for the first time(scrollprize.org)
1697 points by verditelabs 57 days ago | 365 comments | permalink
tl;dr: Researchers have virtually unwrapped and read PHerc. 1667, a carbonized Herculaneum scroll, end-to-end for the first time, using high-resolution X-ray microtomography at ESRF's BM18 beamline combined with machine learning to detect ink. The recovered text—roughly 22 columns of Greek—appears to be a 2nd-century BC Stoic treatise on ethics referencing Aristocreon, disciple of Chrysippus. The team also confirmed prior readings of Scroll 1 via direct 3D ink detection and identified another scroll as Philodemus's *On Gods, Book 8*; all data, code, and transcriptions are openly released.
HN Discussion:
  • Insider team members offering to answer questions and sharing additional breakthroughs
  • Awe at the historical and temporal significance of recovering ancient text
  • Optimism that this represents tech/AI being used for genuinely meaningful purposes
  • Excitement about future potential, especially discovering more scrolls in unexcavated areas
  • ~Skepticism about how translation tone may distort the original writer's voice
20.Om Malik has died(om.co)
1343 points by minimaxir 57 days ago | 169 comments | permalink
tl;dr: Veteran tech journalist and venture capitalist Om Malik, founder of GigaOm and partner at True Ventures, has died. Known for his influential coverage of Silicon Valley from the early 2000s onward, his writing on broadband, Web 2.0, and the tech industry, as well as his photography and personal essays, Malik was widely respected for his insight, integrity, and generosity to founders and fellow journalists. The post has drawn tributes from a wide swath of the tech and journalism community.
HN Discussion:
  • Personal anecdotes of Om's generosity and mentorship to strangers and newcomers
  • Om was a pioneering, honest voice in tech blogging who lifted others up
  • Admiration for his authentic, jargon-free writing style and journalistic integrity
  • Shock and sadness at his death at a relatively young age
  • Sharing links to his final writings and reflections on his legacy