One popular thing that works on Windows and not on Linux is Apple Mobile Device Service. This is the component of iTunes that lets a user sync MP3 albums, such as those bought on itch.io or Bandcamp, onto the Music app of an iPhone over a USB cable. As a driver, it is outside the scope of Wine.

Until the maintainer receives an issue report requesting a package for a particular distribution. For every one report, there are probably dozens of cases where another prospective user considered using a particular piece of software but did not because there was no package.

I read Package Forge's inclusion criteria. In order to get distributed in PkgCache, the application first needs to "be a well known package" and listed on a website called Repology. And Repology's inclusion criteria appear to be meant for distributions' repositories, not individual project maintainers' repositories.

Provided your application already has enough users compiling it from source code to justify packaging it in the first place.

I think allo is referring to the "game-key cards": Nintendo Switch 2 cartridges that contain the title screen and nothing else, where the whole game is shipped as a day-one update. At least on the original Switch, Nintendo required the first few missions of the single-player campaign to be on the cartridge. Among early physical games for Nintendo Switch 2, only Nintendo's first-party games and CD Projekt's Cyberpunk 2077 weren't game-key cards.

It'd look like a subset of C where the compiler emits a diagnostic for every undefined behavior that's practical to detect at compile time and inserts code to catch at runtime everything else the standard calls undefined. The first step toward this is what GCC already does for -Wall -Wextra -pedantic -fsanitize=address,undefined. The second step is that a pointer variable doesn't contain a raw address but instead a base address and index, and every dereference of an array member is bounds-checked against the size of the object it came from. This ends up making the language's strict aliasing rule even stricter, and a lot of pointer casts or union puns become undefined and therefore errors. After programmers become accustomed to stricter pointer provenance, a compiler maker can add a concept of ownership, with a borrow checker to detect use-after-free and the like.

If the only programs with permission to escape the language's type system are system libraries signed by the operating system publisher, it isn't "totally useless." It'd be like Java applets, J2ME phone applications, Silverlight applets, XNA games, and Windows Phone 7 applications. Executables for these platforms are in an intermediate representation that lets the loader tell if unsafe was used. If the executable contains unsafe code but isn't signed with the permission to escape the sandbox, the loader raises a security exception. Applications were expected to perform I/O through first-party system libraries signed with this permission. Third parties weren't supposed to be doing "systems programming" on these platforms.

Yeah, one of the things I like about Claude (and Gemini 3 as opposed to 2.5) is that they really clamped down on the use of "Oh, now I've got it! This is absolutely the FINAL fix to the problem, we've totally solved it now! Here, let me write out FIX_FINAL_SOLVED.md" with some half-arse solution. And yep, the answer to going in circles is usually either "nuke the chat" or "switch models".

Which is why Google waited several years to brick early Nest and Revolv thermostats: the factory warranty had expired.

Exactly. Gasoline, mains power, and batteries are standardized. So are LTE, 5G NR, and Wi-Fi. Compare what Mike Masnick of Techdirt and other Internet user freedom advocates have called "protocols, not platforms."

Though even if there were no cryptographic lockdown of these "smart" devices' system software to interact only with the vendor's server, one big obstacle to running your own server (with proverbial blackjack and hookers) is that so many Internet providers nowadays block inbound TCP connections. T-Mobile Home Internet, for example, puts subscribers behind carrier-grade network address translation (CGNAT) with the whole neighborhood behind one IPv4 address. Even through IPv6, their gateway appliance offers no port forwarding or DMZ option.

Yeah, when thinking of the typical air fryer market, think "working mom with kids who wants to serve something nicer than a microwave dinner, but doesn't have the time for much prep or waiting". You can get those mailard reactions that microwaving doesn't really get you, nice crisping and browning of the surface that you normally get from an oven, without having to wait for an oven to preheat. I don't think anyone disputes that an oven will do a better job, but the air fryer does a better job than a microwave, which is what it's really competing against. They're also marketed as easy-clean, which again is a nod to their target audience.

How costs build up is really staggering. I'm getting into the business of importing 3d filament. In Iceland, it currently sells for like $35/kg minimum. The actual value of the plastic is like $1. The factory's total cost, all costs included, is like $1,50. If it's not name brand, e.g. they're not dumping money on marketing, they sell it for $3 for the cheapest stuff. Sea freight adds another dollar or two. Taxes here add 24%. But you're still at like $5/kg. The rest is all middlemen, warehousing, air freight for secondary legs from intermediary hubs, and all the markup and taxes on those things.

With me importing direct from the factory, sea freight only, I can get rid of most of those costs. Warehousing is the biggest unavoidable cost. If I want to maintain an average inventory of like 700kg, it adds something like $5/kg to the cost. Scanning in goods and dispatching user orders (not counting shipping) together adds like $2,50. And then add 24% tax (minus the taxes on the imported goods). There's still good margin, but it's amazing how quickly costs inflate.

Your "junk" is someone else's "must-have".
Your "must-have" is someone else's "junk".

What's the next step after having been rejected by several companies that had posted entry-level positions in your specialty within 10 km of your home?

It was meddling by both D and R in our economy, both were scared of invisible boogiemen of "something bad might happen".

Fear is a great motivator. Courage is standing in the face of danger understanding the risks might be worse doing nothing than doing something. This is a calculated risk and ought to be rewarded in the marketplace if it is correct.

Conglomerates are neither good nor bad in and of themselves. The good is they offer efficiencies in the marketplace. The bad is they take advantage of those efficiencies and often get "too big to fail" (a lie).

People guessing who have no stake in the market are making bad choices, because of other reasons. Both D and R do this. I call it the "There ought to be a law" reactions. Nobody stops long enough to say "no there shouldn't be".

I mean, from a horticultural perspective, there is some potential to gain more of other nutrients, in that if you have more energy, you can develop a larger root system, or generally more effectively, better feed mycorrhizal associations (fungal hyphae are much finer than root hairs, so can get into smaller cracks, and fungi can "acid mine" nutrients out of mineral grains - as an example, here's a microscopic image showing what they did to a garnet)

That said, yeah, in general if you can provide more energy, you expect the storage of "calories" to grow much faster than the acquisition of other minerals. Also, it's important to note that while more CO2 is generally good for most plants, more heat, or greater periods of drought (land dries out faster, monsoon belts spread) and flooding (atmosphere holds more moisture, monsoon belts spread) are not. In regards to heat as well, there's a lot of details. First off, though we commonly don't think about it, heat management in plants is critical. Their proteins are designed for function within an optimal temperature range, and to maintain it, they have to cool themselves down with transpiration, creating more water stress. Also it's worth noting that C3 plants (most plants) fundamentally don't tolerate heat as well as C4 or CAM plants (there's work to engineer C4 into some common agricultural crops... it's frankly amazing to me that they're getting some success, as it's not a trivial change).

BTW, the reason that plants grow better with more CO2 isn't what most people might think. The TL/DR is that the protein that sequesters CO2 so that (using ATP and NADPH from photosynthesis) - RuBisCo (the most abundant protein on Earth, something that has been evolving for billions of years) frankly sucks at its job. Something like 20-25% of the time (at normal CO2 levels), instead of binding with CO2, it binds with O2 instead ("photorespiration"), which means not only does it not sequester a carbon, but the plant has to *give up a carbon* to regenerate the RuBisCo. This is disastrous in terms of energy efficiency. And as a side effect, you also have to keep the stomata open more, which means more water loss. But as you increase the CO2 levels, the ratio between binding CO2 and binding O2 improves, and photorespiration waste drops. C4 plants "fix" this problem by instead of having RuBisCo directly bind CO2, they first bind CO2 into malate (with high selectivity), then the malate transports into bundle sheath cells, the CO2 is re-released, and THEN - in a high-CO2 environment - RuBisCo takes it up. This reduces photorespiration, but also introduces some more wasteful chemical conversions. (CAM plants to even further by storing malate inside vacuules - at the cost of even more energy - so that they can store it up during the night, and then use it during the day, which - although even more wasteful - lets them keep their stomata closed during the day to conserve water)

(BTW, there are some microbes that have developed a more efficient RuBisCo, but it's proving challenging to engineer it into higher plants)