Yeah, for example a bunch of my system updates began showing scary error notes because somehow there is a header inconsistency between the amdgpu driver and the kernel.

I'm not regretting my choice, but it's also something where the average user can't just call Linux Support and get a "run X and it'll fix it" solution.

One can call Windows support? And get help?

Arguably there are more support options for Windows because it's got fewer derivatives than Linux, and was historically more common on desktop.

I'd prefer the receiving end looks at sender's metadata on the message, and uses that to determine where the line is between recipient-convenience and betrayal.

I suppose you could do both, but "Hey I've got something extra important to send you, but it says need to change your settings first please hurry" seems worse than "sometimes I don't get full notifications on my watch, weird."

Whenever I see something about a single blood sample, I think of the Theranos fraud.

I suppose in this case it's a false alarm, because they're talking about a single type of test (that "consumes" the sample) with a rich result set, rather than many different kinds of tests that each require pristine inputs.

I suppose one could view it as a supply-chain compromise of an alternate chain that's very short.

Back in the 90s I fantasized about a hard drive bay with a physical write-protect switch on the cover plate.

In the mid-2000's I briefly worked for a company that did this at a firmware level ("write-blocked firmware") for USB drive adapters (IDE / SATA / whatever IDE variant laptops were using / etc). This was apparently very valuable for police and investigative services, so they could collect evidence, while being able to show that they did not tamper with the original drive.

Tenable makes some "read only" adapters for hard disks (SATA, PATA, SCSI & FW at least). They're usually sold as part of a forensic analysis kit. I have a couple and they definitely work. I believe there are a couple of other vendors (Wiebetech?) make similar devices.

The alternative (tho not practical in many cases) would be RO media like RW-DVD.

Perhaps this from last year, though it doesn't directly involve code-signing: https://www.neowin.net/news/microsoft-bans-libreoffice-devel...

> Dose rate matters. Particle type matters. Direction matters. Shielding matters.

There's an old story where a professor quizzes his physics class about how to most-safely distribute different kinds of radiation sources. A common variation involves three baked cookies, emitting alpha particles, beta, and gamma respectively. One must be eaten, one must be held in your hand, and one must be placed in a pocket.

A hint, and what I think is the interesting part of the answer, involves the idea that a victim is a lot like shielding. Things which are difficult to block are also things that are less-likely to stop and ruin your day.

I heard about this question in the MIT Ionising Radiation course. You can find it here, around minute 47:30 (wont spoil the answer).

https://ocw.mit.edu/courses/22-01-introduction-to-nuclear-en...

Thanks, I watched it. There are 4 cookies - a gamma, alpha, beta and neutrino emitter. You must eat one, pocket one, hold one and give one away.

Are we protecting the person with the cookies, or everyone else? I'm struggling to think of the best answer

Alpha: In the hand held away from the body would be reasonably safe. In a pocket of a lab coat might provide a little more shielding from the body with the coat material, but it is physically closer. Eaten would be very bad for the user, but protect the outside world the best

Beta: Medium penetration, would likely not be safe in any of these three situations

Gamma: High penetration, definitely not safe in any of these situations, best would be to get it as far away from you as possible, so held at arms length would mean you might only get high radiation exposure in your hand. Hospital visit is probably needed in any of these three situations

Eat the gamma cookie, it's the worst option for any, but with gamma it won't matter much whether it's in your hand, pocket or gut.

Pocket the beta emitter, a little bit of shielding will make a big difference.

Hold the alpha emitter, if you hold it with just a pair of fingers you will be able to reduce you dosage a lot compared to holding it tightly.

wouldn't alpha be mostly stopped just in pocket ? I'd hold beta emitter, less area of body affected by the radiation if you hold it at distance

IIRC:

* Alpha in hand. Eaten, the "shielding" that blocks it will actually be very active living cells, leading to severe health outcomes. Your external layers of dead skin cells will be be fine, putting it in your pocket would be excessive.

* Gamma in hand, because whether it's in your hand or in your pocket, it's roughly the same risk, and most of it is actually going through you without stopping to have an effect. (Compared to other two.)

* Beta in pocket, where the additional clothing layer(s) offer some meaningful protection compared to your hand.

The "twist" behind the exercise involves how people often assume penetrative power is proportional to danger, when in some ways it's really the opposite. (Consider the danger profile of neutrinos.)

Oh, I thought you have to choose one of the options for each cookie, no repeats

I guess in that case it would be to eat gamma? Assuming you're keeping all three long-term, the gamma particles will be washing over you whether they're inside you or out.

You're quite correct, the light of the morning and flu-medicine what I should've written is:

* Gamma is [eaten], because whether [wherever it is], it's roughly the same risk, and most of it is actually going through you without stopping to have an effect. (Compared to other two.)

Naturally there's a whole bunch of unstated "all else being equal" going on, where no cookie's' elements are extra-likely to be permanently incorporated into your bones versus excreted, etc.

Same with x-rays. People tend to think “soft” X-rays are safer because they are quickly absorbed by tissue without passing through.

The radiation that passes through is not the problem.

Sure, using the ambiguous wording to assume the cookies all have such a large activity they are unsafe no matter how they are stored is technically a solution. But it's equally possible the teacher is imagining the cookies as only slightly radioactive such that it is indeed possibly to safely store them according to the alternatives given if you choose the correct pairs.

Can these cookies be weaponized? Are there secret cookie enrichment factories? Would weapons-grade cookies still be good with milk?

Only weapons-grade milk.

Heavy milk (whole) is required for cookie enrichment. But depleted milk (skim) still has value.

Commonly used as a projectile

Take with iodine.

Eat the gamma, pocket the beta, hold the alpha. Nothing in the picture here stops gamma rays so it doesn't matter what you do with those; skin will stop alpha particles, but so will mucosa (and you die of that); in your pocket, the beta source will spall X-rays off your trousers at least some of the time so the beta burns will be mitigated.

> in between rows full of numbers, the text suddenly changes

To tweak the analogy slightly, the person would also need to be on mind-altering drugs, if we want them to be derailed the same way an LLM can be.

A healthy human would still be aware of the simultaneous different ways of interpreting the data, and and the importance of picking the right one. If they choose to interpret it as a cry for help, they're aware it's an interruption and mode-switch from what was happening before.

In contrast, with LLMs we haven't built thinking machines as much as dreaming ones. Your dream-self recovered the poster that was stuck on the elephant's tusk, oh look that's a pirate recruitment poster, now you're on a ship but can't raise the anchor because...

> A healthy human would still be aware of the simultaneous different ways of interpreting the data, and and the importance of picking the right one. If they choose to interpret it as a cry for help, they're aware it's an interruption and mode-switch from what was happening before.

So would an LLM, as far as you can tell (in both cases, you'd have to ask, and both human and LLM would give you a similar justification). But even if not, the problem we're discussing applies to what you described as "healthy human" behavior.

You can't introduce a hard boundary between "system" and "user" inputs in LLMs any more than you could do with a human, for roughly the same reasons.

The problem isn't so much overuse as misuse, as "gaslighting" gets thrown around for almost any kind of falsehood.

Another example would be "Ponzi scheme", which I've seen abused for any situation the speakers seems unsustainable, even when there isn't any records fraud.

TBF that's a little bit apples-to-orchards, since publicly routed e-mails have certain expectable size/frequency characteristics compared to, say, all the videos someone possesses.