peteforde

I don't think that's necessarily true, though. Claiming something has been written with an LLM is certainly peak 2026 Hacker News, though.

In order for what you're asserting as true to be so, someone would have to specifically prompt for an uncomfortable level of Mad Men style casual male grossness. I just don't see it, sorry.

I'm not suggesting for a moment that many/most men - or people, period - don't objectify everyone they see on a conscious or subconscious level. But very few people write it down for public consumption in the modern era.

Whether that comes from tact or just understanding the boundaries between inside thoughts and outside thoughts, I can't say. I'd have to ask an LLM what to think.

Wait, what are you saying here?

I understand why you think that this is a valid point in this context, but 95% of what's hard about creating a thing you can hold in your hands does not happen in KiCAD.

It might be 98%.

I don't need your permission to say words.

I'm really sorry to hear that you're going through this. Any company would be lucky to have you on their team.

My strong suggestion to you: pivot to hardware. I still use Ruby but most of my development efforts are ESP32-P4 and RP2350-based these days.

Learning CAD and mastering various additive and subtractive fabrication techniques, getting PCBs made to make ideas real... it's been a roller coaster but LLMs are remarkably good as engineering mentors these days.

And yes, as several others have already said, this might be the time to learn woodworking and/or video production skills.

You'll find that you still need your programming brain, instincts and intuitions. You just deploy them in a different way in support of creating a device instead of an experience [or yet another CRUD form].

Ryan is a big deal in the Ruby community. He's not imagining his accomplishments and not long ago it's reasonable to say there would have been several companies fighting for the opportunity to have him on their team.

I didn't call you an idiot. And yet...

Absolutely nothing you're saying is in good faith. You're just arguing for the sake of arguing.

So, no: you don't have my permission. You haven't earned it.

It's extremely obvious to everyone else here that they were not talking about literal enterprise software. Perhaps they could have chosen different terminology, but the fact that you've clearly latched on to this relatively minor word choice as some kind of smoking gun says infinitely more about your perspective and motivations than it does about whether some guy is building big-to-him projects or Enterprise Projects.

Tell me: when someone says that the shit hit the fan, do you furiously start looking around for actual shit?

It's been freaking amazing for me. I genuinely love it.

Nobody owes you evidence.

Ladies and gentlemen, the cuck chair is going to be occupied tonight!

Sometimes we make art.

Sometimes we have work to do.

I actually have a fairly simple answer to that: if it doesn't come up in the list of LLMs that Cursor supports, it effectively doesn't exist.

I'm well aware that there's nearly infinite opportunities to yak shave "perfect" OpenRouter setups and some people appear to enjoy bouncing from IDE to IDE as though change costs aren't a thing, but I discovered that I genuinely like Cursor and at least right now it's insanely subsidized by Auto clearly defaulting to whatever Grok's most powerful model is.

I dropped my $200/month subscription to $20/month and stick to Auto for all but really important Plan tasks, and I have basically zero chance of using up my monthly credits even using it 6-10 hours some days.

No, that's not what happened in this thread at all.

There's a recurring pattern on HN where any time someone talks about knocking dozens of personal projects off their list - things that almost certainly would never have actually been addressed in the finite span of a normal life, the way things go - and you AI doomers show up and demand receipts as though that's a total reasonable and definitely not obnoxious request.

It's like if you tell someone that you love your partner and they demand to sit in the cuck chair or else you're obviously lying. I keep hoping people will move past this "prove that you're actually productive" reflex, but it just keeps happening in basically every AI thread.

In reality there are many reasons not to list out projects that you've worked on with LLMs, and while "none of your damn business" is always going to be at the top, the simple truth is that I want my products and projects to be judged by what they do and how well they work, not by how they were made.

Sort of, but mostly not really...

On a small device like the M-VAVE, "sampling" usually means that you take (or record) an audio sample and perhaps if you're lucky, algorithmically pitch it up/down so that you can play it on the keyboard.

There's nothing wrong/bad about that, but when I talk about sampling, I'm talking about usually large libraries of professionally captured sounds. These are often the product of teams of specialists who rent a professional studio and use an array of high-end mics to capture every sound an instrument makes in an intimidating number of ways.

Indeed, they often capture multiple takes of each sound so that if you play them quickly, they don't sound like they are repeating.

Really what I'm getting at is the difference between recording a high-end snare or cymbal at Abbey Road with a fancy mic and putting white noise through an ADSR and a voltage controlled filter. Wildly different strokes for different folks.

I know this will not be a popular take, but surely I am not the only one who is super tired of endless control surfaces for the same FM and Karplus-Strong algos.

I'm a weirdo, but I love the sound of acoustic and/or sampled instruments - especially drums - over the metallic plonks that supposedly sound like "flute" any day of the week.

Don't get me wrong: the M-VAVE is looking like the Flipper Zero of music hacking, and the price is a steal. But at the end of the day, these sounds only really work for certain genres of music and even then there's only so many wavetable shapes before the same-ness has to set in.

Okay, I think I understand. More importantly, I want to say that I don't think you're trying to be jerky about this, though I was initially sort of triggered (admittedly) because you see this sort of sentiment a lot and it hits close to home.

So, the main thing that many/most people do not remotely understand is how incredibly difficult, time consuming and expensive it is to get hardware to the state where it can be sold to strangers. Even HN folks who on the average understand software well frequently just have no way to grasp how much R&D actually goes into even a very simple device. It's playing in hard mode from start to finish.

In a bigger company, there's entire departments whose job it is just to source components. For a small team like pre-CBA Chompi, all of the electronics, firmware, enclosure design, UX, documentation, not to mention the certifications and environmental certs and marketing and packaging and shipping logistics and returns and support and.... two people. I forgot lots of steps, but the number of people stays the same.

The reality is that a product like Chompi can't address a larger market like a product from Roland or Korg can, and everything costs more at smaller scale, and yes, they were doing a lot of the work in the USA so this is directly comparable to the ongoing arguments around categories like fast fashion and how if you wanted to buy a high quality T-shirt made in the USA without any branding it should cost about 5x more than most people think that they should have to pay.

Basically, you want to pay an Ikea price for a limited edition shelf made by a single woodworker. This translates quite literally to stating that you wish Chompi would give you a product that only breaks even at $600 when they sell thousands for $300 because you've convinced yourself that that's what a Chompi should be worth. It doesn't make you bad, it just means your working from a different set of expectations which don't line up with the economics of Chompi. The snarky response is to tell you to just go to Ikea like the other people who don't really care what they are buying but just need a shelf that isn't terrible.

Every year in May the Superbooth festival and trade show happens in Berlin. Almost everyone you meet there is either making or has made gear, so it's an incredibly cathartic opportunity to collectively process shared trauma.

Anyhow, I'm going to leave you with a strong suggestion that you check out this video: https://www.youtube.com/watch?v=2B-AmJUQ1m0

It's a masterclass in empathetic communication around this stuff.

I don't understand the second line/paragraph of your reply. Could you please clarify?

I'm not saying that you've clearly never built and shipped a hardware device as a small team in North America; maybe you have, and maybe that's what you do all day.

However, I have noticed that the only people who look at a product like Chompi and think "overpriced" are people who have never built and shipped a hardware device. Perhaps you'll be the one to prove me wrong.

I'm not saying that Chompi wasn't expensive, or that you couldn't get a mass-produced device like a Roland SP-404 for less on Reverb. I am saying that that is not a fair comparison, however.

I understand that there's always a hesitation to invest in what looks like an expensive and large tool, but if there's one thing that I could go back in time and force myself to get years sooner, it's a stereo microscope.

It's the single largest difference between frustrated and feeling like a ninja. You will, quite simply, never regret it.

I recommend this one: https://www.strangeparts.com/a-boy-and-his-microscope-a-love...

There are lots of pedals these days that are designed as performance instruments in their own right. Check out the Erica Synths Nightverb; it's all chonky knobs and it doesn't even have stomp switches. You'd never put it on your floor.

Storm Summoner takes a pedal with lots of parameters [that would normally be set-and-go] and gives it that missing live performance UX. Singers in particular could use the sensors to modulate their voices.

What excites me is the flag-based system described in the review video at 11:45. You can set up stuff to happen (launch an LFO?) in response to other things that happen, like a value crossing a threshold. That takes it out of the realm of a dumb controller, and we already have enough dumb controllers.

Austria: Federal Environment Agency / approved dual system. P €810 + W €1,110 = €1,920/yr. Battery: +€1,110.

Belgium: IVC/CIE; Fost Plus for household packaging / Valipac for commercial packaging. P €585 + W €585 = €1,170/yr. Battery: +€585.

Bulgaria: Executive Environment Agency / Ecopack or another approved system. P €375 + W €990 = €1,365/yr. Battery: +€1,195.

Croatia: FZOEU / RPPO. P €780 + W €1,315 = €2,095/yr. Battery: +€1,315.

Cyprus: Green Dot Cyprus or applicable national system. P €375 + W €910 = €1,285/yr. Battery: +€910.

Czechia: Ministry of Environment / EKO-KOM. P €385 + W €1,110 = €1,495/yr. Battery: +€910.

Denmark: Danish Producer Responsibility / approved PRO. P €1,110 + W €925 = €2,035/yr. Battery: +€730.

Estonia: Pakendiregister / approved PRO. P €385 + W €775 = €1,160/yr. Battery: +€775.

Finland: Producer Responsibility Register / Rinki. P €810 + W €925 = €1,735/yr. Battery: +€925.

France: ADEME/SYDEREP; normally Citeo or Leko. P €585 + W €985 = €1,570/yr. Battery: +€790.

Germany: LUCID/ZSVR plus a dual-system contract; WEEE: Stiftung EAR. P €360 + W €1,185 = €1,545/yr. Battery: +€705. Germany illustrates why the service price and actual environmental fee should not be confused: LUCID registration itself is free and one current small-volume packaging example is only about €39–€41/year if handled directly.

Greece: EOAN/EMPA / HERRCO. P €790 + W €985 = €1,775/yr. Battery: +€910.

Hungary: national registration / MOHU. P €1,120 + W €1,185 = €2,305/yr. Battery: +€985.

Ireland: EPA / Repak as applicable. P €585 + W €1,170 = €1,755/yr. Battery: +€1,170.

Italy: CONAI and relevant consortium. P €805 + W €1,305 = €2,110/yr. Battery: nominally +€1,305, although this provider offers a WEEE+battery bundle.

Latvia: State Environmental Service / approved recovery system. P €1,290 + W €1,365 = €2,655/yr. Battery: +€1,375.

Lithuania: GPAIS / approved PRO. P €780 + W €1,365 = €2,145/yr. Battery: +€1,375.

Luxembourg: Valorlux. P €375 + W €1,485 = €1,860/yr. Battery: +€1,500.

Malta: ERA / GreenPak or Green MT. P €780 + W €675 = €1,455/yr. Battery: +€675.

Netherlands: Verpact; electronics commonly through Stichting OPEN/Wecycle. P €385 + W €385 = €770/yr. Battery: +€385.

Poland — Packaging/WEEE registration through BDO and appropriate recovery organisation. P €910 + W €1,110 = €2,020/yr. Battery: +€985.

Portugal: APA/SILiamb / Sociedade Ponto Verde or another scheme. P €1,195 + W €1,390 = €2,585/yr. Battery: +€1,390.

Romania: AFM / approved OIREP. P €880 + W €1,195 = €2,075/yr. Battery: +€1,195.

Slovakia: Ministry/ISOH / ENVI-PAK, NATUR-PACK or equivalent. P €750 + W €1,120 = €1,870/yr. Battery: +€1,125.

Slovenia: ARSO / approved PRO. P €1,485 + W €1,615 = €3,100/yr. Battery: +€1,630.

Spain: MITECO / Ecoembes. P €790 + W €985 = €1,775/yr. Battery: +€985. This provider also quotes €595 one-time for obtaining the Spanish NIF where required, plus official/apostille costs.

Sweden: Naturvårdsverket / NPA. P €810 + W €1,110 = €1,920/yr. Battery: +€505.

Using the published list of concierge prices above, covering all 27 EU countries for a batteryless electronic product would cost approximately:

Packaging service: €20,290/year WEEE service: €29,260/year Combined: €49,550/year

Adding battery compliance using the same provider's à-la-carte figures adds approximately €27,445/year.

What is the opposite of something putting a spring in your step?

I'm not part of the Musk fan club either, you can count on that.

However, I've been using Cursor since it came out. I have a massive amount of institutional knowledge locked into their platform, and vastly prefer it to the other options available even if the switching costs were zero.

Going through what would amount to significant effort/time/cost to switch to a different coding tool as a sort of performative political rebuke is just not how I would recommend anyone protest DOGE.

I don't know because I honestly don't know or really understand how/why anyone needs what you're describing.

I am not saying that you're wrong, just very aware that we appear to use LLMs in radically different ways.

Only the most substantial requests run for ten minutes or more, and I would spend an hour or more writing the prompt for that action.

I am always the bottleneck for how quickly Cursor can do what I want, with the level of outcome that I get.

I would genuinely love to be a fly on the wall while you do what you do, because it just doesn't currently make sense to me.

A few months ago folks were understandably annoyed when Microsoft dropped their heavily subsidized per-request pricing model because it was figuratively burning cash.

Well, I'm here to tell you that whatever is going on behind the scenes at Cursor with this Space-X acquisition in the works, the Auto setting is clearly routing all prompts through "Cursor Grok 4.6 High" right now.

This is a degree of subsidy that makes the Microsoft thing look quaint.

I reduced my $200/month subscription to the $20/month level and have proceeded to do what I would have paid about $1500 to do with Opus 4.7 or thereabouts, which is how Grok 4.6 High feels like it compares. I don't have anything remotely like hard evidence to back this estimate up beyond what I'm watching it do and I still somehow have ~10% of my monthly Auto capacity left on my account. It's completely nuts.

Can't say much more because I have more backlog to run before someone comes to their senses.

This is peak HN. It's a criminal shame that nobody else is seeing it.

I feel like you're getting hung up in semantics such that your essay doesn't say the obvious: coding is "easy" relative to confidently knowing how and what to do next at every stage.

The comparison does not imply that coding is an easy thing to do, just that it's easier than being really, really good at the bigger picture.

What Carmack did wasn't hard because writing C is hard.

Let's say you're in the market for an $1800 frequency analyzer such as https://siglentna.com/spectrum-analyzers/ssa3000x-plus/

Well, if you want to use your new tool to analyze some frequencies, you need to start buying activation codes to unlock core functionality.

There are at least five different things your machine can do which need to be unlocked for USD$380-650.

https://siglentna.com/product/emi-activation-license-ssa3000...

Don't worry, though: each unit comes with 128 hours of free preview for these premium features. If the analyzer is powered on, the timer is running. Leave it on when you head home for the day? Whoopsie!

This offends me on so many levels.

That's why we're here. Even the lame posts on here have survived vastly more scrutiny by smart people than >99% of what is published online.

Maybe I'm an outlier, but I find that my main reason for discontinuing Ground News/Nebula/whatever is that I subscribe aspirationally and never actually make time to consume the content. One year I got my partner and I subscriptions to Masterclass and at the end of the year neither one of us had watched a single video.

Basically, I don't know anyone who is ever bored. Who the hell has time to click on and read random posts? The entire value of HN is that other smart people surface the top 1% of 1% of stuff I'd probably find interesting.

The notion of reading Substack posts just because I'm a subscriber seems like fantasy.