Hacker News, Distilled

AI powered summaries for selected HN discussions.

Page 786 of 835

Aeon: OpenSUSE for Lazy Developers

Positioning of Aeon / “Lazy Developers”

  • Aeon is seen as an immutable desktop aimed at users who want minimal admin work: automated updates, rollbacks, containerized apps, and a locked-down base system.
  • Some are confused by the “for developers” label; it looks similar to a “lazy desktop” for non‑developers.
  • Critics say you still need to learn new workflows and packaging models (Flatpak, containers, possibly custom RPMs), and customization options are limited versus what many developers want.

Immutable Model, Apps, and Tooling

  • Base system is updated transactionally; rollbacks and btrfs snapshots are central.
  • Apps are expected to come via Flatpak or containers (distrobox). Some devs prefer using Guix or Nix on top of Aeon/MicroOS instead of Flatpak/distrobox.
  • Concerns: key tools (e.g., Firefox Flatpak on AArch64) and some niche software aren’t always available as vetted Flatpaks, pushing users to “random” sources.

Security: Firewall, Secure Boot, Encryption

  • Controversial: Aeon initially shipped without a firewall; maintainers argue an immutable, non‑listening base doesn’t need one and extra services increase attack surface.
  • Opponents insist a default “deny incoming” firewall is important because user apps may open ports.
  • Some users simply install firewalld via transactional-update despite it being “against the grain.”
  • Tumbleweed and Leap are praised for good Secure Boot + Nvidia support, though others recount painful MOK and driver workflows.
  • New Tik-based Aeon installer reportedly dropped full‑disk encryption support, blocking adoption for some.

openSUSE Ecosystem & Reputation

  • Many commenters run Tumbleweed or Leap as daily drivers (desktops, laptops, HPC, corporate, scientific workloads, NAS-like appliances).
  • Tumbleweed is repeatedly described as “Arch‑like but more stable,” with strong automated QA (OBS + openQA), easy rollbacks (snapper + btrfs), and relatively fresh packages.
  • Some call openSUSE “niche,” especially outside Europe; others dispute this and argue most major distros are now similarly stable.

Rolling vs Immutable & Btrfs

  • Debate over rollbacks: supporters value fast recovery from breakage (especially with 3rd‑party repos); skeptics note rollbacks leave you on old software without fixing root causes.
  • Immutable systems requiring reboots for OS updates are seen as a minor tradeoff by some, annoying with full-disk encryption by others.
  • Btrfs is standard; concern is raised about poor VM/database performance without nocow, and advice is to put VMs/DBs on a separate ext4 partition.

The Death of NYC Congestion Pricing

Overall Reaction to Cancellation

  • Many see the last-minute halt as a major political blunder that angers both supporters and opponents of congestion pricing.
  • Some argue it shows Democrats can’t deliver hard but necessary policies and may even be legally dubious as an executive power grab.
  • Others note congestion pricing was polling poorly, and leaders are reacting to voter anxiety over inflation and rising costs.

Merits of Congestion Pricing

  • Proponents:
    • View it as a way to internalize driving externalities (pollution, noise, deaths, lost time) and free up street space.
    • Expect reduced gridlock, faster buses, safer streets, and better transit funded by toll revenue.
    • Point to other cities (London, Stockholm, etc.) where congestion pricing reduced traffic and later became politically acceptable.
  • Skeptics:
    • Call it a blunt, regressive tax that hits delivery, tradespeople, drivers from outer boroughs/NJ, and ultimately all residents via higher prices.
    • Argue it’s more about revenue than congestion and that rich drivers will just pay.
    • Say planners never clearly defined congestion metrics or built in sufficient exemptions (e.g., cargo, construction, tunnel through‑traffic).

MTA Funding and Governance

  • Broad agreement that MTA is essential but financially messy: overtime abuse, high capital costs, complex legacy systems, 24/7 operations.
  • Some want deep restructuring or even bankruptcy-style cleanup before new revenue; others warn that making funding contingent on “fixing everything” just guarantees chronic underfunding.
  • Debate over whether congestion revenue (earmarked for capital) is a hidden bailout for unsustainable operations; some data cited showing state/local support has already been cut.

Equity, Class, and Modes

  • Disagreement over who really pays:
    • One side: congestion pricing largely charges affluent car users to benefit poorer transit riders.
    • Other side: it taxes working drivers (plumbers, taxi/Uber, delivery) to subsidize affluent Manhattanites.
  • Several note transit vs. driving tradeoffs are highly contextual (borough vs. suburb vs. NJ, family size, off‑peak trips).

Urban Design, Transit, and Density

  • Repeated theme: dense cities need strong transit; roads alone cannot scale.
  • Discussion of chicken‑and‑egg between building transit first vs. density first, with examples from NYC, Europe, and East Asia.
  • Some broader frustration with car-centric planning, bureaucracy, and the high “invisible taxes” imposed by permitting delays and regulatory risk.

200 people charged in $2.7B health care fraud crackdown

Scale and character of the fraud

  • Roughly $2.7B across 200 people ≈ $13.5M each on average; one Arizona wound‑graft scheme allegedly billed ~$900M for <500 patients ($2M/patient).
  • Some suspect many billed grafts were never actually applied, given how implausible per‑patient totals look.
  • Another highlighted case: misbranded HIV meds bought on the black market, relabeled, and resold; at least one patient was hospitalized after receiving the wrong drug.

Medical billing practices vs. outright fraud

  • Several comments explain “chargemaster” pricing: providers bill huge amounts, insurers/Medicare reimburse only up to “allowed” amounts, and write off the rest.
  • Others argue this is functionally fraudulent, since bills sent to patients bear fabricated list prices, and patients who are uninsured or uninformed can be gouged.
  • Debate on ambulance and small‑procedure bills shows confusion and anger about adjustments, negotiated rates, and why nominal prices are so detached from costs.

How much of healthcare is fraud?

  • One side: $2.7B over ~5 years is ~0.01% of total US health spending; main problem is system design, not fraud.
  • Other side cites FBI/Medicare estimates of 5–10% of spending lost to fraud (tens of billions annually), similar to rates in other countries.
  • A former statistician in the space says enforcement focuses only on the largest, easiest cases; environment is “target rich.”

Legal consequences and enforcement

  • Many want prison time, loss of licenses, and full restitution; fear that fines of a few percent and no admission of wrongdoing are common.
  • The airport arrest and possession of books on “disappearing” and criminal law spark discussion about flight risk, evidence of intent, and limits of spousal privilege.
  • Some worry fraud/anti‑kickback laws are complex and ambiguous, making normal business practices feel criminal; others reply that many prosecuted schemes are straightforward identity theft, kickbacks, or fake billing.

System design, universal healthcare, and incentives

  • Several argue US pricing and reimbursement structures (especially insurance intermediation) make legal billing indistinguishable from fraud.
  • Discussion of whether universal or single‑payer care would reduce fraud:
    • Arguments for: lower base prices and simpler rules reduce the upside.
    • Arguments against: this scheme already targeted Medicare; expanding similar coverage might expand the attack surface.
  • A recurring theme: private insurers often tolerate fraud and pass costs via higher premiums, while government programs have more incentive and authority to investigate.

Ethical outrage, anecdotes, and culture

  • Numerous personal stories: spurious newborn charges, traumatic ER visits with surprise bills, and years of paying medical debt; many describe US healthcare as feeling like a scam.
  • Some claim end‑of‑life care is frequently a wealth‑stripping mechanism rather than patient‑centered care.
  • The alleged scammers’ lavish wedding registry (luxury towels, throws, pet leashes, etc.) is widely mocked as emblematic of fast, illegitimate money and distorted values.

Ideas for improvement

  • Suggestions include:
    • Publishing anonymized claims data and paying bounties (e.g., 20%) to third parties who detect provable fraud, potentially using AI/ML.
    • Strengthening prosecutions and penalties to change cost‑benefit calculations for would‑be fraudsters.

Python grapples with Apple App Store rejections

Immediate issue: Python apps rejected on iOS/macOS stores

  • Apps embedding Python are being rejected because binaries contain the string itms-services, which App Review associates with installing other apps and violating guideline 2.5.2.
  • The affected apps do not actually use that URL scheme at runtime; the rejection is based on static string scanning.
  • Current Python-side workaround is a compiler/config flag to exclude the problematic code on iOS, rather than obfuscation, which is seen as risky with Apple.

Apple’s review and security practices

  • Several comments describe Apple’s checks as crude substring scans (“cockroach security”) layered on top of other mechanisms.
  • Others argue Apple likely has more sophisticated checks too, but evidence in this case points to a simple string match.
  • App Review feedback is criticized as vague; developers want precise locations of offending strings and clearer remediation paths.
  • Some question why the sandbox doesn’t just block itms-services for App Store–signed apps, instead of relying on static scanning.
  • Enterprise distribution and out-of-store macOS apps do legitimately use this scheme, complicating a blanket ban.

Why itms-services is in Python at all

  • The string lives in urllib.parse’s list of schemes that have a “netloc”, to correctly parse Apple’s nonstandard URL format.
  • Discussion notes that Python’s URL handling is partly working around other software’s quirks; some argue proprietary schemes should be handled heuristically, not hard-coded.
  • There’s debate over whether Python’s URL implementation is spec-correct regarding empty vs null hosts.
  • Some suggest the immediate fix could be as trivial as changing a test case string or restructuring tests so they don’t ship.

OSS responsibility vs platform bugs

  • One view: OSS can accept small, localized workarounds for big-vendor bugs, but only with a time limit and after filing upstream reports, to avoid being the permanent band-aid.
  • Others are skeptical this works in practice, given perceptions that Apple’s bug reporting is a “black hole.”

Wider ecosystem: signing, AV, and small developers

  • Parallels are drawn to Windows Defender and SmartScreen falsely flagging small, unsigned binaries (including Python/PyInstaller) as malware.
  • Code-signing costs and platform hurdles push some small or OSS developers away from macOS/Windows or from updating apps, and toward Linux or niche distribution methods.

Platform control and user freedom

  • Strong debate over whether Apple’s tightly controlled model is acceptable “custodianship” or “digital serfdom.”
  • Some emphasize that users can choose Android, including de-Googled variants with sideloading and FOSS-only setups.
  • Others argue a duopoly still leaves too little real freedom, and criticize Apple’s browser engine lock-in and limits on running arbitrary code on owned hardware.

Maker of RStudio launches new R and Python IDE

Python IDEs on iPad

  • Several tools are suggested: Pythonista (nice UX but older Python, strong iOS API access), Pyto (more up-to-date, includes clang/LLVM, less approachable), Carnets (offline Jupyter with scientific stack), Juno (notebooks-focused).
  • Hosted VS Code via code-server, especially integrated into the Blink terminal app, is praised as the most powerful option due to seamless remote development.

AI Coding Assistants (Copilot and Alternatives)

  • Some users find Copilot mostly an interruption: frequent trivial or subtly wrong suggestions, requiring vigilance.
  • Others see it as “autocomplete on steroids,” helpful for repetitive boilerplate, dictionary assignments, unit tests, and legacy code comprehension.
  • A common effective pattern: write a detailed comment, trigger Copilot once, accept and then delete the comment (“pseudocode-to-code compiler”).
  • Performance concerns: automatic suggestions can use high CPU and drain batteries; manual trigger via shortcut is preferred by some.
  • Use of Copilot is sometimes employer-driven; attitudes range from skepticism (“solution in search of a problem”) to viewing it as essential when onboarding to large legacy codebases.
  • Local/open alternatives like Tabby are briefly mentioned.

R vs Python and Posit’s Strategy

  • Strong debate over whether R is stagnating versus “already good enough” for its domain.
  • Many see Python outpacing R in overall usage; others report that in analytics-heavy roles (including at large companies), R remains common for exploration, with Python used for production/ML.
  • In life sciences and social sciences, R is still described as dominant for method packages, though some say Python is gaining.
  • Several commenters emphasize Posit’s philosophy of reducing barriers between R and Python rather than picking a winner; many team members and users are “bilingual.”
  • Some feel RStudio’s UI is dated; others praise it as fast, focused, and superior for data exploration and RMarkdown workflows.

Positron vs VS Code and Extension Ecosystem

  • Question raised: why not just a VS Code extension pack?
    • Answer: VS Code’s extension model can’t add certain top-level UI or deeply integrated services; Positron needs modifications to the core workbench.
  • Positron regularly merges upstream VS Code changes to maintain feature parity and extension compatibility, though this is acknowledged as a nontrivial maintenance burden.
  • Lack of Microsoft’s Remote-SSH and devcontainers is a major blocker for some, especially those working with large remote datasets or GPU servers. Open-source equivalents exist but may not yet work smoothly with Positron.

Licensing and Marketplace Concerns

  • Positron is under the Elastic License, explicitly not considered free/open-source by some commenters.
    • Restrictions include prohibitions on offering it as a hosted service and on circumventing license-key functionality.
  • Microsoft does not allow third-party IDEs access to the official VS Code Marketplace; this is viewed as a way to keep control of the ecosystem and makes non-MS distributions second-class citizens.
  • One linked analysis describes VS Code’s design as intentionally leading to fragmentation; some see Posit as becoming dependent on Microsoft’s ecosystem while giving up a differentiated RStudio experience.

RStudio’s Future and Backwards Compatibility

  • Official comment: RStudio development and maintenance will continue; some R-focused features will remain exclusive to it.
  • Some users report Positron does not yet handle .Rproj projects as smoothly as RStudio (e.g., package build/test GUI), though similar commands exist via the command palette. Better GUI tooling is planned.
  • Concerns from users heavily invested in existing RStudio-based R packages or mixed R/Python projects about whether Positron is a full replacement; current guidance implies it is not yet and may never be for all use cases.

Notebooks, RMarkdown, and Related Tools

  • Many prefer RMarkdown/Quarto’s literate programming model over Jupyter, describing it as more elegant, less “over-engineered,” and better integrated with RStudio.
  • Quarto is highlighted as the evolution of RMarkdown, supporting multiple languages (R, Python, others via Jupyter backends).
  • Some ask what VS Code notebooks lack compared to RMarkdown; others say they rely on Quarto or RMarkdown for teaching, homework, and scientific writing due to pandoc + LaTeX integration.
  • Tools like Observable Framework and Evidence.dev are mentioned as RMarkdown-like approaches in the JS and SQL worlds.

Other Tooling Notes

  • For CSV/TSV handling in VS Code, extensions like Excel Viewer and Data Wrangler are recommended, with caveats on large file performance.
  • Some see Spyder as a better traditional Python IDE; others view VS Code (and now Positron) as the new “Eclipse-style platform” for repackaged IDE distributions.

A buried ancient Egyptian port reveals connections between distant civilizations

Overall reaction to the article

  • Many commenters praise the piece as unusually strong for its click‑baity title: rich in detail, clear writing, and good images.
  • Some particularly appreciate its focus on cooperation and mingling among ancient religions and cultures, and note that such stories are underrepresented.

Ancient trade networks and economic scale

  • A cited book on the Roman–Indian Ocean trade claims customs on Red Sea commerce may have made up about a third of Roman state revenue in the 1st century CE.
  • One ship’s cargo after a 25% import tax is said to be worth enough to buy a luxury Italian estate or pay 40,000 stonecutters for a year; reports mention up to ~100 such ships per monsoon season.
  • Several note that these routes and the India–Egypt/Red Sea trade have long been known in scholarship; some criticize media narratives that present them as newly “discovered.”

Navigation, technology, and human capability

  • Commenters are impressed by how ancient societies coordinated production and sailing around the monsoon cycle and navigated the Indian Ocean with limited technology.
  • This is seen as evidence that humans in antiquity had the same capacities and sophistication as people today, minus modern medicine and tools.

Archaeology practice and local knowledge

  • Multiple stories highlight how local farmers or villagers often know sites long before professional archaeologists do, in India, South America, and Europe.
  • Earlier disturbances—from Romans to 20th‑century explorers—are themselves viewed as later historical layers.

Indic manuscripts and temple preservation in India

  • One thread stresses that many manuscripts lie in Indian temple basements, vulnerable to humidity and modern interference; calls for better preservation.
  • Others ask why they would decay now if they survived 2,000 years; replies say old sealed vaults and traditional methods have been disrupted.
  • Serious allegations are made against Indian state temple‑management bodies (especially in Tamil Nadu): claims of corruption, mismanagement of funds and assets, physical damage to heritage structures, and theft/smuggling of idols and manuscripts.
  • Skeptics demand stronger, mainstream sourcing; some provided sources cover idol theft and financial misuse, but there is disagreement over scale and interpretation.

Indian politics, secularism, and religious identity

  • A long sub‑discussion debates whether Indian temples should be state‑run or controlled by devotees, contrasting this with Muslim Waqf boards and relative autonomy for churches and mosques.
  • Some argue India’s current model of “secularism” discriminates against Hindus (e.g., state control of Hindu temples, separate Muslim personal law, lack of a Uniform Civil Code) and advocate for a Hindu nation‑state.
  • Others strongly oppose this, characterizing it as majoritarian or “extremist,” warning that official religious identity would marginalize non‑Hindus and undermine equal citizenship.
  • There is debate over whether secularism requires identical laws for all, or simply equal state treatment even with different personal laws.
  • Historical topics raised include British colonial legal systems, partition, Waqf property rules, and the status of Nepal as formerly Hindu. Opinions remain polarized.

Cross‑cultural religious and philosophical exchange

  • Commenters note that connections between India, Greece, Egypt, and the Near East are well attested: Greco‑Buddhist kingdoms, Buddhist missions westward, and Indian influence in East Africa and Southeast Asia.
  • Some argue that elements of Hellenistic philosophy, Hermeticism, Gnosticism, and even Christian ideas (e.g., salvation through personal enlightenment) may have interacted with or reacted to Buddhist thought, while still remaining distinct traditions.
  • Examples cited include Buddhist stories entering Christian hagiography and Indo‑Greek kingdoms.

Indic history, Dravidians, and ancient languages

  • A thread discusses the term “Dravidian”: whether it refers to language families, ethnicities, or geography, and notes that historic South Indian empires didn’t self‑identify that way.
  • There is mention of genetic models (ANI/ASI) and the date of mixture, but commenters stress complexity and lack of direct ancient DNA for ASI.
  • On the Indus/Harappan civilization, one commenter notes there is no consensus it was Dravidian; some scholars prefer an unknown or Austronesian‑related language, and better evidence is needed.

Linguistic and cultural traces

  • Discussion of the Hindi/Sanskrit word for eye cosmetics and its close Akkadian cognate is used as an example of deep ancient contact between Indic and Near Eastern cultures.
  • Another thread notes that “chai” in English has acquired a specific meaning (spiced tea) distinct from its generic “tea” meaning in many languages.

Terminology and representation

  • Several notice that the article often uses “Indic” instead of “Hindu” for deities and artifacts.
  • Some readers see this as erasing explicit Hindu identity for clearly Hindu figures or chants; others defend “Indic” as a neutral term that avoids modern religious labels and ideological loading when boundaries are blurry.

Modern travel and environment

  • One commenter highlights the southern Egyptian Red Sea coast (including near Berenike) as a beautiful, less‑touristed area with excellent diving and kitesurfing; another asks about current safety in light of regional politics.

Miscellaneous

  • Clarification that “sherds” (not “shards”) is a standard archaeological term for pottery fragments; this is linked to a recent terminology change in a popular video game.
  • Links are shared to lectures, podcasts, and books on ancient Indian trade and empire as further reading, and to images of the Berenike Buddha and Greco‑Buddhist art.

Astronauts take shelter in Starliner, other spacecraft after satellite breakup

Role of Docked Spacecraft as Shelters

  • Crews shelter in their return vehicles (Starliner, Dragon, Soyuz) during debris events.
  • The primary safety function is evacuation capability: maneuvering thrusters, heat shield, and landing system, plus trained crew and planned logistics.
  • Wall thickness or small size matters less than the ability to quickly undock and deorbit if the ISS is compromised.

Starliner Reliability and Optics

  • Thread notes Starliner’s recent issues: multiple thruster failures and helium leaks.
  • NASA has delayed its nominal return to review propulsion data, yet continues to state it is “authorized” for emergency return.
  • Some argue this event shows NASA trusts Starliner enough as an escape pod; others see the article’s emphasis on Starliner as PR spin.
  • Debate over how much the thruster problems actually affect its emergency-return role, given redundancy (dozens of thrusters, only some failed).

Kessler Syndrome and Debris Lifetimes

  • Some see the event as a symptom of a trend toward Kessler Syndrome; others push back that such sheltering is still rare.
  • Several comments emphasize that at ~350–500 km altitude, atmospheric drag removes debris in years or decades, not millennia.
  • Smaller fragments deorbit faster; very-low Earth orbit constellations like Starlink have relatively short lifetimes.
  • A side debate criticizes simplistic “exponential growth” analogies as ignoring real-world carrying capacities and orbital decay.

Details of the Resurs-P1 Breakup

  • The object was identified as the ~5.6-ton defunct Russian satellite RESURS‑P1.
  • It fragmented near the time it passed over Russia’s Plesetsk site, but at least one space historian is skeptical it was an anti-satellite (ASAT) test, partly because using such a massive target would be “crazy and very bad.”
  • Historical stats: hundreds of on‑orbit breakup events have occurred over decades.

Frequency and Handling of ISS Debris Events

  • One view: debris conjunctions and occasional sheltering/attitude maneuvers are now a routine part of operations.
  • Another: actual “shelter in spacecraft” actions remain relatively rare.
  • ISS can sometimes be rotated to reduce cross-sectional area facing a debris cloud.

ASAT Tests and Responsibility

  • Discussion notes that multiple countries (US, Russia, China, India) have done ASAT tests, but their danger depends strongly on target orbit and resulting debris trajectories and lifetimes.
  • Several comments single out Russian tests as particularly hazardous and less constrained by concern for other operators.

Shielding Concepts and Future Mitigation

  • Magnetic shields are deemed impractical: would require multi‑tesla fields, are unsafe for humans, and only affect conductive/magnetic debris.
  • Various physical shielding ideas appear: Whipple shields (already in use on ISS), aerogel layers, sacrificial plates, and “defensive satellite” swarms.
  • Critics note that intercepting or fragmenting debris risks creating more, smaller junk that is harder to track.
  • Detailed discussion explains why hypervelocity impacts behave more like localized explosions than macroscopic pushes; mass and speed dominate damage, and even tiny fragments carry enormous kinetic energy.
  • Some speculate that cheaper access to orbit will eventually allow more robust, heavily shielded, and maneuverable stations, plus dedicated debris-removal satellites, though that may also mean larger “robust” junk unless cleanup becomes part of the ecosystem.

Supreme Court blocks controversial Purdue Pharma opioid settlement

Supreme Court ruling & bankruptcy law

  • Case turns on whether Chapter 11 allows “non‑debtor third‑party releases” that permanently shield people (the Sacklers) who are not themselves in bankruptcy.
  • Majority: bankruptcy courts lack authority to wipe out others’ future claims against non‑debtors; catch‑all “any other appropriate provision” can’t mean “anything goes.”
  • Dissent: code is flexible and equitable; if a global deal clearly makes creditors better off, courts should be able to approve it.
  • Ruling also affects future mass‑tort bankruptcies (e.g., other product or aviation cases).

Impact on victims and settlement

  • Blocked deal would have delivered ~$6B (mostly from the Sacklers) to governments and individual victims, and converted Purdue into a public‑benefit entity.
  • Many victims’ groups, states, and creditors backed the deal as the best realistic option: faster, more certain payouts, less litigation trauma.
  • Others saw it as a “Faustian bargain”: too little money versus harm done, and morally intolerable to sell immunity to billionaires.
  • After ruling, compensation is delayed and uncertain; some expect less money overall, others think new litigation could yield more but over many years.

Sacklers’ conduct and liability

  • Discussion emphasizes internal evidence that Purdue:
    • Knew about addiction risks and designed dosing that fostered dependence.
    • Aggressively marketed OxyContin as safer/less addictive, especially for chronic pain.
    • Greatly increased payouts to the family once liability loomed (“milking” before bankruptcy).
  • Debate over tools to reach family wealth: fraudulent transfer, piercing the corporate veil, criminal charges, asset seizure, garnishment.
  • Many argue fines are insufficient; call for prison and full or near‑total asset stripping.

Doctors, FDA, and regulators

  • Split views on physician responsibility:
    • Some stress doctors knew opioids were addictive, benefited from pharma incentives, and ran “pill mills.”
    • Others note misleading marketing, weak evidence‑appraisal training, time pressure, and system‑wide “treat pain aggressively” policies.
  • Questions about FDA’s role, regulatory capture, and self‑submitted industry studies.

Court politics and institutional trust

  • Unusual ideological lineup (conservatives plus Jackson in majority; Roberts/Kagan/Sotomayor/Kavanaugh dissenting) used both to:
    • Argue the Court is more complex than a simple partisan body.
    • And to argue recent patterns still show strong ideological blocs on many issues.
  • Broader debate over “textualism vs pragmatism” and whether this ruling prioritizes retribution/deterrence over restitution.

Opioids, addiction, and drug policy

  • Extensive side discussion on:
    • Scale of opioid deaths and broader social damage.
    • Comparisons to tobacco and other drugs.
    • Stigma, chronic pain treatment, and how supply‑side crackdowns can push users to more dangerous street opioids (fentanyl).
  • Some argue systemic causes (inequality, consumer culture, inadequate pain and addiction care) go beyond Purdue, even if Purdue amplified the crisis.

Misconceptions about loops in C

Scope of the Paper / Control-Flow Complexity

  • Paper is seen as a practical catalog of loop/control-flow pitfalls in C, especially for static analysis and compilers, not as conceptual “news” to experienced implementers.
  • Main value is as a checklist for building analyzers: shows how constructs like break, continue, goto, return, and complex for/while forms make loop semantics hard to reason about.
  • Several commenters stress that analysis should be done on a control-flow graph / IR, not raw AST; the paper’s CFG-heavy treatment reflects this necessity.

Loops vs Recursion (and Teaching Debate)

  • Strong debate over whether beginners should learn C-style loops or recursion (especially tail recursion) first.
  • One side: loops mirror everyday instructions (“do this N times”, “while dirty, wash”) and are easier for most people; recursion is conceptually harder and frequently misused or feared.
  • Opposing view: recursion is conceptually close to mathematical function application and to recursive structures in language and trees; loops are just an implementation detail over jumps.
  • Some argue functional-first education makes later paradigms easier; others say imperative style is more natural and that “recursion is beautiful” is mostly an in-group aesthetic.

Functional Style, Monads, and State

  • Discussion branches into FP vs imperative:
    • FP proponents highlight map/reduce, immutability, tail-recursive event loops, and clearer reasoning about side effects.
    • Critics note that once you add logging, timing, or I/O, code often falls back into imperative style; monads and general abstractions in Haskell are perceived as non-trivial to learn.
  • Recursion is acknowledged as powerful but often avoided in day-to-day production code due to readability, debugging, and team-comprehension concerns.

Compiler Behavior, Optimization, and Goto

  • Historical note: older compilers recognized syntactic for loops for optimization; modern compilers reconstruct loops from generic control flow (even if written with goto).
  • LLVM and others canonicalize loops into a small set of forms for analysis; not all loops fit, which echoes the paper’s theme.
  • goto is defended in some cases as a clearer way to express complex control flow than deeply nested loops; others see loops as safer, higher-level “domesticated gotos.”

Research Culture / Paper Format

  • Some dispute whether this counts as “research” or more as systematized documentation; others argue research need not be novel, only systematic and understanding-enhancing.
  • Complaints about traditional 2‑column PDF layout and academic gatekeeping; several call for valuing clear technical exposition over novelty posturing.

The loneliness of the low-ranking tennis player

Loneliness and Mental World of Tennis

  • Even top players describe tennis as intrinsically lonely; lower-ranked pros often feel both lonely and alone on the road.
  • The one‑on‑one, zero‑excuse nature of the sport makes failure feel very personal; there’s no team or randomness to “blame.”
  • Some see the article’s protagonist as strongly introverted and self‑isolating, even judging peers for socializing, which may have worsened his loneliness.

Economics of Lower-Ranked Pros

  • Strong consensus that prize money is heavily top‑heavy; Futures/Challenger players often lose money after travel, equipment, coaching, and support staff.
  • Side hustles (e.g., racket stringing, coaching, YouTube, “extreme couponing”) are common; some athletes in other sports report similar economics.
  • Disagreement on where “sustainable” earnings start: some say roughly top‑100; others argue many in the top‑150 can net good money, especially with sponsorships.
  • Dispute over typical monthly expenses (claims up to ~$20k/month vs. others calling that exaggerated for players outside the very top).

Comparisons to Other Fields

  • Parallels drawn to consultants constantly on the road, founders, creatives, esports, MMA, Olympic sports, and academia: high commitment, poor odds, top‑heavy rewards.
  • Distinction: consultants typically get guaranteed pay and transferable skills; athletes face aging, injury, and limited fallback options.
  • Tennis is contrasted with team sports: many more paying roles in soccer/basketball than in a 128‑slot singles draw.

Identity, Talent, and Quitting

  • Many emphasize that serious players are effectively “bred” from childhood, similar to classical musicians; their entire identity is bound to the sport.
  • This makes quitting psychologically akin to renouncing a religion or self‑annihilation, especially for those hovering just below breakthrough level (e.g., ranks ~300–600).
  • Debate on whether intensive training builds broadly useful discipline vs. leaving ex‑athletes “screwed” and burned out.

Status, Hierarchy, and Social Dynamics

  • Ranking strongly shapes social interactions: higher‑ranked players and analogous high‑status workers often ignore those below once they “move up.”
  • Commenters connect this to human tendencies toward hierarchy and ego, not unique to tennis; similar patterns are reported on trading floors and in crypto.

Youth Pipeline and Parenting

  • Junior tennis is described as intense: heavy travel, quasi‑homeschooling, huge parental investment, and sometimes physical overuse injuries.
  • Some parents explicitly keep tennis framed as “just a hobby”; others are seen as pushing children into a narrow, high‑risk life path.
  • Coaching, club “pro” work, and US college tennis are common exit ramps, but not emotionally easy ones.

Entrust Certificate Distrust

Scope of Entrust’s Problems

  • Many commenters note the technical mis-issuances were relatively minor.
  • The core issue is Entrust’s response:
    • Repeated failure to revoke misissued certs within required 5 days.
    • Continuing to issue non-compliant certs after being warned.
    • Incomplete or poor-quality incident reports and remediation plans.
    • Difficulty producing accurate lists of affected certificates.
    • Apparent inability to do mass revocations quickly, implying inadequate resourcing.
  • View: this pattern shows they cannot be trusted to handle a serious incident.

Why Distrust Now, Not Earlier

  • Some argue browsers gave Entrust “every chance” to avoid appearing heavy‑handed or “censoring the internet.”
  • Others think root programs waited until there was clear, historical evidence Entrust’s process improvement rate was inadequate.

Impact on Sites and Ecosystem

  • Many high-profile sites and payment processors reportedly use Entrust, including banks, airlines, government domains, and Cybersource.
  • Chrome is using CT-based SCT timestamps for a phased distrust to minimize breakage; existing certs work until certain dates.
  • Admins can explicitly override Chrome’s distrust, but that override is coarse (trust all / trust none), which frustrates some.

CA Business Model and Governance

  • Broad sentiment that commercial CAs are rent-seeking and exist only because browsers/OSes list them.
  • Several comments blame cost-cutting, non-technical management, and a belief they were “too big to fail.”
  • Others note Entrust has substantial non‑web‑PKI business (cards, HSMs, ID printers, BIMI), so this is reputationally severe but not existential.

Let’s Encrypt vs Traditional CAs

  • Let’s Encrypt is praised for automation, simplicity, and alignment of incentives; still subject to the same rules but has far fewer incidents.
  • Traditional CAs like Entrust mainly serve legacy/manual environments where moving to LE is politically or technically hard.

Certificate Transparency and Security

  • CT is highlighted as a major shift: Chrome/Safari and some platforms require CT-logged certs.
  • Debate on using SCT timestamps for trust decisions; some worry about log backdating, but monitoring tools now check for this.
  • CT considered a strong deterrent to covert MITM, though not foolproof.

BIMI and Logo Certificates

  • BIMI/VMC certs described by some as a “racket” or cash grab; others see value in phishing resistance and visual brand cues.
  • Trademark scope and potential logo duplication are cited as structural weaknesses.

CriticGPT: Finding GPT-4's mistakes with GPT-4

Model quality and “AI to fix AI”

  • Many see a trend where LLM quality problems are addressed by adding more AI (e.g., a critic model).
  • Some compare GPT‑4o with Claude 3.5 Sonnet: experiences differ, with some saying Claude is better at coding, others finding them similar.
  • Explanations floated: better training data, more interpretability work, or just proprietary differences that outsiders can’t really know.

Purpose of CriticGPT and RLHF pipeline

  • Commenters stress CriticGPT is mainly a tool to help human RLHF labelers write better critiques, not a user‑facing product.
  • Idea: detecting bugs is hard; judging whether a proposed bug report is valid is easier. CriticGPT turns “find a bug” into “is this critique correct?” for humans.
  • Better critiques → higher‑quality RLHF data → better base models “at the source.”

Reliability, critics-on-critics, and evaluation

  • Skeptics ask how we know the critic isn’t just adding more errors; some see “critics all the way down” / oracle problem.
  • Supporters note that if a mistake can be caught by either human or critic, overall detection can improve even if each is imperfect.
  • OpenAI’s reported result (human+CriticGPT preferred ~60% of the time) is seen by some as a meaningful gain, by others as only modest over random.

Hallucinations, truth, and terminology

  • Several argue “hallucination” is just LLMs doing what they’re trained to do: generate plausible text, not truth.
  • Others defend the term as a useful shorthand for confidently wrong, fabricated content—especially in coding assistance.
  • Debate over whether hallucinations are a distinct phenomenon, or just incorrect outputs from a probability model; some suggest alternative terms like “confabulation.”

LLMs as coding tools and review targets

  • Mixed experiences: some rely heavily on LLMs for structure and boilerplate, but distrust them for precise API details.
  • Many note LLM‑written code can be overly complex, brittle, or use nonexistent APIs, making strong tests and human review essential.
  • At workplaces, reactions range from banning AI‑generated code to heavily promoting enterprise LLM tools and “prompt engineering.”

Labor, ethics, and safety

  • Discussion highlights low‑paid data labelers in developing countries versus better‑paid expert reviewers; cost and exploitation concerns are raised.
  • Some connect CriticGPT to broader alignment ideas like iterative amplification and recursive reward modeling, with ongoing skepticism about whether such stacks of AI reviewers fundamentally solve safety.

Moaan InkPalm Plus is weird, cheap, small, and my kind of e-reader

Physical buttons vs. touchscreens

  • Many commenters strongly prefer physical page-turn buttons for one‑handed use, reliability, and avoiding accidental turns and smudged screens.
  • Others find touch-only devices (Paperwhite, Scribe, etc.) “good enough” and don’t miss buttons.
  • Some hybrid devices allow both, but not always with a way to fully disable touch.
  • Bluetooth “camera clicker” remotes and volume buttons on Android e‑ink devices are popular hacks to regain “buttons.”

Battery life and lighting

  • Oasis and some modern readers are criticized for weaker battery life versus older Kindles; a week vs. a month is a common contrast.
  • Airplane mode is repeatedly cited as a major battery saver, but some still find life disappointing.
  • Front/back lights with very low brightness and adjustable color temperature are viewed as essential; inability to turn light fully off is seen as a design flaw on some devices.

Form factors: phone-shaped vs. traditional e-readers

  • Phone-sized e‑ink devices (InkPalm, Boox Palma, Hisense A9, etc.) are praised for pocketability and for replacing doomscrolling with reading.
  • Critics dislike the tall, narrow aspect ratio: too few words per line, awkward justification, and “phone-shaped” redundancy given phones already exist.
  • Supporters argue narrow columns can aid eye tracking and that e‑ink is noticeably easier on the eyes and more focused than phones.

Typography and readability

  • Short lines and poor justification/hyphenation are frequent complaints; some blame manufacturer software and prefer KOReader for finer control.
  • Others disable justification/hyphenation entirely due to bad markup in many ebooks.

Use cases and reading habits

  • Transit commuters and backpackers like very small devices they can always carry.
  • Some report reading almost exclusively on phones/tablets for simplicity; others reserve e‑ink for focused, long-form reading and outdoor use.
  • Motion sickness while reading in vehicles leads some to prefer audiobooks.

Large-format e-ink for PDFs and technical books

  • Several recommend 10–13" e‑ink tablets (Boox Max, Kindle Scribe, reMarkable, Sony DPT) for letter-sized or technical PDFs.
  • Trade‑offs include weight, lack of backlight on some models, and limited formats/transfer options.

Software, openness, and Android

  • Android-based e‑ink devices are valued for running KOReader, Instapaper, Kindle/Kobo apps, and even terminals/browsers optimized for e‑ink.
  • There is interest in open, up-to-date, privacy-respecting e‑ink devices (LineageOS on Hisense, desire for Samsung/Google/Apple entries), but current offerings are seen as fragmented and often locked down.

Gemma 2: Improving Open Language Models at a Practical Size [pdf]

Release & Availability

  • Gemma 2 comes in 2.6B, 9B, and 27B variants; 9B/27B are already on Ollama and Hugging Face, with GGUF quantizations from community members.
  • Support has been added to gemma.cpp, and the 27B IT model is available in Google AI Studio (playground; API integration still “soon” / partial).

Architecture, Tokens & Training

  • Uses explicit special tokens (<bos>, <eos>, <start_of_turn>, <end_of_turn>).
  • Discussion emphasizes their role in training: packing multiple short sequences into fixed-length batches to reduce padding and avoid cross-example leakage.
  • Some argue masking could replace them; others note BOS/EOS make large-scale data packing simpler and safer.
  • Distillation is a central technique: student models learn from teacher logits, effectively “learning a whole distribution” per step, which is argued to be like training on many more tokens.

Comparisons to Phi‑3, Llama 3, Mistral

  • Multiple commenters feel Gemma 2 (especially 2.6B/9B) scores worse than Microsoft’s Phi‑3 in standard benchmarks.
  • Others counter that Phi‑3 appears overfit to benchmarks and underperforms in interactive settings (LMSYS ELO, subjective tests).
  • Gemma 2–27B reportedly ranks near Llama‑3‑70B, GPT‑4, and Claude Sonnet on Chatbot Arena; some personal tests disagree and find Llama‑3‑70B clearly stronger.
  • Accusations of “parameter creep” (9B vs 7–8B peers) and mixed views on whether comparisons are fair.

Context Window & Inference

  • 8K context with a 4K sliding window is seen as a speed–quality tradeoff; some criticize it as too small for serious RAG, others say “effective” 8K is preferable to nominal 32K with degradation.
  • Ollama currently halves context when hitting the limit, which appears to destabilize Gemma 2–27B in some setups; maintainers plan to change this to hard-limit behavior.

Quantization & Local Use

  • 4-bit GGUF versions are popular for local deployment; people debate how much quantization actually hurts quality, with papers cited suggesting subtle regressions, especially on factual tasks.
  • Subjective reports range from “indistinguishable from full precision” to visible degradation in edge cases.

Safety, Alignment & Capabilities

  • Self‑proliferation abilities (e.g., autonomously setting up remote LLMs) reportedly score 0/10 in internal tests; some argue this reflects incapability more than true alignment.
  • Training data is filtered for “unsafe” content; some see this as necessary liability management, others question value given the same information is easy to find via web search.

Benchmarks vs Real Use & Task Fit

  • Benchmarks show Gemma 2 9B lagging Phi‑3 Small on many academic tasks; users report Gemma 2 may be better as a general assistant and conversational model.
  • Mixed experiences on coding: some find Gemma 2 structured and pleasant (no verbose preamble), others report severe nonsense on long code outputs or heavy context.
  • Noted strength: 27B appears unusually strong at multilingual translation, including less common languages, though this is absent from the paper’s emphasis.

APIs, Licensing & Cloud UX

  • Licensing matches Gemma 1; terms are proprietary but permit broad use.
  • Some praise AI Studio’s simplicity vs Google Cloud / Vertex; many complain GCP billing, region restrictions, and documentation are confusing compared to OpenAI/Mistral-style APIs.
  • There is a preview OpenAI-compatible endpoint for Gemini on Vertex, but token/auth model makes drop‑in use harder.

Not everything is behavioral science

Acronym, title, and expectations

  • Many readers were initially confused by “BS” meaning “behavioral science” rather than “bullshit.”
  • Several felt the title and opening were clickbait and expected an essay on “bullshit,” not on behavioral science.
  • Some asked for a more accurate, less sensational title.

Perceptions of the article and behavioral science

  • Multiple commenters found the piece smug, sloppy, and full of under-examined half-truths.
  • The article is seen by some as self-serving: an instance of the idea that institutions preserve the problems they exist to solve.
  • Others argue that any field tends to see the world through its own lens; this can be either bias or a useful perspective.
  • There is concern about the reproducibility crisis and “nudge”/behavioral interventions that don’t pan out long term.
  • Defenders say the aim is to highlight overlooked, non-technical solutions and that perception/marketing is often part of “the product,” not an afterthought.

Placebo effect and medical testing

  • Several commenters say the article misunderstands placebo: clinical trials subtract placebo to measure drug effects, not because practitioners want to avoid placebo in real treatment.
  • One detailed comment notes placebo is malleable and can be enhanced, and that drug developers already implicitly exploit side effects that may amplify perceived efficacy.
  • Others stress the difference between testing medicine and treating patients, and compare placebo control to using a naïve baseline in predictive modeling.

Two-dishwasher thought experiment

  • The “two dishwashers, never unload” idea draws extensive criticism on space, cost, plumbing, inefficiency, hygiene, and real-world usage patterns.
  • Some point out it only fits certain lifestyles, house types, and dish inventories, and often fails under realistic load imbalance.
  • Others note aesthetic and psychological reasons to prefer clean cabinets over appliances as storage.
  • A minority argue it can work in specific contexts (small households, new construction) and that people may overestimate the waste vs convenience.

Constraints, thought experiments, and “outside the box”

  • The candle/boiling-water-at-altitude example is widely attacked as contrived: if constraints are not clearly specified, any “solution” becomes possible and trivial.
  • Critics say meaningful problem-solving requires accepting reasonable constraints; otherwise, “thinking outside the box” becomes empty wordplay.
  • Supporters respond that the point is precisely to notice variables assumed fixed (like altitude) and consider changing them.

Solar panels aside

  • Some report very high neighborhood uptake of residential solar, suggesting engineering advances clearly did change behavior.
  • Others, even in very sunny areas, see slow adoption and resistant homeowners, indicating non-technical barriers.
  • Leasing models and contracts attached to houses are cited as deterrents; ownership vs leasing trade-offs are debated.

Meta: disciplines as lenses & limits of rationality

  • One thread generalizes that every field offers a framework (behavioral, architectural, historical, logical) that can be applied broadly; the challenge is picking the right “hammer.”
  • Another subthread discusses how much cognition is unconscious (e.g., juggling), and how this complicates rational analysis and self-assessment of expertise.
  • There is debate over how far rational decomposition scales to complex systems, and whether some domains make it impossible to know when one is wrong.

It's Now Cheaper to Lease a Tesla Model 3 Than a Toyota Camry

Lease economics & pricing structure

  • Current EV market favors aggressive lease incentives over sale price cuts, aided by federal tax credits that apply to leases even when they don’t to purchases.
  • Low Model 3 lease payments are attributed to high assumed residual values plus the $7,500 credit passed through in leasing. Tesla bears residual risk; lessees benefit if those assumptions are wrong.
  • Some note December as a peak for lease deals due to holiday promos and lease cycles.
  • Comparisons with Camry lease/purchase prices are seen as “not fully fair” because of subsidies and differing residual assumptions.

Total cost of ownership (TCO) & maintenance

  • One side argues Camry is cheaper long‑term: proven durability, abundant cheap parts, ubiquitous mechanics.
  • Others counter that EVs, and Tesla in particular, have far fewer moving parts and already show lower 10‑year maintenance/repair costs in some reports.
  • Debate over whether those TCO studies are reliable yet, given Model 3/Y are younger than 10 years.

Battery longevity & replacement

  • Some commenters assume expensive battery replacements every few years; others call this a myth.
  • Multiple anecdotes of 6–7‑year‑old EVs with ~5–10% degradation and no issues, plus references to very high‑mileage Teslas.
  • Claim that modern chemistries routinely support thousands of cycles, implying batteries likely outlast the car; skeptics remain unconvinced, citing high hypothetical replacement cost.

Subsidies, pricing, and fairness

  • Some resent that federal credits make a luxury‑leaning EV cheaper than a mass‑market Camry.
  • Others argue fossil fuels have historically received larger indirect subsidies, and EV credits just shift benefits from producers to consumers.
  • Disagreement over whether tax credits let manufacturers raise prices, or simply increase demand at the same optimal price.

Features, UX, and ownership constraints

  • Missing CarPlay/Android Auto and physical controls are dealbreakers for some; others say Tesla’s native system suffices.
  • Lack of lease buyout on Teslas is seen as a downside compared to other brands, with speculation this is tied to “robotaxi” plans.
  • Access to home/work charging is highlighted as a practical prerequisite; renters without charging are seen as effectively pushed toward ICE/hybrids.

Ethics, environment, and alternatives

  • Concerns raised about lithium mining (including child labor) and China’s dominance in battery supply chains.
  • Counterpoints: fossil fuels have systemic environmental harms; many hybrids (e.g., new Camry) also use lithium packs.

Vehicle preferences & culture

  • Some would still choose a Camry for perceived reliability, cold‑weather performance, privacy, and repairability.
  • Others report test‑driving a Tesla and immediately preferring its driving experience, despite size/identity shifts from trucks or SUVs.

At 50 Years Old, Is SQL Becoming a Niche Skill?

Is SQL Becoming Niche?

  • Many argue “no,” invoking Betteridge’s law: SQL remains foundational and widely used across back-end, data engineering, and data science.
  • A minority view: “good/serious” or “advanced” SQL is becoming niche, especially in new systems where databases are treated as dumb storage and complexity moves into application code.
  • Another framing: SQL-for-development (complex modeling, optimization) is getting more specialized, while SQL-for-access (ad hoc reads, reporting) is increasingly common and democratized.

DBAs, Performance, and Schema Design

  • Teams often treat SQL as a background skill until performance collapses; then they “suddenly want a DBA.”
  • Poor initial schema/index design leads to painful refactors when load grows.
  • There are few DBAs relative to the number of projects; those with intermediate–strong skills report very high demand and pay.
  • Typical fixes: adding/removing/consolidating indexes, restructuring “hero queries,” breaking work into smaller units, and working around ORM-generated SQL.

ORMs, NoSQL, and Abstraction Layers

  • ORMs are seen as productive for 80–90% of CRUD, but often get in the way for complex queries; many developers eventually want to “just write SQL.”
  • A lot of engineering effort goes into avoiding SQL via ORMs, JSON/REST layers, or “hipster” web databases, while a relational engine still sits underneath.
  • Some teams start on MongoDB/DynamoDB for cost/simplicity, then migrate to PostgreSQL when requirements grow; others regret NoSQL as a default and are actively moving back to Postgres.
  • Several note SQL’s expressiveness and concision are hard to beat; alternative syntaxes (ORM chains, pipelines, PRQL, QUEL) are usually less compact or familiar.

SQL in Data Science and Analytics

  • In data engineering and many analytics-heavy domains, SQL is “bread and butter,” used for hours daily and preferred over Python for data processing until genuinely outgrown.
  • Some data scientists underuse SQL, doing only extraction before moving to Excel/Pandas, which others see as a productivity and collaboration loss.

LLMs, Tools, and Skills

  • One view: LLMs already write complex SQL better than most humans, making advanced SQL less worth memorizing.
  • Counterviews report LLMs failing on anything beyond simple joins/aggregates; prompts for real-world complexity are hard to express.
  • Advice recurs: modest SQL study (books, tutorials, RDBMS docs) rapidly differentiates engineers and pays off significantly.

Open-Sora does pretty good video generation on consumer GPUs

Video Quality and Technical Limits

  • Many find Open-Sora’s results “janky,” “blobby,” and morphing, especially in motion and detail; others say it’s impressive given it runs on consumer GPUs.
  • Temporal coherence is a major critique: models like Stable Diffusion applied frame-by-frame produce flicker; Open-Sora is seen as only marginally better.
  • Some suggest hacks (e.g., processing grids of multiple frames) but note GPU RAM limits.
  • Compared to very early AI videos (e.g., “Will Smith eating spaghetti”), Open-Sora is seen as a big step up, usable at least for mood-setting, comic-like storytelling or low-res viewing.

“Pretty Good” and Frame of Reference

  • One camp argues that “pretty good” is fair relative to other consumer-GPU tools and the rapid recent progress.
  • Another argues that being better than other poor outputs is not meaningful to non-enthusiasts; it’s still far from production quality.
  • Some emphasize that Open-Sora is actually available, unlike Sora, which is seen as effectively inaccessible despite demos.

Trademarks and “Open-Sora” Naming

  • Multiple comments question whether using “Sora” violates OpenAI’s trademark, citing registrations.
  • Views range from “you can do anything until you’re caught” to predictions of takedown requests or legal threats against GitHub.
  • Debate over IP more broadly: some see AI developers as dismissive of IP; others argue trademarks help avoid consumer confusion.

Broader IP and AI Ethics

  • Extended subthread on whether IP (especially copyright/patents) helps or harms innovation and culture.
  • Critics say current IP regimes favor large corporations, entrench monopolies, and obstruct open models; defenders argue IP is a moral or practical necessity to incentivize creation.
  • Specific worry: AI firms oppose IP limits on training data but aggressively protect their own models and outputs.

Societal Impact and Sentiment

  • Fears: job loss in video/film, explosion of misinformation, and further detachment from physical reality.
  • Counterpoints: tools democratize filmmaking and design, reduce logistical burdens, and mainly automate repetitive rather than creative work.
  • Some are apathetic or hostile to AI hype, automatically downgrading content illustrated with AI imagery and seeing much of it as low-effort “slop.”

Google Sheets ported its calculation worker from JavaScript to WasmGC

Performance and results

  • Initial WasmGC port of the Sheets calculation engine ran about 2× slower than the existing JS-targeted version.
  • After substantial optimization (devirtualization, speculative inlining, better data structures, and using native browser APIs like RegExp), the WasmGC version is about 4× faster than that prototype and roughly 2× faster than the JS version.
  • Java-on-WasmGC is said to be ~2× slower than Java on the JVM, so there is still a gap to native JVM performance.

Java, JavaScript, and what’s actually being compared

  • The engine is written in Java and compiled either to JS (via J2CL/GWT) or to WasmGC (via J2Wasm).
  • Some argue the article blurs the distinction by talking about “JavaScript vs Java,” when it is really comparing compilation targets for the same Java code.
  • Others counter that the “JS version” is a real JS bundle delivered to the browser and historically replaced a handwritten JS implementation, so comparing “JS vs WasmGC” is fair from a web-performance perspective.

Why WasmGC matters

  • Without WasmGC, GC’d languages must ship their own collectors in Wasm, inflating binary size and complicating cross-module and host integration.
  • WasmGC lets languages reuse the browser’s GC, improves interop (e.g., sharing heap objects across modules/host), and avoids tricky “hybrid” GC schemes.
  • Limitations remain: GC arrays lack some low-level ops (e.g., efficient reinterpretation of bytes as ints), and sharing packed GC arrays efficiently with WebGL/WebGPU/Web Audio is still awkward.

Browser support and fallbacks

  • Chromium-based browsers and Firefox support WasmGC; Safari/WebKit work is ongoing but incomplete.
  • Sheets can still serve a JS-compiled version where WasmGC is unavailable.

Comparisons to Excel and scalability

  • Native Excel (C++) is expected to remain faster, with no 4 GB Wasm memory cap and real shared-memory threads.
  • Users report Sheets becoming slow or hanging with large, complex models, while Excel handles similar workloads better; others mainly notice UI sluggishness, not raw calculation speed.

Offline, native, and platform lock-in

  • Some want a true offline/standalone Sheets app; today’s offline mode requires Chrome/Edge and an extension.
  • This raises complaints about browser lock-in and potential antitrust concerns; others note limited incentive to support more browsers.

Meta: publication dates and credibility

  • Several comments criticize the article’s lack of a clear original publication date and describe a broader trend of hiding dates for SEO/engagement, which they see as harmful to trust and technical context.

Post Office lawyers held secret meeting with judge to stop disclosure

Overall reaction to the scandal

  • Many see the Horizon/Post Office case as a deepening disaster and emblematic of systemic rot, not a one-off error.
  • Strong anger that no one (so far) has gone to jail despite ruined lives and suicides.
  • Some frame this primarily as a failure of the judiciary and its closeness to other establishment bodies, not just the Post Office.

Accountability and leadership responsibility

  • Debate over whether ultimate responsibility should rest at the top:
    • One side argues leaders should be held criminally accountable for crimes committed under their watch, not just “rogue employees.”
    • Others call that impractical and support liability only when leaders knowingly aid, abet, or cover up wrongdoing.
  • Examples from regulated industries and banking are cited where senior managers can face criminal liability for systemic failures.

Class, race, and discrimination

  • Disagreement on whether this is mainly “peers vs poors,” or driven more by racism and suspicion of immigrant sub‑postmasters.
  • Lords debate is referenced: a disproportionate share of prosecutions involved immigrants, especially Asian women, which commenters find hard to reconcile with general prison demographics.

Legal process: disclosure, ex parte, and public interest immunity

  • Many view the secret meeting with the judge and delayed disclosure as perverting justice.
  • Some note that in the US, undisclosed exculpatory evidence or ex parte meetings often trigger mistrials or appeals.
  • Others clarify that the lawyer who sought the meeting is reported to have later stopped prosecutions and that initial nondisclosure was time‑limited due to parliamentary rules.
  • “Public interest immunity” is widely seen as a recurring tool used to shield powerful actors and enable miscarriages of justice.

Public defenders, plea deals, and systemic bias

  • Long sub‑thread on US criminal justice:
    • Public defenders are underpaid and overloaded, often providing minimal or purely procedural defense.
    • Many cases are described as “obviously guilty,” but several commenters insist even clearly guilty defendants deserve robust defense and due process.
    • Heavy reliance on plea bargaining and resource asymmetry between prosecution and defense are seen as producing a “conviction system” rather than a justice system.

Inquiries, reform, and distrust in institutions

  • Skepticism that public inquiries lead to real consequences; they are often seen as slow‑motion deflection until public attention fades.
  • Others argue the “drip drip” can still build pressure and eventually produce accountability.
  • Comparisons to other UK scandals (Grenfell, contaminated blood, undercover policing, Hillsborough, Bloody Sunday) reinforce a pattern of extreme delay.
  • Broader sense that UK institutions (courts, politics, immigration system) are increasingly distrusted and perceived as corrupt or self‑protective.