Hacker News, Distilled

Status of copyright for AI‑generated code

• Many note the US Copyright Office’s position: works “predominantly” generated by AI without meaningful human authorship are not copyrightable. What counts as “meaningful” is unresolved.
• Some argue that prompting, reviewing, and editing AI output can be enough to create a new, copyrightable work; others respond that such code is at best a derivative work, not a fresh copyright.
• Image cases (e.g., Midjourney‑generated comics) are cited: human text got copyright, AI images did not. Several argue code will be treated similarly.
• Others stress that agency rulings and one circuit’s decisions are not nationwide Supreme Court precedent; law remains unsettled, especially on “how much” human input is sufficient.

Employer ownership, work‑for‑hire, and trade secrets

• Consensus: models aren’t legal persons and can’t own IP.
• Ownership today mostly flows from employment and enterprise contracts: the company that directs the work and pays for the tools typically owns whatever rights exist.
• If code is uncopyrightable, contracts can still treat it as confidential work product or trade secret, but that’s weaker: once leaked, anyone else can freely use it.

Training data, infringement, and license contamination

• Strong concern that LLMs are trained on copyrighted and copyleft code (GPL/LGPL, textbooks, GitHub), enabling “copyright washing” of OSS.
• Others argue AI “learns” like humans do, not simply copy‑pastes, though counterexamples of regurgitated code and comments are mentioned.
• Debate over whether provenance from GPL/LGPL/BSD code “travels” into outputs; no clear case law yet, but some assume courts will treat infringing outputs like any other derivative work.

Practical risk, enforcement, and M&A

• Some claim this is mostly academic: very few lawsuits so far, and enforcement (especially of GPL) is rare and expensive.
• Others say the concrete pressure will come from M&A and fundraising: acquirers already ask about AI usage and license contamination; inability to prove human authorship or clean licensing can jeopardize deals.

Ethical views and the commons

• One camp sees AI as accelerating enclosure and exploitation of creators; another sees it as undermining overbroad copyright and pushing more artifacts into the commons, closer to copyright’s original limited‑term bargain.

Impact on software practice and liability

• Developers report “vibe‑coded” codebases, weaker reviews, and erosion of shared understanding when AI writes and reviews most code.
• Others celebrate faster “lone‑wolf” development with agents as power tools.
• On liability, most argue nothing fundamental changes: organizations remain responsible for shipped code, regardless of whether a human or an AI wrote it.

Unclear / open questions

• Exact threshold for “meaningful human authorship.”
• Whether employees can freely publish uncopyrightable, AI‑generated work made at their job.
• How courts will handle provable LLM regurgitation of protected code at scale.

Project concept and intent

• Creator built “Middle Class Museum” as a browser-based, satirical walk-through of things that “used to be affordable” (homes, pensions, cars, Blockbuster, etc.).
• Implemented with vanilla HTML/CSS/JS, no backend.
• Author emphasizes it’s meant as humor/satire, not a rigorous finance explainer, but acknowledges criticism and considers adding “satire” and “not inflation adjusted” disclaimers and possibly revising numbers.

Inflation, wages, and affordability

• Many commenters argue that failing to adjust for inflation makes the comparisons misleading or “disinformation.”
• Some show that old prices, once adjusted, are closer to today’s prices (e.g., 1980 house or car costs) and say that median wages have roughly kept pace overall.
• Others insist that even inflation-adjusted, core necessities (housing, healthcare, education, childcare) have outpaced wages, while many consumer goods (TVs, computers, software) have become cheaper.
• There is sharp disagreement on whether “life was more affordable in the 80s/90s,” with some asserting clear decline in affordability of basics and others saying standards and expectations have simply risen.

Quality, features, and expectations over time

• Critics note that old houses, cars, and products were smaller, less feature-rich, and often less reliable or safe.
• Others counter that older items were easier/cheaper to repair, communities did more DIY maintenance, and that “seatbelts and airbags” shouldn’t be the main benchmark for progress.
• Several point out that people now expect larger homes, SUVs, and more amenities, which drives cost.

Housing, cars, and big-ticket items

• Housing: debate centers on land scarcity, zoning/NIMBYism, and lack of new starter homes/condos. Some say you can still find sub‑$300k homes in many cities; others stress regional crises.
• Cars: comparisons of 1980s station wagons vs today’s compacts/SUVs, interest rates, loan terms, and the shift to leasing/financing. Disagreement over whether cars were really cheaper “in real terms” and whether Americans simply prefer bigger, more expensive vehicles now.

Pensions, retirement, and financial vehicles

• Some nostalgically highlight defined-benefit pensions; others emphasize they were risky (employer insolvency, fraud) and prefer modern defined-contribution plans with low-cost index funds.
• Discussion touches on rising life expectancy making pensions costlier, deferring retirement for higher payouts, and the role of tax-advantaged accounts.

Broader economic and political framings

• Thread includes debates over Marx, labor theory of value, and whether workers should be able to afford what they produce.
• Some blame corporate concentration, tax avoidance, and deregulation for rising costs of essentials and advocate redistribution and unions; others argue narratives of oppression are exaggerated or misleading.
• Monetary policy and stimulus are briefly contested, especially around inflation causes.

Everyday costs, subscriptions, and lifestyle

• Comparisons between Blockbuster rentals vs streaming subscriptions generally favor the present on cost and convenience.
• Some argue people overpay via lifestyle choices (premium phones, streaming bundles, coffee, frequent upgrades) rather than structural unaffordability alone.
• Others stress that cheap entertainment doesn’t offset high housing, healthcare, and education costs.

Design, UX, and factual nitpicks

• Complaints about horizontal scrolling not working well with scroll wheels and copy/paste being disabled.
• Corrections raised about TSA shoe rules, modern savings account yields, current car subsidies, and specific numeric errors (e.g., student loan averages).

Azure, multi-cloud, and Microsoft’s role

• Many read “path to multi-cloud” as an implicit admission Azure alone is insufficient, especially for GitHub-scale reliability.
• Some note other Microsoft properties (e.g., LinkedIn) reportedly backed away from full Azure moves; others say multi-cloud is normal for resilience/vendor-independence.
• Several argue the Azure migration itself likely worsened reliability vs GitHub’s own datacenters; others counter that moving to any cloud is complex but not inherently doomed.

AI/agent-driven load and usage patterns

• Commenters accept that “agentic workflows” and LLM-generated code are driving huge spikes in commits, repos, and PRs.
• Debate over value: some say most of this code “goes to die” or is low-value “vibe coding,” not reflected in better end-user software.
• Several argue this traffic should be metered or priced (e.g., per-commit or per-usage), instead of subsidized as “free” load.

Reliability, user impact, and pricing

• Multiple users report frequent downtime and degraded performance (Actions, search, PR lists, UI glitches), sometimes costing real workdays.
• Paying customers complain that service quality is dropping while per-developer pricing stays high, especially given many workloads are just “static text files.”
• Some are migrating or considering moves to GitLab, Gitea/Forgejo, or self-hosting; others say GitHub’s functionality and ecosystem remain compelling.

Architecture, scale, and technology choices

• Discussion of GitHub’s historic Ruby-on-Rails monolith vs microservices; some nostalgically associate the monolith era with better stability, others say scale alone invalidates simple comparisons.
• Noted efforts: moving webhooks off MySQL, redesigning auth/session flows, shifting hot paths from Ruby to Go, dealing with monorepo pain, and general database/backend migrations.
• Some see frontend complexity (e.g., highly componentized React diff views) as a symptom of wider engineering/culture problems.

Graphs, metrics, and trust

• The unlabeled y-axes in GitHub’s growth charts are widely criticized as manipulative or meaningless; several call this “PowerPoint graphs.”
• New per-service uptime numbers imply only ~97% end-to-end availability; users say this matches their experience and is unacceptable for critical tooling.
• Many find the blog’s tone platitudinous (“we hear you”) and lacking hard numbers, concrete timelines, or genuine accountability, further eroding trust.

Billing change & immediate reactions

• Many were unaware Copilot code review had been using Actions for free; some stopped subscriptions when code review began triggering Actions and slowing reviews.
• Users object to non-Actions activity consuming Actions minutes and to code review now charging both AI credits and Actions minutes, seen by several as double billing and cost obfuscation.
• For at least one 300k LoC repo, a Copilot review uses ~5–10 minutes of Actions time, raising cost concerns for teams with many PRs.

GitHub strategy, goodwill, and “rug pull” concerns

• Multiple commenters frame this as part of a broader trend: AI features initially subsidized to build dependency, then repriced sharply upward (“bait and switch” / “rug pull”).
• Some think this is simply charging closer to true cost; others emphasize GitHub/Microsoft’s size and argue that a de‑facto critical infrastructure provider should behave more like a public utility.
• Perceived “enshittification” of GitHub is a strong theme, with people recalling earlier simplicity and goodwill.

AI economics & pricing debates

• Extensive argument over whether inference is currently profitable:
  • One side: inference alone is likely profitable; training and capex are the real money sink; API token prices may already include margin.
  • Other side: as long as training and subsidies aren’t covered, current prices are still below true economic cost.
• Subscriptions are widely seen as heavily subsidized versus equivalent API usage.
• Many expect further price hikes as funding tightens and providers try to recoup training and infrastructure costs.

Reliability, Actions, and CI/CD alternatives

• GitHub Actions is criticized for reliability (measured availability under 99% and incidents being undercounted), complexity, and security issues.
• Some still find Actions “good enough” and convenient; others report more outages than with self‑hosted CI.
• Solo and small-team developers discuss moving to self‑hosted or cheaper CI (GitLab runners, Forgejo, Woodpecker, Drone, TeamCity, Gitea, home hardware), but note trade‑offs in idle cost, parallelism, and complexity.

Data, competition, and local models

• Concerns about sending code to foreign AI providers versus US‑based ones; perceived IP and regulatory risks differ by jurisdiction.
• Open‑weight and local models are discussed as a future escape hatch, but many say current local models are notably weaker and often not cheaper when hardware and electricity are included.

Pricing, Value, and Market Positioning

• Many see the Seal 08’s spec sheet (≈$42k in China, 1,000 km claimed range, 684 hp) as an exceptionally strong value.
• Expectation that European prices will be 50–200% higher due to taxes and margins, yet still undercutting comparable European EVs.
• Desire expressed for similar battery tech in cheaper, down-market models (~€30k, >750 km real range).

Range, Use Cases, and Test Cycles

• Some question the need for 700–1,000 km range in daily life; others argue it removes anxiety when the car isn’t fully charged and for long trips.
• Real-world range is debated: China’s CLTC test is described as more optimistic than WLTP/EPA; direct comparisons with BMW/Mercedes require same-cycle or real-world testing.

Charging Infrastructure and 1 MW “Flash Charging”

• Enthusiasm for 5–10 minute “flash charging” and BYD’s Blade 2.0 pack, especially if rolled out across everyday locations (e.g., convenience stores, fast food).
• Proposal: stations buffered by large on-site batteries (often using second-life packs) plus moderate grid draw and some solar, to mimic gas-station throughput.
• Counterarguments stress that 1 MW per stall is a huge power level; scaling to “many small lots” requires substantial grid upgrades, transformers, and local storage, which may take decades outside China.
• Clarification that early 1 MW sites in China already rely on internal batteries to smooth grid demand.

EV vs ICE Economics and Policy

• Several argue EVs have far lower energy and maintenance costs than ICE, especially with home charging; others note high electricity prices and commercial fast-charging fees can narrow this gap.
• Debate over carbon policy: some favor tech-neutral carbon taxes; others see EU/DE mandates as coordination tools to save domestic automakers.
• Discussion of battery material origins (e.g., Australia via China) and extended producer responsibility / battery recycling in China.

Industry Competition and Geopolitics

• Repeated theme that Chinese EV makers (especially BYD) are outpacing US/EU in model variety, scale, and technology; Beijing Auto Show cited as evidence.
• Concern that Western protectionism (e.g., restricting Chinese EV imports) reflects inability to compete.
• Some see EU/US legacy makers as oversized, expensive, or slow; others point to ongoing efforts like BMW’s Neue Klasse.

Tesla, Autonomy, and Future of Car Ownership

• Strong disagreement over Tesla’s prospects:
  • One side: Tesla is “cooked,” Chinese EVs have won on cost and scale; Tesla’s valuation no longer justified.
  • Another side: Tesla is pivoting to robotaxis and “cybercab”/“cybervan” concepts, anticipating a collapse in consumer car ownership as self-driving ride-hail becomes cheaper per mile.
• Long subthread on whether self-driving fleets will significantly reduce private car ownership:
  • Pro-AV view: shared AV rides could be ~40¢/mile vs ~75¢/mile for personal EVs; many households, especially lower-income and infrequent drivers, might give up cars or reduce from 2–3 cars to 1.
  • Skeptical view: people value private cars as personal space, storage, and brand; dislike shared, potentially dirty vehicles; cost savings may not be decisive.
  • Questions raised about vehicle durability (million‑mile batteries, suspensions, interiors) needed for high-utilization fleets.

Technical and Safety Aspects of Batteries and Charging

• Blade 2.0 praised as a mechanical/pack-design improvement that maximizes cell fraction and cooling. Pack size (~92 kWh) seen as large but not unprecedented.
• Some characterize such packs as “reusable bombs”; others counter that gasoline tanks store far more energy and are more explosive, while EV batteries are primarily fire hazards with less explosive gas volume.
• Discussion of 1 MW charger specifics: ~1,000 V and 1,000 A; actual operating voltage/current negotiated between car and charger.

Tires, Weight, and Maintenance

• Mixed anecdotes on tire wear: some EV owners report high wear (due to weight and torque, plus “fun to drive hard”), others report normal lifetimes (~80,000 km).
• Clarification that EVs still require maintenance (gearbox oil, steering, bearings, etc.), but consensus leans toward lower overall maintenance than ICE.

Charging Reality vs Marketing Claims

• Concern that advertised 400 km in 5 minutes assumes ideal 1 MW chargers, which are rare outside China.
• Some European drivers report routinely achieving >200 kW on existing 250–350 kW chargers; others (notably in the UK) report difficulty finding public chargers that deliver >100 kW “consistently.”
• Thread consensus: infrastructure is lagging but improving; China is moving fastest, with other regions expected to follow over years, not months.

Personal outcomes from quitting or cutting back

• Many report long stretches of abstinence (months to years) and say they don’t miss alcohol, sometimes wishing they’d quit decades earlier.
• Others stopped unintentionally (just lost desire) and noticed little or no obvious benefit.
• A subset returned to occasional drinking after long breaks and felt they could now keep it rare and controlled.
• Some never drank much in the first place (a few drinks per year) and don’t see alcohol as central to their lives.

Sleep, health, and weight

• Better sleep is a recurring theme: fewer awakenings, more restful nights, and improved sleep-tracking metrics after stopping. Some note effects extending for days after drinking.
• Several mention weight loss that occurred “automatically” after quitting, attributed to lost liquid calories and associated snacking.
• Others see clear improvements in bloodwork and inflammation markers; some are unsure which lifestyle changes mattered most.
• There are strong anecdotal warnings linking heavy drinking to serious illness (colon/breast cancer, pancreatitis, neuropathy), alongside reminders not to overgeneralize from single cases.
• A few argue that for low, infrequent consumption the marginal health benefit of total abstinence may be small.

Social life and culture

• A common downside: social events feel less fun when sober around drinkers; some find work dinners with drinking colleagues “a special kind of hell.”
• Others note that social norms are slowly adapting, with more respect for non-drinkers and better non-alcoholic options.
• Several link their decision to hobbies (e.g., early-morning cycling, running) that are incompatible with hangovers.

Control, addiction, and psychology

• Experiences diverge: some find moderating trivial; others see “one drink” as a cliff, not a slope.
• Commenters highlight self-medication for anxiety, trauma, ADHD, and sensory overload; for these people, quitting is much harder than for casual social drinkers.
• Replacement compulsions (especially sugar) are common after quitting.

Substitutes and “having a thing”

• Many resonate with wanting “a thing” more than alcohol itself.
• Popular replacements: non-alcoholic beer, coffee, a wide variety of teas, kombucha/kefir, sparkling water, and desserts—though some worry about swapping in unhealthy sugar.

Overall reaction to the essay

• Many readers find the message inspiring, humane, and “the best honest advice” they’ve seen, especially the emphasis on ethics, love over fear, and caring about craft.
• Others see it as naive or nihilistic: too pessimistic about industry, too absolutist on AI, and not practical for students who need jobs and have debt.
• There’s debate over whether publishing this required “courage”; some say there are real career risks in academia, others see little downside for a tenured professor.

Ethics, responsibility, and education

• Multiple commenters stress the importance of explicit ethics training, citing engineering disasters and software safety case studies.
• Others are cynical: mandatory ethics courses often become box-ticking or “communications” classes and don’t meaningfully shift behavior.
• Some describe leaving tech for more overtly ethical fields (e.g., nursing), framing software as structurally misaligned with public benefit.

Generative AI and “LLM vegetarianism”

• The essay’s categorical refusal to use LLMs is highly polarizing.
• Supporters see LLMs as built on labor exploitation, unlicensed data use, and high resource consumption, and as tools that outsource thinking and erode agency.
• Critics call this hyperbolic, inconsistent with using modern hardware, and suspect shifting goalposts even if energy costs drop or training data is “clean.”
• A few hope for “ethically trained” small models that such people could study without compromising principles; others argue there is no obligation to engage with LLMs at all.

Craft, refactoring, and “going slowly” vs industry reality

• One camp embraces the advice: deep thinking, refactoring, and documentation are seen as key to maintainability, profitability, and genuine engineering. “Slow is smooth and smooth is fast.”
• Another camp argues this is misaligned with commercial incentives: entry-level engineers who “go slowly” and polish endlessly risk getting fired or never hired.
• Several note that industry often values shipping and “product as the artifact” over code as craft; automatic coding tools and agents intensify this trend.
• A recurring theme: high-level system designers who can’t code are already ineffective; relying on LLMs without technical depth will be worse.

Deep work, distraction, and life habits

• Many resonate with the call to carve out distraction-free time; some say they only grasped how pervasive distraction is once they tried to fight it.
• Exercise and reading are cited as surprisingly powerful enablers of deep work and better time use.

Academia vs industry and preparing students

• Some criticize academics with little or no industry experience for giving career advice; they see the essay as detached from “messy” commercial constraints.
• Others reply that education isn’t solely job training, and that students should be encouraged to define success beyond “succeeding in this market.”
• There’s disagreement on whether ignoring current trends (especially LLMs) is responsible preparation or principled self-marginalization.

Luddism, inevitability, and tech pessimism

• Several label the stance “Luddite”; defenders counter that Luddites had coherent, labor-focused critiques and that opposing harmful tech is rational.
• Some warn against accepting narratives of inevitability (“we’re never going back to manual coding”); others argue that economically powerful automation will not be reversed.
• Broader worries surface about “move fast and break things,” enshittification, surveillance, and software’s role in social harm.

Careers, money, and meaning

• One axis of disagreement: study CS for beauty, curiosity, and social good vs. study it primarily as an income-producing skill.
• Some argue non-instrumental motivations are a luxury if you’re not rich; others share anecdotes of leaving software for lower-paid but more meaningful work and not regretting it.
• The essay’s core challenge—setting ethical boundaries early and caring more about people than profit—is seen by some as necessary moral grounding, and by others as incompatible with current hiring and productivity expectations.

Local hardware & deployment

• Several commenters discuss VRAM needs. 20GB is borderline for 13B BF16 weights, though splitting layers across CPU/GPU via llama.cpp is possible but slower.
• Some compare high‑VRAM GPUs vs large shared‑RAM desktops; consensus: GPUs give more “usable” local LLMs, but you won’t “make your money back,” so buy what you’re happy to pay for.
• No GGUF is yet available; people note it should be convertible from the PyTorch checkpoint for use with tools like Ollama.

“Vintage” concept, data leakage & contamination

• The authors frame “vintage LMs” as trained solely on pre‑cutoff data to avoid benchmark contamination and post‑date knowledge.
• Commenters point out evidence of temporal leakage (e.g., anachronistic political facts, terminology, and future knowledge), arguing the model doesn’t fully meet its own “vintage” standard.
• Distinction is drawn between contamination by benchmark answers vs generic post‑cutoff text; some see them as nearly the same issue.

Behavior, style & capabilities

• Many are charmed by the 19th/early‑20th‑century prose: ornate, confident, discursive, and very different from modern LLM tone.
• Examples show it:
  • Treats “computer” as a human job and distinguishes “digital” as “using fingers.”
  • Gives period‑typical takes on India, empire, American Civil War causes, women, yoga, industrialization, etc.
  • Produces speculative future visions (2025/2026, moon travel, computers) that feel like historical futurism.
• Users note a common pattern: first sentences may be accurate; then it drifts into plausible but wrong explanations, so it can “pollute your brain” if you don’t know the answer.

Historical bias, racism & ethics

• Commenters report explicitly racist, colonialist, and sexist outputs and stress that these reflect the surviving texts and power structures of the era.
• Some see this as historically honest and even desirable for future “uncensored” historical models; others find it troubling and question the value of partial moderation layered on top.

Epistemic snapshot & scientific testing

• Strong interest in using such models as “time capsules” or “epistemic snapshots” of a given era, comparable to other history‑only LLM projects.
• Several propose research uses: training models before key breakthroughs (e.g., relativity, nukes) to see whether they can rediscover them or predict events, though many doubt current LLMs could.

Speculation, simulations & future models

• People imagine combining era‑locked models with VR or personal archives to simulate past periods or one’s younger self, edging toward “time travel” or “simulation” experiences.
• Some are excited; others push back on simulation talk as philosophically dubious or psychologically risky.

Cost and practicality

• Back‑of‑the‑envelope FLOP and cloud‑pricing estimates suggest pretraining costs on the order of tens of thousands of dollars, seen as impressively affordable for bespoke models.

Nature of the device and “first time” claims

• The device behaves like a Stingray/fake base station but was used to push phishing/spam SMS rather than traditional surveillance.
• Several commenters argue media coverage was sensational, noting similar tech is already used by governments and regulators.
• Others clarify that officials likely meant this was the first documented use for fraud in Canada, not the first existence of such devices overall.
• Some remain skeptical, comparing this to pretending large-scale scam/spam operations are “new.”

---

Telecom protocol weaknesses

• Core issue: phones often connect to the strongest cell without authenticating the tower, especially on 2G (GSM).
• 2G lacks mutual authentication and can be forced via downgrade attacks; turning off 2G is possible on many Android devices and in iOS Lockdown Mode, but not generally exposed or default.
• There is debate whether SIMs/carrier profiles can effectively hide or disable 2G/3G; some reports suggest behavior differs by carrier and country.
• Commenters emphasize that SMS sender identity is weakly verified and relies heavily on carrier trust.

---

Motives for using an SMS blaster

• Avoids carrier spam filtering and other anti-abuse systems.
• Provides highly localized targeting of nearby devices.
• Leaves little or no trace in carrier logs, since messages never traverse the operator network in the normal way.

---

Law enforcement, government use, and double standards

• Some argue it’s hypocritical for authorities to treat this as “unprecedented” while using similar gear themselves and rarely prosecuting official misuse.
• One view is that prosecutors truly “haven’t seen” such devices only because cases against agencies don’t get brought.
• Others claim the strong reaction is less about protecting users from spam and more about preventing unsanctioned, potentially encrypted communications outside approved channels.
• There is specific concern about blocking or mishandling 911 calls; some speculate the attackers would have forwarded emergencies, others are doubtful.

---

Relation to SIM farms and grey markets

• SIM farms (racks of SIMs/phones) are distinguished from SMS blasters:
  • SIM farms use legitimate network access for grey-market VoIP, 2FA receipt, “grey route” SMS, or even research against botnets.
  • SMS blasters impersonate base stations, harvest numbers in real time, and inject messages directly.
• US examples of SIM farms provoke debate about how actively law enforcement and regulators respond; some cite significant FCC fines, others see lax enforcement.

---

Global scope and user impact of spam

• Similar SMS-blaster scams are reported in several countries (e.g., Switzerland, NZ, France), sometimes allegedly involving foreign operators; actual attribution is described as unclear.
• Multiple commenters describe pervasive spam calls/SMS in places like Brazil, India, and the US, driving users toward apps like WhatsApp.
• “Flash” or class‑0 SMS (full-screen, ephemeral messages) are noted as another abused channel that can appear like system notifications.
• Many report coping strategies such as silencing unknown callers and ignoring traditional telephony, which some see as evidence the public phone system is becoming dysfunctional.

---

Mitigations and open questions

• Suggested mitigations:
  • System-level toggles to disable 2G without extreme “lockdown” modes.
  • Stronger UI warnings or indicators when connected to downgraded/suspicious towers.
  • Cryptographically secured SMS with carrier-backed certification, though commenters note this doesn’t solve 2G fallback.
• Several participants highlight the tension between stronger telecom security and perceived government/carrier incentives to keep networks wiretap- and metadata-friendly.

Perceptual Differences and Anecdotes

• Many describe recurring arguments (often with partners/family) over whether objects are blue vs green, turquoise vs blue, blue vs gray, etc.
• Some find the test result (e.g., “greener than 95%” or “bluer than 98%”) matches long‑standing disagreements in real life.
• Others say they cannot honestly label mid-range hues as either blue or green; they experience them as “teal/turquoise/cyan” and end up guessing or quitting.

Test Design and Methodology Critiques

• Core complaint: forced binary choice between blue and green with no “neither/both/unsure” option, especially when shown obvious cyan/teal.
• Several argue this measures linguistic classification, not raw color perception, and is “scientifically junky” as a single-item forced-choice instrument.
• Concerns about anchoring and binary search: prior colors bias later answers; repeated runs can yield different boundaries.
• Some suggest Likert-style “more blue vs more green,” multi-point sliders, randomization, context colors, or intermediate “palette cleansers” to reduce adaptation.

Devices, Calibration, and Environment

• Many note results vary significantly across monitors, phones, brightness levels, HDR/SDR, blue-light filters, Night Shift/f.lux, and ambient lighting.
• Some test across devices and see large shifts; others report factory-calibrated displays are often “good enough,” but cheap panels can be wildly off.

Biology and Individual Variation

• Reports of different color balance between left/right eyes, cataract implants, age-related yellowing of lenses, and mild color vision deficiencies.
• Red–green colorblind participants share results, sometimes surprisingly near the population median.
• Discussion touches on cone sensitivities, opponent color processing (blue–yellow, red–green), and how that yields “families” like cyan or magenta.

Language, Culture, and Color Categories

• Extended discussion of languages that group blue and green, or split blue into distinct basic terms, and how this shapes categorization.
• Examples include historical lack of distinct words for blue/green or orange, culturally specific terms (e.g., traffic lights called “blue”), and banknote colors.
• Some emphasize that “turquoise/cyan” may be a basic category for them, making the test’s framing feel wrong.

Philosophy of Perception (Qualia)

• Multiple comments revisit the classic “is my blue your blue?” and qualia/inverted spectrum questions.
• Some argue there may be no absolute inner color, only learned relations among stimuli; others highlight this as part of the “hard problem” of consciousness.

Perceived Hypocrisy and Distrust

• Many commenters see the “principles” as marketing spin rather than binding commitments.
• There is strong skepticism that a profit‑driven AI company will prioritize anything above growth, influence, and valuation.
• Several argue that the organization’s history (nonprofit → capped‑profit → more conventional for‑profit behavior) shows it lacks stable or meaningful principles.

Democratization and Openness

• The stated goal of “democratization” is widely criticized as incompatible with closed models, proprietary data, and centralized control.
• Commenters suggest the only convincing proof of democratization would be open‑sourcing models and research or at least building a genuinely open ecosystem.
• Some note that framing “we will democratize” implies a gatekeeping role: deciding who is “empowered” and on what terms.

AI, Power, and Inequality

• Many see AI as likely to concentrate power and wealth among a small elite (founders, investors, large corporations), not “universal prosperity.”
• Concerns include permanent dynasties of AI‑owners vs. a precarious majority with weakened bargaining power and fewer jobs.
• Comparisons are drawn to existing failures to distribute cheap, effective medical treatments: tech progress does not guarantee fair access.

Military, Surveillance, and “Kill Bots”

• Commenters highlight the absence of explicit commitments not to support autonomous weapons, mass surveillance, or cyber‑warfare.
• Some argue military and security uses are among the most lucrative and closely aligned with the capabilities of AI systems, raising doubts about any unwritten restraint.

Optimistic Visions vs. Allocation Problems

• A minority of comments articulate a hopeful scenario: AI‑driven robotics, automated farming, cheap energy, advanced medicine, and near‑zero‑cost services (education, healthcare).
• Even those acknowledging this possibility question whether existing economic and political systems can translate such productivity into broadly shared flourishing.

Corporate Strategy and Timing

• The release is widely interpreted as reputation management: pre‑IPO positioning, response to recent controversies, lawsuits, or military contracts.
• Some see it as an attempt to reassure employees and the public without changing underlying incentives; others dismiss it as “principles that will be revised whenever convenient.”

Past Promises and AGI Race

• Earlier commitments to halt competition and assist a “more advanced, safety‑conscious” AGI project are recalled.
• Several commenters doubt these would ever be honored in practice, or expect conditions to be defined such that they never trigger.

Scope and motives of the WotC / MTG Arena union

• Many assume unions form only when workers feel underpaid, overworked, or insecure; several infer that Arena staff must be unhappy with layoffs, RTO mandates, and AI pressure.
• Supporters stress unions as a way to rebalance bargaining power, not just to protect jobs, but to negotiate on workload, mental health, benefits, IP rights, and process issues.
• Critics argue unions are less compelling where pay and conditions are already good, and worry that unions can entrench poor performers and create rigidities.

US vs Europe, tech vs games

• Europe is cited as having high union coverage; US tech is seen as relatively non-union due to high salaries, mobility, outsourcing risk, and strong anti-union culture/propaganda.
• Others counter that games industry pay and conditions are notably worse than big-tech SWE, making it a natural site for organizing.
• Debate over whether industry structure favors unions: some say games are non-critical and easily offshored; others note game dev is highly specialized and not trivially replaceable.

Customer impact and business risk

• Some MTG Arena players fear unionization could raise costs, hinder layoffs, and endanger the product’s long-term viability.
• Others reply that unions don’t ban layoffs, just constrain arbitrary ones, and that stable, motivated staff tend to produce better, more consistent products.
• A few view the move as evidence WotC/Hasbro mismanagement has pushed workers to the brink.

IP, side projects, and “free time”

• Many report contracts in tech and creative fields that broadly assign all IP produced during employment (sometimes even off-hours), or require approval for side projects.
• Some see this as standard risk-management; others call it predatory overreach and refuse to sign such clauses.
• The union’s demand that off-hours creative work remain the worker’s property resonates strongly with commenters.

AI and RTO demands

• Workers’ pushback on generative AI is framed as protecting creative integrity, jobs, and copyright clarity; opponents view resistance as Luddite and competitively dangerous.
• Mandatory return-to-office is widely criticized; supporters of the union see RTO as sufficient cause to organize, while opponents insist management should decide what’s productive.

Nostalgia and Historical Role of ZSNES

• Many recall ZSNES as their first or main SNES emulator in the late 1990s–2000s, often on very weak hardware (486/Pentium-era PCs).
• It enabled access to fan translations and Japan-only RPGs, and popularized features like savestates, fast-forward, slowdown, and layer toggling.
• Users describe working around missing features (e.g., broken transparency) and hardware limits to finish games like Chrono Trigger and Final Fantasy titles.

Super ZSNES: Goals and Feature Set

• Seen as a “return” of ZSNES, but acknowledged as a new, separate emulator sharing only the lineage.
• Key selling points: GPU-powered enhancements, widescreen support “where available,” per-game visual and audio upgrades, high-res Mode 7, shaders, texture/audio replacement.
• Some find the new UI jarring compared to the iconic “snowy” legacy UI; others call the old one “timeless.”

GPU, Unity, and Architecture Choices

• Debate over whether GPU use is necessary for such an old system.
  • Pro: required for heavy visual enhancements, shaders, high-res tricks; Unity offloads low-level GPU work and eases multiplatform support.
  • Con: adds hardware/driver complexity; project may exist largely “because it’s fun.”
• Several are surprised or wary that it’s a Unity-based, closed-source binary, with minor concern about malware; others point to the developers’ long-standing reputation.

Accuracy, Performance, and Technical Critique

• Some worry GPU tricks and tile/line-based rendering may compromise PPU accuracy compared to cycle-accurate CPU emulators.
• Others note that high-accuracy SNES emulation is “solved” elsewhere, so new projects can build from that baseline.
• One commenter analyzes decompiled code and calls the implementation alpha-quality, with improved sync but limited optimization and modest GPU usage (mostly blending/bump-mapping).

Audio Enhancements and Philosophy

• Enthusiasm for uncompressed audio/sample replacements and references to existing fan projects restoring original instrument samples.
• Counterpoint: many SNES soundtracks were composed for the hardware’s limitations; “restored” or uncompressed versions can lose character or sound wrong.
• Some suggest a middle ground: remasters that intentionally re-introduce SNES-style compression/effects.

Open Source, Business, and AI-Free Positioning

• Old ZSNES remains GPL and forked; Super ZSNES is closed-source, with Android monetization cited as a reason.
• Community generally accepts closed source here because other open SNES emulators are mature.
• “No vibe coding / classic development style” sparks debate: interpreted as avoiding AI-driven codegen rather than all tooling; some see “handcrafted code” as a selling point for hobby projects.

ROM Legality and Usage

• Discussion of “legal” play: dumping one’s own cartridges with hardware readers vs. simply downloading ROMs.
• Many argue practical/legal risk for decades-old games is negligible, while others mention homebrew ROMs as a clean option.

Outage symptoms and user experience

• Issues, PRs, releases, stars filtering, projects, and Actions views intermittently show zero results or “no items,” despite content existing.
• Pages sometimes load correctly after repeated refreshes, implying only some servers are healthy.
• Failures are often “silent”: lists appear empty rather than showing clear errors, which users view as dangerous and misleading.
• Some users report partial data (not all PRs) even when lists load.
• Official status characterizes this as “intermittent,” but several commenters say failures are ~90% of page loads.

Perceived reliability trends

• Many feel GitHub outages have become weekly or “evergreen,” with the last 6–12 months called out as notably worse.
• Shared third-party uptime charts and historical links are used to argue reliability has degraded; others note pre-acquisition outages existed but may have been under-tracked.
• There is criticism that postmortems are rare or shallow given outage frequency.

Speculated causes

• Migration from GitHub’s own datacenters to Azure is frequently blamed; some suggest Azure itself is unstable or overloaded.
• Several tie issues to increased AI usage: Copilot features, AI-generated code volume, and coding agents hammering the platform.
• Some commenters generalize that modern practices (shipping faster, less testing, AI-assisted coding) lead to more production bugs.

Impact on teams and business

• Outages delay deployments and releases; some teams explicitly postpone production changes.
• Enterprise migrations from self-hosted GitLab or internal Git servers to GitHub are now viewed as risky, especially when profit is tied to tight release windows.

Alternatives and self‑hosting

• Many report positive experiences with self-hosted Gitea, Forgejo, GitLab, gitolite+cgit, Gerrit, and SourceHut, citing:
  • Better perceived uptime and performance.
  • Lower or more predictable cost.
• CI is highlighted as the hardest part to replace; GitHub Actions is still seen as strong, but Gitea/Forgejo Actions and Woodpecker are mentioned as viable.
• Some mirror code to multiple forges to reduce GitHub as a single point of failure; others note that migrating issues, PRs, permissions, and links is the real challenge.

Centralization and ecosystem concerns

• Several argue GitHub’s central role makes its outages disproportionately harmful to the global dev ecosystem.
• There is tension between the convenience/social network of a single popular forge and the resilience benefits of diversification and decentralization.

Overall Tone and Pessimism

• Many comments are skeptical, reflecting a broader sense that Canada is underperforming and “unrecognizable.”
• Some push back that this is just expectation-setting based on track record, while others criticize what they see as default HN/internet negativity, especially toward non‑US stories.

What Kind of “Sovereign Wealth Fund” Is This?

• Several note this is unlike classic sovereign wealth funds, which are usually funded by resource surpluses; here it’s seeded with debt.
• Critics call it effectively a “debt fund” or an infrastructure bank rebranded as a wealth fund, potentially a slush fund for mega‑projects.
• Supporters argue it’s at least better than one‑off spending and could create long‑term assets if well run.

Debt, Macroeconomics, and Domestic Focus

• One line of argument says a heavily indebted country shouldn’t run a wealth fund at all; others counter that Canada’s public debt is far below “300% of GDP” and that GDP isn’t the only measure of wealth.
• Some say Canada has vast natural resources as implicit collateral and could have Norway‑style surpluses if it chose different royalty/tax regimes.
• Debate over whether concentrating investments domestically overexposes Canadians to local economic risk versus providing extra tax and growth benefits.

Comparisons to Norway, Saudi Arabia, and Others

• Norway’s model is praised: resource profits saved abroad to shield the domestic economy.
• Saudi’s fund is described as a policy tool for large infrastructure; Norway’s is portrayed as a pure investor.
• Other pension approaches (New Zealand, Australia, UK, US Social Security) are referenced to contrast funding discipline and demographic risks.

CPP / Pension-Fund Performance Debate

• Some warn that if the new fund is run like the CPP Investment Board, it may underperform benchmarks while paying high fees and bonuses.
• Others counter that CPP has been relatively successful internationally and has helped Canada avoid large unfunded liabilities.
• There is a long subthread on whether lower volatility justifies underperformance, the role of active management vs. indexing, and how funds should manage depression‑level drawdowns.

Political, Governance, and Equity Issues

• Concerns raised about potential conflicts of interest and financial-sector influence, though defenders note mechanisms like blind trusts.
• Suggestions that wealth taxes and nationalized resource revenues should feed a true sovereign fund, possibly used to cut income taxes.
• Pushback that national control can deepen regional grievances (resource provinces vs. major cities) and intersects with Indigenous reconciliation claims.

Project Selection and Execution Risks

• Worries that it could degenerate into politically motivated job programs or stalled infrastructure (NIMBY resistance, poor pipeline of “investable” projects).
• Some see it as a way to bypass normal political constraints to fund strategic projects; returns are expected to be middling at best.

Pricing changes and mechanics

• Copilot is moving from per-request subscription to usage-based, token-priced “AI Credits” that roughly mirror underlying API costs.
• Plan fees (e.g., $10 Pro, $39 Pro+) stay nominally the same, but now just prepay that dollar amount in credits each month.
• Annual Pro/Pro+ subscribers keep the old “premium request” model until renewal, but model multipliers jump sharply (e.g., Sonnet 4.6 from 1×→9×, GPT‑5.4 from 1×→6×, Opus from 3×→27×), effectively slashing allowed high-end usage.
• Autocomplete and basic “next edit suggestions” remain unlimited within plans; agentic chat, code review, and containers consume credits (and GitHub Actions minutes for reviews).
• New models and features will not be added to legacy annual plans; users can get prorated refunds or convert to monthly.

User sentiment and behavior

• Many individual users say the deal went from “incredible value” to “not worth it overnight” and plan to cancel or switch.
• Heavy agentic users report they were effectively getting hundreds of dollars’ worth of Opus/Sonnet tokens monthly for $10–$40; they see this as a massive (often 10–100×) effective price hike.
• Some will keep Copilot solely for autocomplete and VS Code integration, wishing for a cheaper autocomplete-only tier.

Economics of inference and “enshittification”

• Broad agreement that flat per-request pricing was unsustainable once “one request” could trigger hours-long, multi-agent runs.
• One camp frames this as classic “enshittification”/bait‑and‑switch after a subsidized land‑grab; another says it’s just the end of loss-leading and alignment with real compute costs.
• Concern that token anxiety will make users self-censor use of stronger models, degrading experience and adoption.

Alternatives and competition

• Many mention moving to:
  • Direct APIs (Anthropic, OpenAI, DeepSeek, Kimi, etc.).
  • Routers like OpenRouter (with ~5–5.5% markup but unified API, failover, easier billing).
  • Other IDE/agent tools (Claude Code, Codex CLI, Cursor, Windsurf, Cline, OpenCode).
• Enterprises may stick with Copilot due to Microsoft ecosystem lock‑in and data-governance approvals, even at higher effective cost.

Legal and contract concerns

• Some argue mid-term changes for annual plans (especially 6–9× multipliers) may violate consumer-protection laws in regions like the EU/Australia; others note that prorated refunds are offered and ToS often allow such changes.

Local models and longer‑term outlook

• Growing interest in local/open-weight models (Qwen, Gemma, DeepSeek, etc.) via tools like Ollama, llama.cpp, Unsloth, especially to avoid variable bills and privacy risks.
• Debate over whether cloud inference costs will keep rising or be undercut by increasingly capable, cheap-to-run open models on consumer hardware.

Relative Safety of Glyphosate / Roundup

• Many argue glyphosate is among the least harmful widely used herbicides, especially compared to older chemistries (organophosphates, DDT, etc.).
• Others counter that “less bad than alternatives” is not a strong endorsement, likening it to comparing small vs large calibers: both can still cause harm.
• Several note that Roundup (the formulated product) is not the same as pure glyphosate; surfactants and other additives may add risk.

Health Risks, Cancer, and Gut Microbiome

• Some commenters say evidence for glyphosate causing cancer is weak, especially for normal dietary exposure; risks may be higher for workers with long-term, heavy contact and for pre-harvest desiccation uses.
• IARC’s “probable carcinogen” (2A) classification is contrasted with EPA’s historical “not carcinogenic” stance; classification context (similar to red meat, hot beverages, night shifts) is emphasized.
• A long subthread debates gut microbiome disruption: mechanisms like inhibition of bacterial pathways and gut inflammation are cited, but others insist current data is mostly mechanistic or animal-based and not clearly tied to human disease.
• Additional claims involve glyphosate’s metal-chelating properties (e.g., with aluminum) and possible neurological links; some point out that at least one highly cited paper here is considered very weak.

Agricultural Dependence and Alternatives

• One side argues modern population levels rely on fossil-fuel-based fertilizers, pesticides, and herbicides; removing them abruptly would cause mass starvation, citing Sri Lanka’s failed rapid “organic” shift.
• Opponents say weeds can be controlled mechanically (more labor and fuel, higher prices, more CO₂) and that organic methods and future robotic/mechanical weeding could reduce chemical use.
• Debate continues over how much increased costs would translate into real-world hunger.

Resistance and Other Herbicides

• Glyphosate-resistant weeds (e.g., Kochia) are already widespread; some see resistance as inevitable but manageable by rotating chemistries.
• Newer or alternative herbicides (e.g., glufosinate/Liberty) are mentioned; environmental breakdown rates and resistance patterns are concerns.

GMOs, Patents, and Farmer Economics

• Strong thread on patented “Roundup Ready” crops:
  • Critics object to seed patents, annual purchasing, and market power concentration; see it as a “subscription model” for seeds.
  • Others respond that hybrid seeds and IP protections long predate GMOs and that farmers choose them because yields and economics are better.
  • Worries are raised about contamination and liability; countered by claims that case law punishing accidental contamination is largely myth.

Law, Federal Preemption, and the Supreme Court Case

• Multiple commenters stress the case is mainly about federal–state preemption and failure-to-warn torts, not a direct safety ruling on glyphosate.
• EPA’s position, state labeling regimes (e.g., California), and Prop 65 history are discussed as context.
• Some see state-level labeling as potential trade barriers; others defend states’ right to stricter warnings and the precautionary principle.

Public Trust, Precaution, and Regulatory Capture

• Comparisons are drawn to tobacco, leaded gasoline, and other chemicals once declared “safe,” fueling skepticism of EPA/FDA due to perceived regulatory capture.
• Others insist that long, global use without clear, large-scale harms is itself strong evidence of relative safety.
• Broader worries about microplastics and cumulative pollution appear, with disagreement over how much evidence is “enough” to justify strong regulatory action.

AFP & Time Capsule Deprecation

• macOS 27 is expected to drop AFP client support and SMB1, breaking network Time Machine backups to older Time Capsules and NASes that only speak AFP/old SMB.
• Some see this as overdue: AFP’s been deprecated for years, Time Capsule hardware is old, disks and PSUs are failing, and SMB2/3 are long-established.
• Others argue Apple could afford to maintain old features and that this erodes trust and “platform commitment,” especially for users heavily invested in older gear.
• A few point to community workarounds (e.g., Netatalk, Samba on Time Capsule) as paths forward, but note this requires technical effort.

Time Machine: Reliability, UX, and Future Direction

• Many describe Time Machine as unreliable over the long term: silent failures, corrupted catalogs, forced “start over” events, and issues over SMB/AFP.
• Some use Time Machine only as a secondary backup, preferring tools like Carbon Copy Cloner, rsync, restic, or Acronis + NAS snapshots.
• UI/animation is criticized as buggy and distracting; full-screen “space” metaphor and poor multi-display behavior make restores harder to use.
• There’s speculation that Apple might eventually push iCloud-based Mac backups, tying into services revenue, though others think enterprise reliance on local TM slows that.

SMB, NFS, and macOS Networking Quality

• Strong complaints that Apple’s SMB implementation is slow, buggy, and fragile (sleep, reconnect, small-file performance, extended attributes).
• Some report excellent throughput on modern SMB3 and 10GbE, suggesting performance is highly workload- and setup-dependent.
• AFP is often reported as faster and more predictable than SMB; NFS can be faster still but is buggy in Finder and occasionally causes panics (e.g., krb5 auth).
• Discussion notes Apple’s move away from GPL-licensed Samba to its own SMB stack, widely viewed as a regression.

Security & TLS 1.2 Baseline

• The move to require TLS 1.2+ for certain connections is broadly seen as overdue and positive.
• Concerns about “e-wasting” older LAN-only devices are raised, but others note the documented scope of the change does not appear to include typical local printer/scanner web UIs.

Broader Reflections on Apple Software Quality

• Multiple commenters feel macOS quality and polish have declined over the last decade, especially compared to older releases.
• Some attribute this to services-first business priorities and reduced attention to “power user” and networking workflows.

Nature of SVG and Risks

• Commenters stress that SVG is a markup language, not a simple image format; it’s as dangerous as HTML when untrusted.
• Main hazards: embedded <script>, event-handler attributes, <foreignObject> with HTML, external URLs in attributes/CSS, <image>/feImage, DTD and namespace tricks.
• Browsers already treat the same SVG differently depending on context: <img> disables scripts/external loads, while inline and <object> can execute them.
• Because the same file can be used as both “image” and “document,” simply “allow SVG uploads” is risky.

Subset and New-Format Proposals

• Many suggest supporting only a restricted SVG subset that covers common “paths + fills” use cases.
• Existing or proposed subsets/formats are mentioned: SVG Tiny, SVG Tiny PS, SVG Native, BIMI profiles, Android’s vector format, TinyVG, and ideas like “SVG-ES” or “SSVG.”
• Trade‑off: security and simpler tooling vs losing features such as SMIL animation, filters, and drop shadows. Some see that loss as acceptable for wide adoption; others do not.

Real-World Cases and Tooling

• Scratch’s history illustrates the pitfalls: initial XSS via <script>, then repeated sanitizer bypasses (including regex-based filtering).
• Scratch inlines SVGs so it can call getBBox() because many input files have bad viewboxes.
• There are references to SVG sanitization libraries and to the browser HTML Sanitizer API, which allows a limited SVG subset but bans style by default.

Security Mitigations Discussed

• Strong preference for CSP over ever-more-complex sanitizers: script-src 'none', sandboxing, and applying CSP directly via <meta http-equiv> in documents or iframes.
• Iframe sandboxing plus strict CSP is seen as robust but underused, partly due to poor docs and integration friction with modern frameworks.
• Some propose blocking any http(s) URLs in attributes or constraining SVG by headers like Content-Security-Policy and Sec-Fetch-*.

Embedding Modes and API Ideas

• Frequent wish for a simple, declarative “safe mode”:
  • Attributes like sandbox or exec="false" on <svg>.
  • Type flags (<img type="ssvg">) or special SVG versions/namespaces that guarantee no scripts or external fetches.
• Others argue that <img src="file.svg"> already provides a mostly-correct safe image behavior, but lacks dynamic styling (e.g., CSS variables) and DOM interactivity.

Broader Reflections

• Several see SVG’s scripting and HTML-like complexity as “overreach” when most users just want crisp logos.
• Others note SVG’s origins as a richer, potentially Flash-like interactive medium, which explains (but doesn’t excuse) today’s security headaches.

Scope of the Case & Data Types

• Thread focuses on geofence warrants for app/OS-based location histories (e.g., Google) rather than classic cell-tower records.
• Key question framed as: is location history more like a bank record (weakly protected third‑party data) or a “digital diary” (strongly protected personal papers/effects)?
• Some note this case involves Google searching hundreds of millions of accounts to identify a handful of devices near a crime scene.

Fourth Amendment, Privacy, and Third‑Party Doctrine

• One side argues the 4th Amendment protects property-like interests (“persons, houses, papers, and effects”), and data held by third parties belongs to those third parties, so subpoenas/warrants on them are easier to justify.
• Others push back:
  • Location data should be treated as the user’s “papers/effects,” even if held by a provider.
  • The “right to be secure” must adapt to modern mass surveillance, not just physical house searches.
• Debate over whether “feelings of security” matter or only objective “searches and seizures.”

Geofencing vs Other Surveillance Tools

• Comparisons drawn to:
  • Bank cameras/license plates (seen as narrower, more contextual).
  • Flock/ALPR networks and DNA databases (also dragnet‑like).
• Critics emphasize scope: a geofence in a dense area can implicate millions and intrude into private spaces (e.g., inside a church).
• Supporters say police already have very limited tools, and geofencing can be crucial for solving serious crimes when other leads are exhausted.

Due Process, Warrants, and Abuse Concerns

• Some argue geofencing is acceptable if backed by a judge and probable cause; better than warrantless data purchases from brokers.
• Others counter that:
  • Judges often rubber‑stamp broad warrants.
  • People swept up never learn they were searched, so can’t challenge it.
  • Parallel construction and illegally executed warrants undermine any formal safeguards.

Tech Company Behavior & Google’s Changes

• Noted that Google stopped storing centralized location timelines and now keeps data on-device, partly in response to legal pressure (e.g., Carpenter) and abortion‑related prosecutions.
• Mixed reactions:
  • Privacy advocates welcome it; some see data as “toxic waste” given government access risks.
  • Others miss lost features (Timeline history, web access) and personally don’t fear courts.
• Skepticism remains about telcos and data brokers continuing to sell or share location data.

Courts, Originalism, and Democratic Legitimacy

• Several comments doubt this Supreme Court will meaningfully limit surveillance, citing originalist tendencies and result‑driven reasoning.
• Others argue change should come via legislatures and constitutional amendments, not by re‑imagining what the framers “would have written” about modern tech.