ErigmolCt

I know that feeling. Some interests never really go away, they just stop being the thing that pays the bills

That's a neat form of accidental crowdsourcing. Instead of trying to predict where something interesting might be, you let people's travel plans generate the sampling distribution for you.

I think "origins of major parts of life" is still a bit too broad. It's more like the origin of one major cellular innovation that later became essential to plants and algae

The actual science is interesting enough that it doesn't really need the "origins of life" framing

The Paulinella part is much more interesting to me than the motel-room angle. We effectively have an independent, comparatively recent case of primary endosymbiosis, which means researchers can study molecular details that were completely erased by 1.5 billion years of evolution in plastids

I get the joke, but this is kind of why the security issue matters

Exactly. The scary part isn't any single bug, it's the mismatch between the sensitivity of the system and the apparent security model

And this seems designed around the assumption that the physical camera can be trusted indefinitely

Apps, browsers and "smart" features are disposable software, but a good panel can remain useful for a decade. Makes you wish manufacturers treated the smart layer as optional from the start.

I'm inclined to believe the shutdowns were a bug rather than an extremely patient friend waiting for the perfect gameplay moment. But modern TVs having enough remote-management machinery that the prank theory isn't immediately absurd is not exactly reassuring.

At that point network isolation stops looking like a privacy preference and starts looking like an adversarial relationship with a product you own

I wouldn't sell yourself short on the last one... the comment itself is a decent example of it :)

Maybe the failure mode isn't the PhD itself, but environments where publishing becomes the objective rather than a byproduct of getting unusually good at understanding and building things

You probably aren't building the next Airbus in a dorm room, but you might build a weird propulsion prototype, simulation tool, control system, manufacturing technique that eventually becomes the reason a company exists

I think side projects still matter, but maybe in a slightly different way now. They don't necessarily have to become the company themselves. They can be how you discover the problem, test an approach, meet the people you eventually build the company with

The irony is that universities are probably best at producing founders when they're not explicitly trying to produce founders. Teach difficult things, put smart people together, give them access to equipment and enough free time to build weird stuff. The hard part institutionally is that "we gave students fewer assignments and left them alone" is much harder to sell than an Innovation Center with a nice sign on the building...

I suspect rare here often means out of print or commercially obscure, not unique

Copyright holders are certainly responsible for keeping unavailable works inaccessible, but shredding is mostly an industrial scanning decision, not a copyright requirement

A genuinely useful project would identify publications that are actually rare, poorly catalogued, or held by only a few libraries, then prioritize those for careful preservation

Bluesky trick exposes how strange the abstraction is

Once apps can present one thing to the user and another thing to the captured image, there are some pretty weird possibilities

Agreed… yet I meant "needs" in terms of product features, not maintenance or moderation

Yeah, 2009-2014 feels more like the old social web to me

2009 is old web now, and this is how we find out we're old... Yet I think maybe "middle-aged web" is the more accurate term

HN is a good example of how little a website actually needs once you remove the requirement to constantly squeeze more engagement out of it. Though to be fair, its audience is unusually self-selecting :)

I think the audience could absolutely shrink back to something resembling the old web. I'm less sure the old web itself comes back

The interesting number here isn't 76.7% dead, it's that 23.3% alive is an upper bound. A server answering HTTP doesn't mean the thing somebody shared in 2011 still exists

I think the revenue number needs a lot more context for exactly this reason

And I'm writing my comment to understand it better...

The really haunting part is that Europeans often arrived late enough to see the collapse, but not early enough to understand what had collapsed