saturnV's recent activity

  2. Comment on Rob Jetten becomes Netherlands' youngest ever PM in ~society

    saturnV
    Link Parent
    income inequality measured by gini coefficient has been falling for the last decade, is now fairly low compared to rest of europe and economic growth seems to be on par with the rest of europe

    income inequality measured by gini coefficient has been falling for the last decade, is now fairly low compared to rest of europe and economic growth seems to be on par with the rest of europe

    1 vote

  3. Comment on Why are American passenger trains slow? in ~transport

    saturnV
    (edited )
    Link Parent
    strongest counterargument I've found so far is that the difference in cargo type affects the incentives. US still transports a lot of coal whereas this is less common in Europe, which allows for...

    strongest counterargument I've found so far is that the difference in cargo type affects the incentives. US still transports a lot of coal whereas this is less common in Europe, which allows for slow, long trains which are described in the article, but as coal gets phased out those trains will become less common as freight is dominated by more time-sensitive loads, undoing some of the negative incentives
    As of 2023, coal accounted for 28% of total rail tonnage
    Also the article mentions but underemphasises the importance of how big & empty the US is compared to europe, making rail freight inherently more attractive independent of regulatory choices
    oh also swiss rail sorta damages the argument because they have both plenty of passenger and freight on their railways, even though the conditions are slightly different in some obvious ways
    Also it's very handwavy about why US has to intermingle freight and passengers and can't just build separate railways
    overall I think the article is factually correct and useful but talks insufficiently about upstream factors that create the premise that freight and passenger rail have to be shared
    (this is obviously partially attributable to the fact that this article is written for a conservative journal, something I only checked after reading the article but had some suspicions)

    12 votes

  4. Comment on Why are American passenger trains slow? in ~transport

    saturnV
    (edited )
    Link Parent

    The American choice made sense given American conditions. The geography here favored freight: vast distances between population centers, bulk commodity production (grain, coal, minerals) in the interior requiring long-haul transport to coastal ports, and low population density that makes frequent passenger service uneconomical outside a few corridors. Ownership structures favored freight: private railroads answering to shareholders had to earn returns on capital, and that meant maximizing returns from freight, since Amtrak had assumed responsibility for passengers. Competition favored freight: trucking could not and cannot match rail’s cost advantage for bulk commodities over continental distances. American freight railroads move cargo at roughly one-quarter the cost per ton-mile of trucking; for bulk goods over long distances, rail has no effective competitor.

    Given different conditions in Europe, Europeans made different choices. Geography there favors passengers: shorter distances between major cities, higher population density, and an extensive coastline that made coastal shipping competitive for freight. Ownership favored passengers: state-owned railways accountable to politicians naturally chose to serve citizens, not shareholders, meaning that passenger service was politically visible and salient in a way that freight was not. And competition favored passengers: high fuel taxes made driving expensive, and rail remained competitive with air for journeys of two hundred to six hundred kilometers where travel times between city centers favored trains over the airport-to-airport-plus-security-ritual of flying.

    Amtrak’s statutory dispatching preference exists on paper, but enforcement has been vanishingly rare. Until 2024, the Department of Justice had brought exactly one lawsuit to enforce Amtrak’s statutory rights: a 1979 case against Southern Pacific over the Sunset Limited between New Orleans and Houston, which ended in a consent decree without a final court ruling. In July 2024, the DOJ filed a second suit, this time against Norfolk Southern (NS) for its handling of the Crescent, which settled in 2025 with explicit commitments to prioritize Amtrak trains and require supervisor approval for any dispatching decision that does not. Two lawsuits in forty-five years is not a robust enforcement regime. And Amtrak has paid a price for it: freight train interference remains the leading cause of Amtrak delays on host railroads, causing over 850,000 minutes of delay in 2024, more than any other single category of delay.

    The realistic paths forward are more modest. On specific corridors with sufficient traffic, like Michigan, Virginia, North Carolina, or California, public acquisition and sustained investment can produce genuine improvement. The toolkit exists: purchase strategic segments, fund ongoing maintenance, transfer dispatching authority, design investments that serve freight interests alongside passenger interests (as Virginia’s Potomac crossing does for CSX). These interventions work and can be replicated.

    But they will not produce a European-style national network. That gap is structural, driven by geography, commodity flows, population distribution, and the competitive position of alternative modes. These are facts about the physical and economic landscape, not mere policy failures we can correct.

    4 votes

  5. Comment on Why are American passenger trains slow? in ~transport

    saturnV
    Link Parent

    These are particular examples of a general problem. Across the American rail network, passenger trains run slower today than they did before the Second World War. What happened?

    None of the obvious explanations, like underinvestment in rail assets or the ubiquity of car culture, solve the puzzle. If car culture were determinative, we cannot explain why Europe maintains faster rail service than the United States, even though 74 percent of European intercity trips are taken by automobile, only modestly below America’s 85 percent. And political will cannot be the binding constraint either, since bipartisan infrastructure bills consistently include rail funding, but none have reversed the decline.

    These explanations fail because they assume something has gone wrong. If we start instead with the premise that something has gone right, the puzzle solves itself. American passenger trains are slow because the United States established a framework after 1970 that made optimization for freight service inevitable. Private railroads, freed from mandatory passenger service, rationally invested in what paid: moving cargo. That optimization succeeded spectacularly: today, America moves over five thousand ton-miles of freight per person annually by rail, a rate nine times that of Europe’s and fifty-four times Japan’s.

    Put another way, the United States has excellent intercity rail service . . . for freight. The United States features the most productive freight railroad in the world, and by optimizing for that, it has chosen not to pursue passenger speed. Doing both simultaneously on shared infrastructure is difficult, if not impossible. Four interlocking barriers—incentive structures, infrastructure standards, regulatory requirements, and operational arrangements—keep passenger speeds low: not by accident, but by design.

    6 votes

  6. Comment on Why are American passenger trains slow? in ~transport

    saturnV
    Link
    tl;dr piece argues reason US passenger rail is slow is because of rail-freight being better value for private railroads, thus causing US railways to be optimised for freight and making it...

    tl;dr piece argues reason US passenger rail is slow is because of rail-freight being better value for private railroads, thus causing US railways to be optimised for freight and making it nigh-impossible to have fast&cheap passenger rail
    Europe is different because shipping is more practical (more coastal ports) + more regulatory intervention by government to encourage/subsidise passenger rail

    13 votes

  8. Comment on Building a C compiler with a team of parallel Claudes in ~tech

    saturnV
    Link Parent
    I partially agree with this, but most of the interestingness of this mentioned in the article is the fact that previous models couldn't do this but the newest one can (and anyways it's not an...

    and the 100k lines of code are ridiculously high.

    I partially agree with this, but most of the interestingness of this mentioned in the article is the fact that previous models couldn't do this but the newest one can (and anyways it's not an insane amount of overhead given the fact it handles so many edge-cases), e.g. tcc is ~40k lines, maybe you'd expect this one to be less because rust is more expressive, but still only an order of mag overhead (generously) which isn't that bad

    also there really aren't that many non-toy c compilers, I would be careful making it out as trivial, and I think much of your comment reads as disingenuous when reading the article and seeing how plain and uninterested in hype it is, e.g.

    The compiler is an interesting artifact on its own, but I focus here on what I learned about designing harnesses for long-running autonomous agent teams: how to write tests that keep agents on track without human oversight, how to structure work so multiple agents can make progress in parallel, and where this approach hits its ceiling.

    The compiler, however, is not without limitations. These include:
    It lacks the 16-bit x86 compiler that is necessary to boot Linux out of real mode. For this, it calls out to GCC (the x86_32 and x86_64 compilers are its own).
    It does not have its own assembler and linker; these are the very last bits that Claude started automating and are still somewhat buggy. The demo video was produced with a GCC assembler and linker.
    The compiler successfully builds many projects, but not all. It's not yet a drop-in replacement for a real compiler.
    The generated code is not very efficient. Even with all optimizations enabled, it outputs less efficient code than GCC with all optimizations disabled.
    The Rust code quality is reasonable, but is nowhere near the quality of what an expert Rust programmer might produce.

    The resulting compiler has nearly reached the limits of Opus’s abilities. I tried (hard!) to fix several of the above limitations but wasn’t fully successful. New features and bugfixes frequently broke existing functionality.

    Also I think it's relevant to this that the author nicolas carlini is a very skilled programmer in his own right, and has engaged deeply with the strengths and weaknesses of AI for years now (see https://nicholas.carlini.com/writing for both)

    4 votes

  9. Comment on SpaceX is acquiring xAI in ~space

    saturnV
    Link Parent
    It makes sense to do it before the IPO to avoid the regulatory hassles that come with publicly listed companies. While they're both still private, elon has zero accountability and can easily do...

    Suspiciously close to the rumored IPO?

    It makes sense to do it before the IPO to avoid the regulatory hassles that come with publicly listed companies. While they're both still private, elon has zero accountability and can easily do this, whereas after, he'd end up getting sued by shareholders and have to endure a long legal battle

    5 votes

  11. Comment on How AI assistance impacts the formation of coding skills in ~tech

    saturnV
    Link
    Evidence from anthropic around the tradeoffs of different ways of using AI, where the more you delegate the more your skills will atrophy. However, using AI to generate code snippets can still be...

    Evidence from anthropic around the tradeoffs of different ways of using AI, where the more you delegate the more your skills will atrophy. However, using AI to generate code snippets can still be net positive. n=52 isn't super conclusive but still an interesting topic, looking forward to future research.

    Also interesting that this is something anthropic researchers feel comfortable publishing about when it could be pretty easily spun into something anti-AI

    11 votes

  14. Comment on China have a new sixty-centimeter dome Terahertz telescope in Antarctica, a two week trek from their station in ~space

    saturnV
    Link Parent
    update, author replied:

    update, author replied:

    The HEAT telescope did make some observations but they were never published, so scientists have found it hard to determine exactly what HEAT accomplished. I probably should have mentioned HEAT in the story.

    6 votes

  15. Comment on China have a new sixty-centimeter dome Terahertz telescope in Antarctica, a two week trek from their station in ~space

    saturnV
    Link Parent
    Wow yeah. do you mind if I email the author of the article about this? I feel like this needs a correction, or at least get re-edited to mention HEAT

    Wow yeah. do you mind if I email the author of the article about this? I feel like this needs a correction, or at least get re-edited to mention HEAT

    15 votes

  16. Comment on China have a new sixty-centimeter dome Terahertz telescope in Antarctica, a two week trek from their station in ~space

    saturnV
    Link
    It is very difficult for astronomers to scan the terahertz frequencies between infrared and radio waves because water vapor in Earth’s atmosphere absorbs them. Astronomers in China have managed to...

    It is very difficult for astronomers to scan the terahertz frequencies between infrared and radio waves because water vapor in Earth’s atmosphere absorbs them. Astronomers in China have managed to make it work by putting a terahertz telescope in the driest place on Earth, but it wasn’t easy. The instrument, the Antarctic Terahertz Explorer, has a 60-centimeter dome and has been painstakingly hauled out to Dome A, the highest point on a plateau four kilometers above sea level on the East Antarctic Ice Sheet. The 1,228-kilometer trip from Zhongshan Station to Dome A took two weeks by tracked vehicle. The ATE60 is the first step of many, as the Chinese Academy of Sciences has agreed to fund at least two one-meter telescopes at Dome A.

    13 votes

  20. Comment on Postal arbitrage in ~finance

    saturnV
    Link

    As of 2025, a stamp for a letter costs $0.78 in the United States. Amazon Prime sells items for less than that... with free shipping! Why send a postcard when you can send actual stuff?

    5 votes