Here's an idea: All graphical operating systems published by gatekeeper-class companies (as defined in the Digital Markets Act or foreign counterparts) would need to support, at minimum, some specific GUI API for developing local applications. For comparison, the US government used to require POSIX compliance. Microsoft delivered the bare minimum POSIX support in Windows NT versions 3.1 through 4.0, though initially not enough to be practically useful because it lacked networking and graphics.

How would one go about building the FOSS alternative to Apple Mobile Device Service, the component of iTunes that synchronizes music into the Music app of an iPhone? That's probably the biggest thing keeping my roommate on Windows. She wants to play purchased songs (ripped from a CD or purchased on Bandcamp or Itch.io) and rented songs (from her Apple Music family plan) in the same playlist. Because Apple Mobile Device Service is a driver, Wine doesn't run it properly. Last I checked, libimobiledevice for Linux could write files but not the music database used by the Music app, and VLC could play purchased songs stored as files but not rented songs.

Most Windows licenses included with laptop computers sold in major big-box electronics chains are not Pro.

Wine runs user-mode applications. It does not run kernel-mode drivers.

Such a law would have to be very carefully written so as to distinguish between "bots" and assistive technologies used by human beings with disabilities. I can't think of where to draw a bright line.

I guess it's the whole industry's "stupid choice" to sell products on Amazon then. When I worked for an online toy seller in 2010-2019, Amazon's marketplace service provided an Excel spreadsheet with macros used to perform local preflight validation before sending a product listing or inventory change file to the server. You'd put in product listing information, and it'd tell you what you would probably need to change before the server would accept the file. This was optional but highly recommended to reduce server-side processing errors at least until I wrote my own validation routines in Python. Excel could run the macros; LibreOffice Calc could not.

I disagree that phones and tablets are both good enough and affordable.

On their own, phones and tablets are not good enough because they run phone operating systems. These trade off flexibility for reliability, in part because people expect a phone to be their primary way of reaching emergency services (1-1-2, 9-1-1, 9-9-9, etc.). Phone operating systems are locked down with strict W^X in such a way that makes them not very capable, for example, for lightweight programming projects. Change my mind.

One can circumvent the lockdown by using the tablet as a remote terminal for running applications elsewhere. This requires adding a mobile data subscription and a virtual private server, or adding a mobile data subscription and a virtual private network (like Tailscale or Hamachi) to connect to your home computer behind an ISP-managed firewall. That adds a recurring fee in the tens of dollars per month or hundreds of dollars per year.

The practical problem with Chromebooks at the time was that for several years, between the debut of Chromebooks in 2011 and widespread support for Crostini (a GNU/Linux virtual machine) in 2019, a Chromebook couldn't do anything other than browse the web without threatening to wipe all your data (because "OS verification is OFF") every time you turned it on. That isn't very conducive to offline use while (say) riding a bus. I had been using my netbook for lightweight hobby programming projects.

The disadvantage of a bigger laptop is that a bigger laptop is less convenient to use in a cramped space, such as on a bus commute to and from your day job. It's also less convenient to pack in a cramped space, such as your tiny personal locker at your day job. A 10.1" laptop fit in (say) a locker in the back of a Walmart Supercenter, and a 11.6" laptop did not. That's part of why I was so disappointed that manufacturers suddenly discontinued 10.1" laptops at the end of 2012. I remember recommending that people affected by this discontinuation buy a cellular iPad, a Bluetooth keyboard case, a VPS, and an expensive data plan, and use the iPad to remote desktop to the VPS. I rejected that workaround as cost prohibitive at the time.

YES! Many happy users of Proton on Steam Deck will answer in the affirmative for many games. This is what Valve's 30 percent cut of Steam sales pays for.

As if Windows 11 doesn't have its own host of "config issues that require an hour of reading to fix."

One use case for a VPN is allowing incoming connections to a home PC that is behind an ISP's firewall and/or NAT. Such a PC might be running a game server or a low-traffic web server. Another is connecting to the IPv6 Internet from an IPv4-only home ISP such as Frontier.

The <figure> element might help with that.

I happened to be aware of the existence of a extension made by someone else that offers domain-level opt-in consent to run script in a particular web browser. I cited the extension's title and author and deliberately left out any URL. I thought that would have been adequate to imply lack of conflict of interest. A user has implied to me that it is not. What means of citing a source would have been adequate?

Most User-agent strings that don't contain "bot" nor "Mozilla/" were in Anubis's allowlist last I checked.

Yet. Because Anubis is free software, and because its hash happens to be the same as the proof of work of the cryptocurrency Bitcoin, someone could modify Anubis to tie the SHA-256 puzzle to the Bitcoin block that a mining pool is working on.

Until someone writes a browser extension to offload solving the hashcash to WebGPU.

Laptops tend to have an always-on CPU spike monitor: the exhaust fan. So do phones and tablets: they get warm. So do older, less expensive, or small-form-factor desktop computers: they get stuck on the interstitial for up to a minute.

This already exists. Use an extension to make script-in-the-browser opt-in per domain, such as the Firefox extension "Javascript Control" by Erwan Ameil.