Linux already has dm-integrity, a layer that emulates a block device with per-sector tags that store integrity information. If you insert it between a block device and any filesystem, you get the same result, including a read error if the data cannot be corrected.

I don't see an explanation of why you would need a whole new filesystem.

I had this exact problem. I was alone in my car, headbanging to Samael’s Rebellion on a mountain road I knew well. I was going a bit too fast down a long straight stretch with a hairpin at the end, but I had plenty of time to slow down.

I saw the turn immediately but dismissed it. I was so taken by the music that I felt invincible. It was only halfway through that I registered the danger and slammed on the brakes. I knew there was no way I was going to make the turn, but thankfully I managed to cut my speed roughly in half, to about 50 km/h, before leaving the road and hitting a tree. I walked away without a scratch, thanks to very solid German engineering.

I haven’t listened to music while driving since. I also keep a window cracked open now, just to have a little reminder that I’m sitting in a car, not a tank!

Unfortunately, refusing to speculate and waiting for the release of the final report, way after the accident has been forgotten, doesn't sell ads or subscriptions.

Exactly. Even if you did steal something once, this is scary. You are essentially barred from buying food, ever.

This distinction doesn't exist in everyday language. However, that is how these terms are defined in mathematics. See Sphere and Ball.

5 POPS = 5000 TOPS.

I am still waiting for my Ford Nucleon.

There are safety-critical languages with a whole ecosystem behind them, like SPARK. It would take an enormous effort to bring Rust to that point, and for what benefit ? I'd rather see existing tools be improved.

You are wrong. Satellites do not transmit the P code directly, but the Y code, which is an encrypted version of the P code. Advanced military receivers can lock on the Y code directly. They don't need the C/A code and you cannot spoof the Y code, thus they are not affected by spoofing. Advanced civilian receivers cannot lock on the Y code, but if they can lock on the C/A code, then they can use it even without decryption. Thus they are affected by spoofing of the C/A code. Regular civilian receivers, like those found in cell phones, cannot make use of the Y code.

That has already been done: https://www.nbcnews.com/id/wbn...

That is what happens when you use a PDP-11 assembler in 2022. There is no shortage of languages that perform automatic index checking (e.g. Ada, C#, Java, Python, Rust, ...). And yet, billions in damage are incurred every year with those bugs and nothing changes. To quote Hoare:

Sure, Tannenbaum for wanting to keep Minix small, Stallman for GCC and Thompson & Ritchie whose work was cloned.

Or rather, Uniswap and co. The ridiculous fees of Ethereum don't make these things very practical yet. What I find annoying in the media coverage (including here) is the sole emphasis on the potential value of coins or tokens as investment. The important thing is the technology and as you pointed out it has considerably improved since bitcoin appeared. It still has a long way to go. What's most interesting is that we don't yet know how to build a really good blockchain-based network, nor do we know which applications (and they go far beyond financial services) will succeed, with many yet to be invented.

It's too specialized to be repurposed.

The capsule's centre of gravity is deliberatley off-axis. These are counterweights that put it back on the vertical axis. This is needed from launch all the way to separation of the cruise stage, after which they are jettisoned. Thus the capsule enters at an angle, and generates lift. It is therefore aimed a bit short. If you look at the EDL animation you'll see nitrogen thrusters on the back shell that adjust the orientation of the capsule during entry. If the trajectory is short, the capsule pitches up thus increasing lift. If it is long, it pitches down thus decreasing lift. Cross-range errors can also be corrected to some extent. All this significantly reduces the size of the landing ellipse, and allows the rover to drive to its target in a reasonable amount of time.