You could say that about credit reports too. They're just someone's opinion of how credit worthy you are.

Practically speaking, that ship has sailed. We've already collectively decided if someone is collecting data and making a decisions which can have profound effects on someone, that someone has a legal right to see what data went into that decision. However, the law as written may or may not apply to AI because AI wasn't a thing at the time. If it doesn't, IMHO, that's a job for Congress or the California legislature to address. I'd prefer they explicitly added AI and other algorithms to the legislation rather than having a court invent a legal right.

Bear in mind, I am not a lawyer and I haven't read the relevant legislation and case law. I have absolutely no idea how a court will decide. I doubt many of us here do either. Case law especially is very complicated so be cautious with what you think you know.

I didn't read TFA, I just started some back of the envelope math. I was wondering what you build this thing out of and what material has a tensile strength to weigh ratio to hold anything at 2,000 g's. Turns out carbon fiber can easily do it.

Assuming I have a 1m high strength carbon fiber rod with a cross section of 1 cm^2, that's 100 cm^3. According to the Googles, carbon fiber has a density of about 1.6 g/cm^3 so the whole rod has a mass of 160g.

Also according to the Googles, an ultra-strong version of this rod will hold up something like 70,000 kg (!). Even allowing for a hefty safety margin, the rod can hold over 40,000 times it's own weight, well above what this new centrifuge needs.

That's crazy. I knew carbon fiber was strong and light but had no idea it was that strong.

2,000 g's is definitely a ton of gravity.

In the consumer space, I think the vast majority of systems are laptops with less than a TB of local storage. The number of systems which had multi-TB is a small slice of the market, probably just gamers and media creators. That's why I'm surprised: I wouldn't think there were enough of them to buy 150 million units.

Maybe I shouldn't be. How many professional photographers and videographers can there be? Tens of millions? Probably not 100 million.

Right. But how many casual creators keep that locally versus beaming it to a cloud? Not the ones I know, which is admittedly a small slice.

That's the thing. You personally might have an enormous home storage footprint. I'd be hesitant to extrapolate from that. People who hang out in /. and other nerd forums tend to be weird outliers in many ways.

I beg to differ. In the enterprise space, "large" arrays are measured in petabytes and they all use SSDs now. I don't think you can buy a primary storage array which still supports HDD. Even the small enterprise arrays, tens to hundreds of TB, are exclusively SSDs.

That was popular 15 to 5 years ago. Last I looked, SSDs and NVMe devices had gotten large and cheap enough to entirely replace HDDs except for limited applications. Well, maybe not "entirely" but for huge swaths of the market. At least, that was my impression. Given the sales numbers, I wonder if I'm wrong.

All that's in the enterprise/data center space. I have very little insight into the SOHO market and what happens on deskside systems. That was a different part of the company.

That's the thing. My impression was the class of applications where a SSD wasn't appropriate had shrunk to just backup and archive storage.

Right. I was differentiating between trains and subways/light rail/street cars. I don't know if that distinction makes much of a difference and I don't know if that's what the OP had in mind.

In the US, the two mean quite different things and are used for very different purposes. I think that's true world wide but don't know for sure.

Not really. What you want is fast (for streaming writes) and cheap per byte. What you can buy is slow and cheap or fast and expensive. No one actually wants slow devices, that's just what they settle for.

And it turns out with very clever software and careful hardware design, you can make many slow disks pretty fast.

I used to work for an enterprise storage company making secondary storage devices (that is, destinations for backups). High capacity and low cost were the selling points (and write performance and reliability, but those don't figure in here).

For the last 15 years, the shift in the enterprise storage industry has been to SSDs for everything, just like for home use. Around the coffee machine, we talked about how secondary storage was the last industry segment which still used spinning rust. Last I checked, even that division was selling purely SDD based systems.

I checked some HDD market stats. Apparently something like 150 million units shipped last year, compared to a peak of around 600 million in 2010. I'm surprised there's even that much of a market for HDDs left.

No doubt. Back in the day, the HDD market was notorious for booms, busts, and profit margins thinner than the oxide layer on the platters.

Supply has also got to be tight. Manufacturers are going to tightly plan production to keep costs in line, thus making it hard to rapidly ramp up production. In those circumstances, even a small shift in demand can lead to large jumps in price.

Trains may work for city to city transit. They're not so great getting people from point A to point B in an urban center. Even light rail isn't great in terms of coverage. It's really a backbone which needs surface transit to fill in the last few miles.

I haven't been to Japan but every city I've been in with a subway also has busses and/or taxis.

I'd be happy with just converting from full sized busses to twice as many 15-seat passenger vans. Driver salary may make that cost prohibitive today. If we could make busses autonomous, then we'd be on to something.

As you say, you could just make this autonomous Uber Pool: when you whistle up a bus, you put in your destination and some clever route planning software could group riders. One could even imagine a range of vehicle sizes, from single seat to 45, depending on demand.

I guess I'd have to see a demo. Every time I've tried it, Linux GUIs look amateurish compared to macOS and Windows.

Thing is, from the summaries, Linux ain't there yet. The summary of the summaries was "yeah, then I had to do this annoying workaround which would flummox any non-techie and everything was great!" Do hear echoes of Jerry Pournelle and Chaos Manor?

  "Looks worse and harder to use" isn't a great sales pitch.

I wanted to start BadgeDash. You give your badge to some guy, he drives across town with a box of other badges, and swipes everyone in.

No one else seemed to think this was a good idea. Don't know why.

My last three or four managers have worked on a different coast from me.

Sheesh. I think the last time I had a manager in the same building was..2018, something like that?