plutokras

What if Hacker News had topical sub-HNs so people could post things that aren't Hacker or News? (:

I was skipping Tahoe too after seeing all those posts about the differently rounded corners.

This time I took the plunge and upgraded, and I have to admit the whole system, especially Safari is much more snappy than before.

AFAIK Brave Search is also using its own index.

> It feels quite slow

It definitely is. The extension-based redirect in Safari is also unreliable. Furthermore, many of my searches (and others', apparently) unexplainably defaulted to Groningen in the Netherlands, even though I'm nowhere near the place.

I ended up not renewing my annual subscription. Since then, I vibecoded a small userscript that replicates domain ranking for DuckDuckGo – plus DDG already lets you turn off ads natively.

I remember them advertising Svelte as the framework that disappears in build time.

The SimpleUI plugin helps a lot with the clunkiness.

> Eventually, this arms race ends with a computer vision model that looks at the screen, classifies visual elements as ads, and draws a rectangle over anything that looks like an ad.

That would be a killer feature for AR glasses. Imagine no billboards, no political posters, no neon storefront clutter.

This feels like obvious setup for regulatory capture. It won't be long before missing "safety" watermarks are cited as the pretext for restricting Chinese models.

What specific technical signals make you think we're close to a shift like that?

> My friend didn’t have the vocabulary to ask the AI to write code

I had the exact same realization using Claude Design. It’s great that "anyone can design" now, but I don't have the domain knowledge to describe what a good design consists of. I know it when I see it, but that is far from enough.

With the recent wave of posts about Claude rewriting entire codebases from Zig/Electron to Rust/Swift, I can't help but question the prompter's actual confidence in the final result. Gaining a thorough understanding of the generated output still demands an enormous amount of effort. I suspect the reality is that most people just chain a few review agents together, and if they output "LGTM", the code gets committed and pushed.

I have been doing daily agentic coding myself since last November, but lately Im having concerns. We are on the verge of becoming Warhammer 40k Tech-Priests: chanting sacred prompts to appease the machine spirit, praying for a good result. The foundational knowledge is lost, meaning that even if we do look at the code, we won't understand it anyway.