chuckadams

As the old saying goes, comedy is when someone else walks into the open manhole.

Most successful bank heists are in and out in less than five minutes. Or were anyway, there's a reason you don't hear about them these days.

> I think the HN/web forum ethos of "a comment should stand on its own" has been a grave mistake, as this strips the surrounding context from what someone is saying.

I get what you mean about quoting something out of context, but I always read the intent of that guideline as the opposite: that a reply should quote enough of the original to provide the context. Really only needed near the top of long/deep threads where the reply might be multiple screens below the nested replies above it. Summarizing also works, which always risks misinterpretation, but it's also good to get that interpretation out in the open (again assuming good faith).

One thing I still can't stand reading is the usenet-era "point-for-point" style that interleaves a sentence or two of quote alternating with response, back and forth for the whole post. It's not only noisy, it tends toward a very snippy tone as well.

Huh, that makes more sense, thanks. Whodathunk that The Simpsons wasn't an authority on translation? Let down again... :-/

I'm old enough to remember candy cigarettes.

Schadenfreude literally means "shameful joy", so it's kind of built into the word itself. Someone who doesn't feel at least a little tinge of wrongness is someone who's broken.

Hanna Arendt coined the best term of all for it: The Banality of Evil.

No who remembers WAP has any desire to see anything like it return. A 'lite' subdomain is far easier and more effective by not trying to reinvent the entire stack.

As for WML, I'd say transport-level compression made that pretty much academic.

The "But" and the quotes suggest that the GP reply might be sarcastic?

Poe's Law in action I guess.

"It's life, Jim, but not as we know it."

On Apple devices it's L4/Darwin, aka "sepOS".

Agreed. There was frequently a better answer that was ranked lower, but the top one was rarely actually wrong. SO had a litany of failings that many to this day are still denying, but that wasn't one of them.

"We're not answering your question because it has words another question has, you're wrong for saying it doesn't answer your question, and you're wrong for thinking we were wrong."

That's a bit clearer. Bit long to engrave on a tombstone though, but SO will remain unburied because after all, most coffins are pretty similar.

I like what Discourse does: it checks for similarity in your subject line (and can extend it to the body) and shows you a list of similar posts next to it. Whereas SO's unspoken motto that it lived and very much died by was always "gotcha": for askers, answerers, and moderators alike. Those meta-moderation traps were real, and it quickly got to where I just didn't bother doing it anymore. Selecting moderators by whether they agreed with the existing moderators was just another ingredient in the poisonous cocktail that SO mixed up itself and gulped down every day.

> What the hate-leaders tell us to think about the mullahs and Iran, we could easily have thought about our own nations just a half a lifetime ago...

Or where they'd still like to take us. The only problem they have with Ayatollahs is that they call God by the wrong name.

Yep, Florida ordering you to remove every one of your cameras from state highways didn't happen. Entire states moving to ban your product isn't a thing. And of course there are no tanks in Baghdad.

Unless your name is Raymond Chen, you probably want to at least use ATL/MFC, which are frozen but still supported. Me, I'd take Qt any day, or wxWidgets if I wanted to go minimal.

It's definitely a better language than Matlab, and Numpy made it actually fast for that purpose. AI research really took advantage of that even before LLMs hit the scene.

I remember falling in love with Python when I first learned it at version 1.5, liked it even more with 2.0, but when 3.0 came around, that's when I noticed the ecosystem falling apart. Not for the obvious incompatibility schism: I do think most of 3.0's changes were good ones, and it finally is the standard default version now, but in many other ways it feels like the ecosystem has still never really recovered. Virtualenv is still the same grotesque hack, where rather than doing what other languages like Perl and Ruby do and making some wrapper around setting PYTHONPATH, it relies on python's behavior of locating the library path relative to the install, thus hardlinking a whole python installation in and making you "activate" it first. That nothing has managed to replace it in 20 years speaks volumes.

It's really uncommon now, but it's still possible. I hit it two days ago with deepseek 4.1 flash, and I've had coworkers report that happening with Gemini.

My own anecdote: I have seen frontier models get stuck repeating the word "the" until it fills up the context. I really don't care to argue the capabilities of LLMs or their somehow simultaneous lack thereof; I'll just reiterate that ultimately the operator chose to ship a cheat.

It was certainly assisted by AI, but there isn't an agent out there that does that kind of cheating without deliberate prompting by the operator.

I've taken a vibe-coded project pretty far, with even better testing than most other things: the main factor is that it's big on falsifiability, i.e. it has to see a test for a new feature actually fail before it considers the test any good. Not just the red/green from being originally unimplemented, it does a mutation test to ensure it flips back to red. But wow did it ever make that test framework complicated, byzantine, and ultimately brittle until I stepped in and forced a consolidation and simplification pass (twice now). Part of the complexity came from it being a polyglot monorepo, but I've found even in other places, you have to follow an expand-contract cycle of development when vibe-coding. Not all that different from any other system you've built up organically, just on a much faster cycle.

But back to my original point, none of this really applies to the pgrust benchmark, which was obvious and blatant cheating on the human's part.

Of course, I didn't mean to imply it took zero effort to copy one, though I guess that's how it read. You'd have to find a skilled jeweler that's also willing to commit treason, so for sure the barrier wasn't insignificant there.

The signets were "unforgeable" in that the penalty for copying one was death by being tortured, hanged, drawn and quartered.

That sounds pretty similar to what Valgrind does, basically runs your program in a VM with that kind of instrumentation. The details of how it works are way past the boundaries of my own knowledge though :)

Sanitizers come with a lot of overhead, and the tech wasn't readily available in SimCity's day, certainly not for Windows 95.

A sanitizer is that runtime debug check. The compiler just inserts those checks.

C was adequate for its era, though it had major problems from day one: like array pointer decay, lack of counted strings, aliasing just about anywhere and everywhere, and a legacy of squishy types it inherited from BCPL which was about as well-typed as Forth. Even so, it did yeomans work, but it's now long past its use-by date.

I still want C to burn in a fire, but I suppose it's at least earned an honorable Viking funeral (yeah I know that's ahistorical but you get the idea).

Why can't simpler programs be written in languages that at least make an attempt to be memory-safe?

> C is not to blame for programmers using its features unwisely

"Make no mistakes" works about as well for humans as it does for AI, and we have decades of lived experiences to draw on for that verdict.