SwtCyber

A wrong model is not the same thing as fraud, and biology is full of plausible models that later turn out to be incomplete or misleading

It's not just that one paper may have been wrong; it's that a fragile consensus can redirect billions of dollars and decades of work. Alzheimer's is also a brutally complex disease, so the opportunity cost of narrowing the search too early may have been enormous

True, court cases are one of the few times details actually surface

You don't, at least not without forensics tools

Pretty sure that's just iOS behavior + app design. If notifications are off, apps will occasionally prompt again to make sure you didn't disable them by accident or miss something

Originally enabled it just to avoid awkward moments

This is one of those cases where the "secure app" narrative collides with how messy real systems actually are

Whatever system you design (market, regulated, hybrid) it still ends up creating incentives that shape behavior

Most real-world work sits somewhere in the middle

I'm not convinced "token efficiency" maps cleanly to "good code". You can have very compact code that's extremely hard to understand or modify

Studio for desks, something rackable for scale

But I think the Mac Pro was never really trying to be on your desk in the first place. For a lot of its target users, it lived under the desk or in a rack, and the size wasn't about aesthetics so much as airflow, expansion, and serviceability

The interesting question is whether they'll lean into it intentionally (better tooling, more ML-focused APIs) or just keep treating it as a side effect of their silicon design

It totally makes sense. Mac Studio basically ate the Mac Pro's lunch. But it's still kind of sad. The Mac Pro used to represent this idea that Apple cared about the absolute high-end, no-compromises workstation crowd

Ten years ago a lot of this would have required expensive proprietary systems or a lot of custom engineering

This feels like one of those cases where DIY ends up being simpler than waiting for the perfect product to exist

There's a difference between automation for reliability and automation for tinkering

This is a pretty common experience with Home Assistant

"Find a problem for Home Assistant" is a dangerous starting point. Next thing you know you've built a small distributed system to water parsley.

E-ink feels weirdly well-suited to generative art

There's something very satisfying about solving the whole problem with a couple of scripts instead of building a full framework around it

Makes me wonder why there isn't a whole category of "public display-friendly" e-ink devices with first-class APIs...

IKEA moment is a perfect real-world demo of how much the outer geometry and boundary conditions matter

I suspect the connection isn't "paper roof holding weight" so much as "geometry that can turn thin sheet material into a stiff, self-supporting form"

I've noticed that learning something like advanced math as an adult feels slower, but the understanding tends to be more intentional

The age makes for a catchy headline, but the real signal is the time investment and sustained curiosity

Pretty impressive work for 14 and more importantly, a great example of curiosity turning into real engineering thinking

I do think clearer norms would help

The bigger issue is whether users feel they have clear, informed control over what's collected and who can access it

Home security devices sit in an incredibly sensitive place. If users feel like the scope of data use is drifting beyond what they originally agreed to, that's a big deal