Can anyone comment on the seemingly impossible transition from quantum computers on the order of ~150 qubits available today to systems with the 500,000 physical qubits required to break classical encryption?

Seems like quite the handwave, especially with a target as close as 2030.

Ofcom doesn’t really wanna block websites though, they want websites to either comply or block themselves, both of which legitimizing Ofcom’s extraterritorial enforcement.

Reddit isn’t trying to protect user content here. They’re suing to make sure they’re the only ones who can monetize it.

Not sure how you’d reach the conclusion that it would harm SerpApi’s case. They’re the ones being sued.

Is SerpApi asking each user for permission to use their posts if they are saying that the rights of the posts belong to the user?

Copyright protects copying. Scraping content does not violate copyright if the content is not republished. Otherwise Google and all search engines would be illegal.

Ads weren’t that much of a problem when they were contextual. I remember video game websites younger me used to visit having their background plastered with latest release by a AAA studio. This is contextual advertisement. It has no privacy concern.

The issue is that ads now are behavioural, privacy invasive and centralized. No matter what site you visit you’ll get unrelated, possibly scam, advertising that depends on a profile built by a large American corporation. It’s just not reasonable in this context to avoid using an ad blocker.

So yes, the problem is indeed Google (and Meta etc) who monopolized the advertising market. I would say the root cause is lack of antitrust enforcement.

Interestingly, Vitest now supports visual regression testing

Came here to point that out too. Looks super useful. There are entire platforms built around just this kind of feature.

I have come to the same conclusion as the OP: people like tailwind not for its API or ergonomics but for its design system.

Tailwind can actually be decent for templating systems and it’s likely why it’s shipped with Phoenix for example.

But for React components, it’s definitely a step back from CSS-in-JS, which gave you style encapsulation out of the box. Not sure what’s up with the knee-jerk reaction against it in the comments. With vanilla-extract (mentioned in the OP) or PandaCSS, there is no runtime at all.

Just components that defined their structure and style, in a readable way, with less abstraction than Tailwind.

I just use tailwind inside components. This problem is because of no components + tailwindcss, not because of tailwind usage alone. Nothing is stopping one from using components that use tailwind.

"Nefarious" is indeed a great pick.

As a native French speaker, my favourites are those words that capture concepts better than their French equivalent.

Something as simple as "mind" for example. In French, you're stuck with "esprit" which also means "spirit" or "facultés mentales" which is quite annoying to use in conversation.

That’s been my experience as well. It’s especially jarring when asking for a refactor as it will leave a bunch of WIP-style comments highlighting the difference with the previous approach.

Did you miss the DX of React Server Components at all when switching to Inertia?

I haven't so far. I’ve preferred the new way:

With RSCs, I’d have code scattered throughout components that would hit APIs to get data.

With Rails + Inertia, that code is all in a Rails controller - usually with a serializer.

I’m still getting used to using InertiaRails.optional there which is what’s needed for a Suspense like experience w/Inertia. Having everything in controllers again is nice.

Seems like it still won’t replace Sidebery for me as this provides only a single grouping layer.

To replace Sidebery Firefox would need to clone like half of Arc Browser.

Arc is dead, long live Arc. Best we can hope for while The Browser Company is toying with their investor-bait "AI browser" is for other browsers and extensions to copy the good parts of it.