Hacker Newsnew | past | comments | ask | show | jobs | submit | DowsingSpoon's commentslogin

I don't think it's strange at all that the scientific framework of today is capable of investigating physical phenomena in the real world. It would be strange if it _couldn't_.


PsyOps, you say? Operations by which organizations? Which specific individuals were involved in the planning and coordination? How are they funded? Who got paid and who paid who for this PsyOp work? To what end?

Who approached Margaret Atwood, for example, and proposed she write a novel as part of a desensitization “PsyOp?” When?


If Handmaid’s Tale was supposed to be desensitization, she failed.


If you’re looking at the AI-generated output then you’re not Vibe Coding. Period. Let’s not dilute and destroy the term just as it’s beginning to become a useful label.


Wait, are people not reading the AI code they use?


People of course often do read (and even modify) the model-generated code, but doing so is specifically not “vibe coding” according to the original definition, which was not meant to encompass “any programming with an LLM” but something much more specific: https://simonwillison.net/2025/Mar/19/vibe-coding/


The whole point of the term is to convey that you are only looking at the output and not looking under the hood. If vibe coding a UI, you only look at the UI, not the CSS.


Nope. That's the "vibe" part of Vibe Coding™.

> The developer does not review or edit the code, but solely uses tools and execution results to evaluate it and asks the LLM for improvements. Unlike traditional AI-assisted coding or pair programming, the human developer avoids examination of the code, accepts AI-suggested completions without human review, and focuses more on iterative experimentation than code correctness or structure.

https://en.wikipedia.org/wiki/Vibe_coding


RISC vs CISC is nonsense.

Pre-RISC CPU designs were pragmatic responses to the design constraints of their time. (expensive memory, poor compilers)

RISC was a pragmatic response to the design constraints of its time. (memory becomes less expensive, transistor budgets are tight, and compilers are a little better)

Post-RISC designs of today are, also, only pragmatic responses to the design constraints of today. Those constraints are different than they were in the 80s and 90s.

The supposed dichotomy is just utter horseshit. It was invented as a marketing campaign to sell CPUs. It was canonized by the most popular text books being written by /Team RISC/.

I wish, as an industry, we’d just get over it, move on, and stop talking about it so much.


I had been planning to explore Lima tonight as a mechanism to shackle CC on macOS.

The trouble with sandbox-exec is that it’s control over network access is not fine grain enough, and I found its file system controls insufficient.

Also, I recently had some bad experiences which lead me to believe the tool MUST be run with strict CPU and memory resource limits, which is tricky on macOS.


Wait, does lima do isolation in a macos context too?

It looks like linux vms, which apple's container-cli (among others) covers at a basic level.

I'd like apple to start providing macOS images that weren't the whole OS.. unless sandbox-exec/libsandbox have affordance for something close enough?

You can basically ask claude/chatgpt to write its jail (dockerfile) and then run that via `container` without installing anything on macos outside the container it builds (IIRC). Even the container-cli will use a container to build your container..


Neat, I've not tried https://github.com/lima-vm/lima


Just two days ago, I asked Claude Code (running as a restricted non-admin user) to generate a unit test. I didn’t look too closely at exactly what it wrote before it ran it for me. Unbounded memory use locked the system up so hard it stopped responding to all user input. After a few minutes, the machine restarted automatically. Oof.


How so? In this context, “tedious” clearly means “not very exciting for the reader.” Were you hoping for more relationship drama, or romance? I’d be down for that. Though, it doesn’t seem to be what Banks was most interested in writing about.


It is often remarked, by those who may be assumed to have some insight into it, that the difficulty in writing Superman is to convey his perfection and goodness, but that for people who do not understand Superman well the difficulty is that he is too perfect and unbeatable and thus boring.

I put forward thus that in the same way Superman can be tedious so can Utopias, and Utopias can be interesting in the same way that Superman can.


> Were you hoping for more relationship drama, or romance? I’d be down for that

It’s tedious to a modern audience. Hemingway but in space isn’t really an improvement. (The closest we have is Ted Chiang.)


While I've fortunately never had this happen to me, I'd be tempted to say something like, "Wow. Well, I sure hope you don't get fired over this. Good luck. We'll scope it out and let you know how much time we'll need."


Having been on on the customer side it's frustrating how often the situation is: Me: "So, you got a bid which offers features A, B, C, and D we asked for, and you say it also has X and Y and hit our budget?" / Buyer: "Yes".

A week later. "OK, their install team says it can't technically do C yet, however there's an early 2026 preview scheduled which addresses most of C. The D feature isn't in the edition we have, our buyers are talking to their sales people and we may need to pay extra to unlock D. And you're correct that two other organisations in our industry confirm X is dogshit and you'd be better off without it but it can't be disabled. Still A does work, and we have filed bugs about the known defects with B so hopefully we can get those fixed"

Every time I buy a product as an ordinary consumer I marvel at how much worse my huge employer is at buying products than I am. I reckon if they were sent to the store to buy a whole roast chicken with a £20 note they'd come back with six expired chicken sandwiches and no change.


> Every time I buy a product as an ordinary consumer I marvel at how much worse my huge employer is at buying products than I am.

It's the size of the deal that matters. Most of the consumer goods you buy are sold on a take-it-or-leave-it basis. No individual sale is worth the vendor forming a "relationship" with that customer or promising bespoke features. B2B sales are often large deals that require months of negotiation and may be worth millions. Bullshitting in order to land the deal is incentivized on both sides, to the point where both only have a fuzzy idea of what exactly is being bought and sold.

But consumers get this experience as well when they make larger purchases. When I buy a car, maybe I fail to mention the unreported fender bender my trade-in was in, and maybe the salesman tries to charge me $1200 to etch "anti-theft tracking numbers" on the new car's windows, citing some dubious statistics about vehicle recovery rates.


> consumers get this experience as well when they make larger purchases

Or as I like to do, buying random things on AliExpress and Temu knowing full well that some of the things will not meet the expectations you’d have from the product listings.

Sometimes I’m lucky and the stuff is good. Sometimes I’m a little unlucky and it’s worse quality than I’d like.

At least I quickly learned to read carefully what was said to realize that what’s depicted is not exactly what’s being sold. Some sellers do this misleading trick where they have some amazing photo up front but there are either multiple variations of it or the thing being sold is only some component for that thing. I still sometimes see product reviews from other buyers that were upset that they didn’t get what they thought they were buying and I don’t blame them because it can be pretty misleading at times, but if you read carefully and look at all the pictures and check what the “color” or similar option dropdown says etc you will usually spot it when they are selling something different than what it might look like at first. So I haven’t had that kind of misfortune for years now. But sometimes you still get products that are lower quality than you were hoping for, even when the product listing was pretty accurate. Some kinds of bad quality is just not possible to judge unless you see the product in person.


Maybe they exist but I haven't worked in a company yet that wouldn't fire an engineer or manager for refusing to implement a feature that some salescritter already sold. One of them made the company money (on paper, sure) while the other is threatening to undo the deal. It's not hard to guess which one the c-suites would send packing first.


oh you agree to do it but you laugh, literally laugh, at their deadline. and you say, you can fire me but that's not going to get your software done on time. in fact it will delay it.

they shut up. it's done when it's done.

I've done this many, many times. Oh you promised it by the end of the week and didn't ask me? lol, that sounds like a YOU problem.


The Whitehouse once called my team at Microsoft and asked for some features.

We said yes, we'd get right on it. :-D

We were all too stunned to have any real feedback.


Fortunately it doesn't happen too often, and some can be attributed to our somewhat complex feature matrix that differs by regions due to reasons.

On the other hand, in our niche customers usually don't swap software providers often due to integration work needed.

When an opportunity arises, it's usually because the yearly license expires. So we got to either sell it now with a hard deadline in the near future, or wait 5+ years till next time they switch.

So that can lead to sales being a bit optimistic when making the pitch.


I've been on both ends of this workflow. Sales always wins.

"Wow. Well, I sure hope you don't get fired over this. Good luck. We'll scope it out and let you know how much time we'll need."

"We'll see."

The big-screen TV in the modern glass conference room showed the final slide: “Questions?”.

"I.. I'd like to add that this feature we sold is not in the product and we can't just go around adding features that Sales makes up out of the blue just... just to close a deal. I mean, we gotta plan these things, there's a procedure, we should get product involved..."

Head of Sales, interrupting: "Can't we, Jeff?"

Jeff, the middle-manager, shuffled his feet: "Uh. Yeah. Right. I think we shouldn't. Hey! Haste makes waste, that's what they say, right?"

Head of Sales: "Can't we Barbara?"

Barbara, the boss: "I don't know. Let me call Pradeep"

(Barbara presses the "huddle" button in Slack on her big iPhone. A few rings and a bored voice replies)

"Yeah?"

"Sorry to jump on you like this, Pradeep. Would you mind coming to meeting room seven for a second?"

Less than a minute later Pradeep walks in, his thick glasses casting a green hue over his eyes, his arrogant demeanor preceded him like a shadow.

"Pradeep, did you read the feature request I messaged you?"

"Yes."

"How fast can you do it"

"Just merged it this morning."


That’s a bad analogy. It works against your point. It seems entirely, entirely reasonable to avoid a clothing store that refuses to stock hip styles simply because they’re “offensive”. For an example using the cliche, many people find a naked ankle to be completely acceptable.


Yours is wrong. Yours would work if you liked all the clothes there but then one day they stopped selling Kanye West’s Nazi shirt that you liked because people found it offensive. And then you stopped shopping there because of that.

The analogy requires you giving up what you want because they stopped carrying something others find offensive. Not that they don’t sell what you like all up. In which case it makes perfect sense not to shop there.


It’s highly dependent on what I’m viewing, of course. Some people just… talk… so… slooowly. If the content is interesting then they get 1.5x. Some people talk a mile a minute with dense information and they get 1x. For myself, 2x is usually too fast to be enjoyable.


This is the problem: the settings should let you specify an absolute rate like average words or phonemes per minute.

Maybe some coefficient of word/thought complexity.

Not a relative multiple of the base rate. Garbage in, garbage out and all.


My experience has been that it's not just the cadence of the speaker, it can be their intonation and pitch that determines whether I can speed it up. And some deep voiced person from Essex who is recording in their bathroom ... not a prayer, because I can't find the word boundaries from the background noise


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: