tensility

This.

Oh, it would be nice if it were just feature size. Over the prior 15 years, the nand industry has doubled its logical density three times over with the trick of encoding more than one bit per physical voltage well, making the error bounds on leaking wells tighter and tighter and amplifying the bit rot impact, in number of ECC corrections consumed, per leaked voltage well.

At the physical device layer (i.e. what a nand controller vendor programs to), a nand flash device that supports ONFI 2 SLC Mode can be configured to use (some of) its blocks in SLC mode rather than MLC/TLC/QLC/etc. This allows one to divide the array into high-reliability versus high-capacity regions.

There are also expensive high-grade SLC NAND flash chips available that offer significantly higher retention time than the cheaper commodity channel TLC NAND (i.e. 10 years versus 3 years).

In general, though, whether NAND or NOR, the fundamental way that flash works is by creating an isolated voltage charge for each bit (or several bits, for TLC), making it effectively a vast grid of very tiny batteries.

Like all batteries, no matter how well-stored, they will eventually leak energy to the point where the voltage levels change enough to matter.

Further, it's not enough to simply make power available to this grid since the refresh of the cells requires active management by a nand controller chip and its associated software stack.

Read off all live data on the drive. This should cause the nand management firmware to detect degrading cells via ECC and move the data in order to refresh isolated cell voltage levels.

Good advice; however, past experience suggests that conventional magnetic hard drives suffer problems of stiction when left in cold storage for too long. I wouldn't trust either technology for long-term archival purposes.

Doubles as the sock outfit for the next Red Hot Chili Peppers tour.

Meanwhile, can I have multi-message selection back in (iPad) Mail? Whoever decided to axe that feature apparently has a spam-free inbox.

This company has become such a joke. Maybe Apple should start being concerned about building computers that Just Work well again rather than continuing to flounder after Cook's obsession with bad fashion.

I suppose the underlying message here is that, if you can no longer innovate, shill overpriced purses instead.

More accurate to call it the iPhone Purse.

I'm far more fond of knights than kings. So, yes, I would much more happily train another wave of humans at my craft than salt the earth behind me.

For whatever it's worth, this is exactly the kind of awful that I never want in any code base that I'm working on:

<https://github.com/simonw/tools/blob/473e89edfebc27781b43443...>

At least run a pretty-printer on the code so that it can be reviewed by anything but a robot.

Part of developing a good software system is about exercising taste in vendored-in libraries and, especially, the structure around them.

P.S. I've gone to look at other chunks of Javascript and see that I was unlucky to grab this steaming pile first.

Actually, my criticism was the result of my own misunderstanding of what you were claiming. My apologies for that, although I'm still unlikely to use these tools based upon the example when my own personal counterexamples have shown me that it's often as much or more work to get there via prompting than it is to simply do the thinking myself. Have a good day.

I'm surprised nobody else has gone meta yet, so I suppose I must. Anyway, "ChatGPT said this" ... about this thread.

----

In many of the Hacker News comments, a core complaint was not just that AI is sometimes used lazily, but that LLM outputs are fundamentally unreliable—that they generate confidently stated nonsense (hallucinations, bullshit in the Frankfurtian philosophical sense: speech unconcerned with truth).

Here’s a more explicitly framed summary of that sentiment:

⸻

Central Critique: AI as a Bullshit Generator

Many commenters argue that: • LLMs don’t “know” things—they generate plausible language based on patterns, not truth. • Therefore, any use of them without rigorous verification is inherently flawed. • Even when they produce correct answers, users can’t trust them without external confirmation, which defeats many of the supposed productivity gains. • Some assert that AI output should be treated not as knowledge but as an unreliable guess-machine.

Examples of the underlying sentiment: • “LLMs produce bullshit that looks authoritative, and people post it without doing the work to separate truth from hallucination.” • “It costs almost nothing to generate plausible nonsense now, and that cheapness is actively polluting technical discourse.” • “‘I asked ChatGPT’ is not a disclaimer; it’s an admission that you didn’t verify anything.”

⸻

Philosophical framing (which commenters alluded to)

A few participants referenced Harry Frankfurt’s definition of bullshit: • The bullshitter’s goal isn’t to lie (which requires knowledge of the truth), but simply to produce something that sounds right. • Many commenters argue LLMs embody this: they’re indifferent to truth, tailored to maximize coherence, authority, and user satisfaction.

This wasn’t a side issue—it was a core rejection of uncritical AI use.

⸻

So to clarify: the strong anti-AI sentiment isn’t just about laziness.

It’s about: • Epistemic corruption: degrading the reliability of discourse. • False confidence: turning uncertainty into authoritative prose. • Pollution of knowledge spaces: burying truth under fluent fabrication.

It's called a "scam". You're welcome.

"But the comments that quote it directly are almost always trash."

Because the output is almost always trash, and it takes re-doing the work and then claiming that it came from LLMs for it not to be.

These tools are being sold to me as similar to what a Jr engineer offers me, but that's not at all true because I would fire a Jr that came to me with such bullshit and needing such significant hand-holding so often as what I see coming out of an LLM.

Further, grep (and any of its similar siblings) works just fine for such a task, is deterministic, won't feed you bullshit, and doesn't charge you tokens to do a worse job than existing free tools will do well. Better yet, from my experience with the dithering pace of LLMs, you'll get your answer quicker, too.

Why do you believe it's disingenuous when a capable person tells you that they don't need the crutches that you are advertising? We've accomplished millennia of engineering without the need for a bullshit generator, but you somehow assume everyone needs such a tool just because you are such a fanboy for it? Talk about disingenuous...

That tendency to hallucinate that you so conveniently downplay is a major problem. I'll take reading the reference manual myself all day rather than sifting through the output of a bullshit generator.

There's a very good argument for not using tools vended by folks who habitually lie as much as the AI vendors (and their tools). I don't want their fingers anywhere in my engineering org, quite honestly. Given their ethics around intellectual property in general, I must assume that my company's IP is being stolen every time a junior engineer lazily uses one of these tools.

"This tool uses the WebAssembly build of Perl running actual SLOCCount algorithms from licquia/sloccount."

The best form of (AI) plagiarism is to simply wrap the original tool in your own facade and pretend like you built anything of value.

Is this intended to be a bad joke?

Not at all true, though. You see, I expect the Jr will grow and learn from those off-loaded tasks in such a way that they will eventually become another Sr in the atelier. That development of the society of engineers is precisely what I do not wish to ever outsource to some oligarch's rental fleet of bullshit machines.

It's also a good reason to not believe what's being said given the still extremely high rate of hallucinations.

As far as I understand, many users are better off with GPT-4o anyway. Amusing to be charging premiums for an objectively bad upgrade, but I guess that's the kind of bullshit economics that hype cycles create.

[flagged]

The winning feature of LoC as a metric is that it can be "eye-balled" easily without tools. Cyclomatic complexity is a more interesting metric; however, it's not the easy, quick, and rather dirty spitball that Lines of Code is.

Oh, boy! Byzantine generals in the traffic lights... Wonderful.

If your Scheme were a native compiler rather than an interpreter (i.e. like Chez) and written in (a constrained subset of) Scheme, then you could use an itty-bitty Scheme interpreter written like this to bootstrap the core of your native Scheme compiler so that the interpreted version of it could compile itself to native binary.

Given that this was the path that one of Dr. Dybvig's post-doc acolytes, Dr. Michael Ashley, taught us in his Scheme-based compiler class, I must guess that this was also Kent's path and that that was the intent for this little interpreter. I suppose I should (finally) take the time to read his dissertation.

A lie is far worse than a gap. I'd much rather see someone simply be honest about the fact that they've taken a year off to focus on their personal health due to a serious medical condition. Further, if the lie is uncovered at any point, trust is likely broken forever. This is awful advice, in my opinion, and I certainly don't want to hire anyone who agrees with it ... ever.

I sincerely hope that the attempt to obscure your identity in the resume wan't intended to be serious, A.C. The fact that you left all your former employers' names other than the current one made you a trivial single-hit search on LinkedIn, for whatever it's worth. If you were serious with your attempt at operational privacy, tighten up your information hiding because you're an easy match right now. If you weren't serious, I'd suggest removing the obfuscation altogether. Best of luck.

Oddly, I thought that the irony was that this was released at the same time as Anthropic changing Claude's terms of service to eliminate user privacy. So much for Apple being a privacy-first company. I guess it's okay as long as they are just piping their developers' code to a third party, eh?