lambdaone

This is very cool work. Nice to see the adults in charge somewhere, planning for managing and mitigating anticipatable disasters that haven't happened yet. Much better than disaster mitigation after the fact.

It's pretty difficult to operate a CT scanner with pen and paper, to name just one thing that fell over yesterday. CT scanners are life-critical.

Absolutely. The risk being managed is the risk to the CEO/CTO's jobs, not the risk to life.

An awful lots of apparently useless bureaucracy exists because many people, left to themselves, are often very, very stupid.

Bureaucracy certainly stops smart people from doing the right thing, but more often, it stops stupid people from doing the wrong thing. Hack away at bureaucracy at your peril.

Given this: https://www.theregister.com/2024/07/19/crowdstrike_update_nh... someone has certainly been selling Crowdstrike software to be used on life-critical systems.

I had to get someone life-critical medicine yesterday. My GP practice's computers were down because, presumably, of Crowdstrike. Manual pen-and-paper processes saved the day.

I wonder how many people didn't get so lucky?

They're silly, creepy, harmless (and of course fictional) fun with a side-order of existential dread. What else might they be?

Exactly. It's like the brown M&Ms in the Van Halen rider; it's not that the M&M's were the problem, but that it was a test of diligence. People who don't care about detail are likely to screw up the big things just as badly as they screw up the little things.

Being a multi-million dollar company and using unsafe languages today is not a good look. But everyone gets away with it because everyone else is doing it.

Good. Microsoft doing something sensible.

From that point of view, there are several million amateurs, some of them quite highly paid, out there writing terrible code.

Yes. The term "normalization of deviance" comes to mind. Even just a phased rollout would have caught this one with just a tiny fraction of the damage observed.

You can indeed write buggy/unsafe code in any language. But it's a lot easier to do in notoriously unsafe languages like C/C++, which for some maniacal reason we seem to have based the world's digital infrastructure on.

C++ was a terrible, terrible mistake.

And human beings can indeed write safe "unsafe" code. But to do so consistently, you have to be very smart, very cautious, and somewhat lucky.

Again, there are safe ways of doing this. For example, Wuffs exists: https://github.com/google/wuffs

At the very least, big money security software companies should be parsing untrusted content with some kind of rigorouly safe approach, not just squirting it through a big pile of C/C++.

And don't get me started on the whole concept of undefined behavior in those languages. To quote I. I. Rabi, "Who ordered that?"

Memory-safe lanuages (for goodness' sake, even the crap I write in Python qualifies!) are the very minimum that is needed; not to use them for anything critical is simply crazy. Yes, do all the other things, but at least put out the blazing fire in your basement while you are implementing your fire-safety strategy.

I didn't know that. So that makes this two strikes?

It's the equivalent of not writing about Boeing until the day a 737 MAX crashes right in front of your newpaper offices.

You're making the mistake of assuming that the people running those companies care about anything other than their job security, and buying in solutions is the best way to have a ready-made scapegoat when things go wrong. The mantra "no-one ever got sacked for buying IBM" still holds, you can just substitute "Oracle", or "Microsoft", or now - apparently - "Crowdstrike".

This has been an open secret for decades. Just a handful of major OS and browser vendors, constantly shipping patches to their systems and most software having such vast software supply chains that it's effectively impossible to audit anything, let alone truly certify anything as safe, and "security" software just expands the attack surface.

Everyone in the industry knows this.

Interesting to see the NYT just catching up.

That's all fine, until one day Crowdstrike is the threat.

I think it would take Crowdstrike doing five, six of these in a row before they tanked. Look at Boeing; two planes full of passengers died for quite scandalous reasons, followed up by a whole bunch of other engineering scandals, and while the value of Boeing shares took a 50% hit, the company's still in business, and people are still buying planes from them.

For me, the word "smart" is a red flag. It generally means "unnecessary feature that's designed to screw you".

I'd be surprised if they were toast. Big companies can fuck up with impunity - just look at Boeing. They can take a bit of an insurance hit, chalk it up to the cost of doing business, and pass the cost back to their customers in the long run.

Most software in indeed trash. There's neither the budget nor the will to fix it. The existence of "security" software is a symptom of systemic sickness, not the underlying disease.

It is certainly possible to write bug-free code, in terms of meeting a formal specification of behavior, and guaranteeing no behavior outside that specification. It requires formal methods, and it's much more expensive than ordinary software development.

Creating exploit-free code is another matter - you have to be able to craft exploit-free specifications, and there's no real understanding what that might even mean. But bug-free software would be a start.

There certainly are systematic solutions to at least _some_ of the real problems, based on formal methods, of which Rust's checkers are a simple example, and it's an inexcusable moral failure that they are not used more widely.

Reliable software is expensive to develop; more expensive than we are willing to admit. You can either take the expense at development time by increasing development cost, or you can shift it until later and externalise the cost by making the end-user and society at large suck it up in terms of unreliability and occasional disaster.

We are at the stage the Victorians were with exploding boilers, or people in early Medieval times were with collapsing cathedrals. They fixed their problems, and so can we.

It's a counsel of despair. I stopped reading when I read the words "As a non-scientist, I do my own kind of modelling."

Given what existing models can do right now, that moment might come sooner than you think.

It is, indeed, just a tool like any other. And just like any other tool, like a gun or a knife or a pepper spray, having one does not give you the right to use it on other people.

Your right to swing your fist stops at my nose.

The parent poster is right in that pure fructose is probably worse for you than an equivalent amount of pure sucrose. But the way that fructose is delivered to your body by digesting fruit - slowly, and in conjuction with other materials, in a way that we have evolved to tolerate - is entirely different from eating the pure chemical.