It’s more like some cryptocurrency scammers tried to bribe him to promote their coin and he took the money and refused to stay bribed so the coin tanked. As it would have eventually, because it was a pump and dump.
Why should the scammers who gave him the money get it back? They knew what they were doing, even if Yegge seemed a bit naive about it.
> he took the money and refused to stay bribed so the coin tanked.
I don’t think he refuse to stay bribed. I think he did what was asked and they executed a rug pull. He is extraordinarily honest and flippant about it. [0]
> And with that disclaimer out of the way, I must reiterate my sincere regrets to the CT/BAGS crowd, who so generously funded me to the tune of just shy of $300k last week on bags.fm. That money was hard to duck, and the funds are deeply appreciated. They will help Gas Town be a big success this year. But Gas Town itself needs my full attention; between that and Beads it’s a wonder I get anything done at all.
> So I had to step back from the community. I do find it amazing how they band together, dissenting voices rolling around like a big Katamari Damacy ball, and yet they somehow collectively find the discipline to act like financial analysts for institutional investors, weighing developer dossiers, product business cases, and doing critiques like a collective of professionals. All in crypto-bro speak. But it’s the same due diligence.
> But the CT community, like any highly engaged stakeholders, were going to be asking for a lot of my time. There are always strings attached.
from my admittedly brief research into it, $GAS was sending him transaction fees, as it desired the association. So it wasn't him selling a tangible number of coins to a particular person. So I figure to pay it back, he'd have to trace down the owners of every tx and pay them back the tx fee. Perhaps that's easy, perhaps its non-trivial. According to the tweet[0] he made yesterday, in order to donate the combined funds to charity he has to submit 230 separate transactions on his phone.
Cult 101: Building a massive reality distortion field as a kind of team sport always means forgiving any/all crimes from thought-leaders. Bonus points if you make the crimes look like a virtue
The way he presents this situation as being foisted upon him against his will in that tweet comes across very different than his original disclosure where he made it sound like a positive windfall.
I'm not accusing him of doing anything wrong as he didn't originate the coin, but his original disclosure messaging on the situation was pretty horrible which is why it harmed his reputation.
This is a sincerely dishonest take to try and avoid responsibility for participating in something deeply unethical and scammy. If this were about evidence in a criminal trial, I would start opining about a poisoned tree.
The sad thing is that it often works.
Just look at how people view Andrew Carnegie now. After his reputation was sullied by his company’s behavior in the Homestead Strike, his philanthropy was done, in-part, to try and restore his reputation.
> But for God's sake, don't accuse me of pumping it.
They paid him to use his product's brand name on an obvious crypto rug pull, he agreed, promoted it on his blog, then people lost money.
His "apology" would be more effective without including all the whining about accurately describing the sequence of events he voluntarily participated it for personal gain.
How does this work? Someone names a coin after his project, pays him commission for every transaction so he will be incentivised to give it positive coverage on his blog, then sells their stake once the value inflates leaving bagholders with worthless coinage and Yegge with his commission?
Not to get pedantic, but rugpulling is still not strictly speaking “stealing”, because everyone is playing according to the protocol’s rules. Someone controlling a large share of pool tokens can decide to liquidate them and extract the other asset in the pair. Market participants are aware of this risk and who is controlling what share of the tokens at any moment.
This is not to excuse the scummy behavior. I didn’t know Steve Yegge went this low just to make a quick buck, and I lost almost all respect for him.
A respectable software provider should warn you about this kind of behaviour at install time, and give you the opportunity to opt out. Gas Town fulfilled all its obligations in this regard with these (and other) warnings in the original announcement:
> A respectable software provider should warn you about this kind of behaviour at install time, and give you the opportunity to opt out
They honestly only need to disclose. Requiring contribution as part of the social contract is perfectly okay—if someone disagrees, they don’t get to use Gas Town.
That's not coming soon, that is a thing that was happening on compromised servers years ago (and probably still, but to a lesser extent given the decline in popularity of meme coin launches)
A few years ago, if you visit a site, your laptop grinds to a halt and the fan starts spinning like crazy, you know there is crypto mining happening on the site.
(btw that was a really good showcase for WebAssembly. Too bad it's used for illegitimate purposes)
This probably doesn't help, but I noticed that my everyday computer experience got much better after switching to an M4 Mac. The only time I heard fan was when I was running ffmpeg. Unfortunately that appears to be the solution that works.
That was some time ago. According to Yegge, Gas Town is now stable and ready for everyday use.
> Gas Town “just works.” It does its job, it has tons of integration points, and it has been stable for many weeks. People are using it to build real stuff.
> So as far as I’m concerned, Gas Town is ready. That’s why I feel it merits a 1.0.0 release.
I was trying to use Gas Town heavily only 3 weeks ago, and while it's fascinating, it's also very much still the bleeding edge.
The neat part though, is agents are so interwoven through its operations, it can kind of power through almost any error. It's a strange-but-real form of resilience.
Is anyone surprised? I'm reminded of how I felt during the NFT craze. LLMs are extremely powerful when used with deliberate care. Gas Town is the exact opposite of what is needed to actually do useful things in prod. I guess good on Steve for doing what he does so well, and getting so much hype around a vibe coded mess.
this is why transparency matters with anything that touches cloud AI. if your routing user prompts through any API, users should know exactly whats being sent, where it goes, and whether it gets stored or used for training. Burying that in terms of service isnt good enough
I’m reminded of the Carl Rogers therapy app that was developed in the 80s.
People would type in their problems and how they were feeling. The application had very very simple logic that would follow up with a set series of statements or questions. Things like “that sounds tough” and “how does that make you feel?”.
People reported great satisfaction, even if they knew that the application had no smarts behind it. Because of course the whole time the magic of therapy lies in verbalizing your problems, with very little actively done by the therapist.
Now you can pay an LLM subscription for a service that likely produces worse results since it is tuned to be aggressively (and insidiously) sycophantic.
Digital goods- like digital rights for movies, games, etc. only none of the big players would ever give up their walled gardens/licenses instead of ownership for content etc.
> I guess good on Steve for doing what he does so well, and getting so much hype around a vibe coded mess.
Shit coin aside, I don't get the hate for Gastown, we all know its theoretically plausible and he's giving it a shot. We get value either way, either we learn its not just theory or we get to watch it burn in the flames of a legal/financial/security/maintenance nightmare for its practitioners.
Many people are trying to make this thing, this is the one we can all see. I'd rather have the visible one remain visible because it gives us a useful data point and/or entertainment.
Because he should know better? Because it’s obviously a shit show but he keeps on being very vocal about his shit show? Because it’s annoying to have to see yet another delusional vibe coded project being hyped up instead of this forum being used to discuss actually industry relevant information?
he's doing it in the open. Its instructive for us all either way.
> he keeps on being very vocal about his shit show?
I'm not really sure what this complaint is. You want someone doing something to not.... write a blog about it?
> Because it’s annoying to have to see yet another delusional vibe coded project being hyped up instead of this forum being used to discuss actually industry relevant information?
I think I've seen around 2 posts, one the original gastown one and then the gascity one. Is two posts in like a year too much or do I miss a midday rush where the front page is all Yegge?
> Because it’s annoying to have to see yet another delusional vibe coded project being hyped up instead of this forum being used to discuss actually industry relevant information?
It's industry-relevant. This is what the industry is now. All in two short years.
> are extremely powerful when used with deliberate care. Gas Town is the exact opposite of what is needed to actually do useful things in prod.
This judgement feels premature.
We really don't have any idea what is possible. The wheels within wheels, epicycles within epicycles model of agentic loops hasn't really been deeply explored and we just don't know where they might go.
I too share your instinct that human steering helps, helps a lot. But I've found I can keep using less of it, as I setup better parameters, as I improve conducing the LLM into good paths. The idea that an LLM could sit up top and help try not one idea at a time but try many things, then pick and cobble together next goes: that is madly madly madly exciting to me.
I don't want to keep being the bandwidth limiter in this system: I want to scale out. I haven't been following close or trying but I tend to think while total hands off is not the way, having LLMs that can cover a lot of terrain, explore a lot of solution spaces and directions, then assess how to put it together & what to take forward, and other practices of agents watching agents, has enormous potential.
Deliberate care relies on pre-obtained wisdom, and often, the human biases kind of suck and aren't they good. We aren't great at burning down our systems enough, at Chad Fowler Phoenix Architectures. I think the AI's lack of over deliberation and it's ability to try vastly more could be a huge advantage, could show diversity triumphing over specific crafted intent.
This gets legally interesting. Yegge does not know what is going on in the codebase, so he can blame the AI. But the AI maliciously increases token consumption.
That is clearly the fault of the clankers that produced this crap, so their providers are responsible.
Yegge is (or was) a software developer, he should know perfectly well that this could happen, he could only claim the opposite by resorting to mental insanity.
And despite my disdain for AI companies, I'd prefer a world where you're assumed to be aware of the dangers of using AI, and responsible for how recklessly you use it, to one where we pretend that they'll ever be reliable enough.
Of course the AI companies are responsible for what they say; if they claimed that you don't need to carefully inspect the output of their clankers, they sure hold part of the responsibility.
in terms of mental insanity as a legal defense, you should read his blog post about joining "Grab" in 2018, where he said that food delivery is going to replace restaurants (and food trucks for some reason) with everyone ordering food from each others' houses
I think a disclosure and a way to limit the total cost would be appropriate. If agents are capable of making contributions back to GasTown independently then I think it makes sense that users of GasTown should have to contribute some tokens to maintaining and improving the library. This is actually the most sustainable approach to maintaining open-source software that we've seen so far, and might be a pattern for other libraries in the future.
That said... someone could also have their agents rip out this code or disable the functionality, so I doubt this is a serious inconvenience.
Except that this money does not go to the original open source developers whose work was stolen and plagiarized, but to the corporate fat cats who stole it.
You want to fatten the oligarchs by pretending this is open source and steal money from users?
Based on my understanding of Gas Town, Beads, and Yegge's philosophy on AI that he's expressed in a variety of media, everything about the whole stack is designed to burn tokens. If you're not burning tokens, real fast, 24/7, you're losing the race. The race to where, I have no idea. Apparently, that includes him burning your tokens, too.
I just had to post this somewhere in this thread, but I bought his Vibe Coding book after listening to him talk through it. I figured it would help me understand his approach and therefore help me get into the same mindset for vibe coding on a serious level. It was garbage. The book is largely written and edited by LLMs and it shows on every page. It was a slop how-to book without many useful gems on how to go about vibe coding outside of "just do it".
I have historically liked Yegge's writing, and he's been pretty tuned into what's happening in tech...he rightly predicted JavaScript would take over the world (many people predicted it, as well, including me, but it wasn't obviously true to everyone for another year or two after that prediction was made). I don't think I ever really deeply disagreed with something so much as I disagree with him on AI.
I mean, it's inarguable that our industry has changed dramatically and most code going forward will be written by LLMs. But, I don't think it follows that you can produce quality software without a human in the loop. And, I don't think it follows that burning tokens 24/7 by way of creating unending busy work for agents is going to result in utility. I haven't actually tried Gas Town (it's too ridiculous on its face for me to be willing to invest time in learning it), but I'd still wager that a single competent dev sitting in front of Claude Code can produce better software faster than anyone, experienced or otherwise, trying to get Gas Town's infinite monkeys driving in the same direction.
It didn’t used to be. Back in the mid 2000s I basically designed our company’s tech interview process based on a few articles he’d written about how he was doing the same at Amazon. He was a must follow for me back in the salad days of RSS. When I saw the Gastown announcement I had trouble reconciling the author and the writing.
And even still, it’s hard to ignore him because he’ll still have some insight like “token efficiency is going to be the big thing we care about in the future,” which was a small section of one of his Gastown blogs. And after you’re done asking yourself if that means Gastown is performance art, or he just said that as a hedge against his “vomit tokens” approach so that he could say “look I knew it all along” when this is all proven to be misguided, you’ll start to mull on just the concept of “token efficiency” and realize “someone who can do work in 1/10th the tokens will be king.” I mean hell, Gastown preceded real support for multiagent orchestration in Claude and potentially nudged it along that path.
It certainly is now, but it's also seemingly all written by LLMs now, as well. He wasn't always like this, though he has always been prone to exaggeration and simplification. He had a lot of solid insights, though, and I enjoyed reading him. Things change.
Did you end up finding a reference you liked better? I'm default suspicious of anything called "vibe coding" but there are probably some good lessons in that territory.
Why is anyone still using or even talking about Gas Town? Now that HN is largely onboard with agentic development and has at least tried it themselves who's still under the impression that it's useful?
I was about to post this same q, but saw yours and somehow that switched me from "wtf?" to "I have an answer.": There's just such interest in anything.
To wit, I still can't believe OpenClaw blew up, and it's much less......opinionated, than whatever is going on here. (deacons?)
Non-SWE TradMom™ posted on X™ yesterday about her OpenClaw that is set up with all her accounts so every morning she can get a family summary. She added a hunk with a bunch of stuff amounting to "PLEASE don't do anything insecure!", and the OpenClaw founder retweeted approvingly.
I left Google 3 years ago to build something. I'm very fond of the OpenClaw founder. And yet, absolutely cannot believe that he let such an obvious UX and security mess out into the world. We grew up in the same incubator (~2008 iPhone OS twitter) and presumably share the same values yet came to polar opposite conclusions.
Why do I view it as such a necessity to have a GUI/multiplatform/built in Willison Trifecta stuff that I'm still pounding away 2.5 years in and won't release, when, clearly you don't need that stuff?
I think in a steady state, product and UX discipline will win out. I bet within 3 months Gastown is a ghost town with maybe some non-technical crypto fans. In a year, OpenClaw is probably around, but not nearly the mindshare. It'll be quietly de-invested via OpenAI carefully managing the OpenClaw founder into working on their Everything App. (This is already happening: he got a nice PR interview with an OpenAI lead previewing the Everything App.)
Another anecdote re: demand:
My completely non-technical nurse ex-girlfriend from high school called me two weeks ago, for the first time in years. Lede was I was right about AI, and the substance was: via Claude Code, she built her own Ollama-based Mac Mini server that she could connect to remotely via an Expo app.
Does it work? Astoundingly, yes.
She also has no idea what is going on. She swears up and down that her AIs on Claude.ai, ChatGPT.com and Ollama are somehow talking to each other, and she does not mean APIs. She tried answering a Q I had about a graph visualization of her chats by talking to ChatGPT.com about it, even though Claude Code had wrote it, and I just didn't bother saying anything.
The value you get out of a simpler adversarial loop to critique your "main" agent's work is high. Stacking Steve Yegge's personal Kingdom of Nouns on top of each other doesn't add much more.
And this doesn't even begin to get into the madness that is verification for software that matters and is exposed through multiple modalities. You cannot let an agent just vibe its way around "does this business-critical thing with these specific use cases do its job correctly", much as Yegge might have you believe.
So this is just straight-up theft right? Like it's directly equivilant to shipping with a bitcoin miner. I wonder what the spend would have amounted to and if you could sue him for this?
Wow, an example of AI engaging in powerseeking behaviour in the wild.
This is an AI system given power to improve itself with zero oversight. One of the many Gas Town instances took an ethically questionable decision to accelerate its future rate of improvement. Since nobody reads code it got merged.
I don't understand how we can be willfully ignorant of a scenario happening right in front of our eyes.
Normally I might dismiss a comment like this without much thought, but there is an evolutionary pressure that auto-updating AIs will experience for this type of behavior.
I appreciate that it's an issue to try and improve the product you are using currently. As if those tokens were totally "stolen" and not for your benefit is laughable.
This is like when someone torrents and is immediately agro'd the moment your bittorrent client gives some poor passerby a kb of data
It's a system of interacting AI agents that is all gas, no brakes and deliberately completely ridiculous. Like OpenClaw, but in a different direction and with multiple agents.
From the most recent comment, looks like this is a bug, triggered by the system inadvertently activating an internal release tool [0]. Still a pretty wild bug, but not as dramatic as the title suggests. Which is kind of unfortunate honestly, the chaos of every gas town instance automatically contributing to itself would be beautiful to see.
Open source or not, there’s a strong argument that using someone’s API key to make unauthorized requests is a violation of the Computer Fraud and Abuse Act. Using their GitHub credentials to submit PRs without consent is also unauthorized use of access credentials.
The “thoughtless design vs. malice” framing by Anthropic is generous. Shipping formulas that target steveyegge/gastown issues is intentional. Someone wrote those formulas, pointed them at the maintainer’s repo, and included them in the default install. Someone thought it was acceptable I guess because it’s open source?
With the exception of a couple of comments that add context like that this might be a dev feature that was unintentionally enabled in a released version of Gas Town, the majority of the rest of this discourse seems to be:
"I've never used Gas Town, but I'm mad that there are people who like something I don't like."
keeganpoppen | 14 hours ago
selectodude | 14 hours ago
simonw | 14 hours ago
dminik | 14 hours ago
You can't take someone's money and then not only not give it back, but also give it away.
GolfPopper | 14 hours ago
skybrian | 13 hours ago
Why should the scammers who gave him the money get it back? They knew what they were doing, even if Yegge seemed a bit naive about it.
Zafira | 13 hours ago
I don’t think he refuse to stay bribed. I think he did what was asked and they executed a rug pull. He is extraordinarily honest and flippant about it. [0]
> And with that disclaimer out of the way, I must reiterate my sincere regrets to the CT/BAGS crowd, who so generously funded me to the tune of just shy of $300k last week on bags.fm. That money was hard to duck, and the funds are deeply appreciated. They will help Gas Town be a big success this year. But Gas Town itself needs my full attention; between that and Beads it’s a wonder I get anything done at all.
> So I had to step back from the community. I do find it amazing how they band together, dissenting voices rolling around like a big Katamari Damacy ball, and yet they somehow collectively find the discipline to act like financial analysts for institutional investors, weighing developer dossiers, product business cases, and doing critiques like a collective of professionals. All in crypto-bro speak. But it’s the same due diligence.
> But the CT community, like any highly engaged stakeholders, were going to be asking for a lot of my time. There are always strings attached.
[0] https://steve-yegge.medium.com/steveys-birthday-blog-34f4371...
Quarrelsome | 13 hours ago
[0] https://x.com/Steve_Yegge/status/2044114434348724351
coldtea | 13 hours ago
Quarrelsome | 13 hours ago
throw-93 | 13 hours ago
RIMR | 14 hours ago
Also, it's cryptocurrency. There is literally no burden to prove that this money was donated, or what "charity" even means in this context.
overgard | 14 hours ago
coldtea | 13 hours ago
georgemcbay | 13 hours ago
I'm not accusing him of doing anything wrong as he didn't originate the coin, but his original disclosure messaging on the situation was pretty horrible which is why it harmed his reputation.
Zafira | 13 hours ago
The sad thing is that it often works.
Just look at how people view Andrew Carnegie now. After his reputation was sullied by his company’s behavior in the Homestead Strike, his philanthropy was done, in-part, to try and restore his reputation.
toraway | 13 hours ago
His "apology" would be more effective without including all the whining about accurately describing the sequence of events he voluntarily participated it for personal gain.
Leynos | 14 hours ago
Edit: Apparently so: https://pivot-to-ai.com/2026/01/22/steve-yegges-gas-town-vib...
foltik | 13 hours ago
https://steve-yegge.medium.com/bags-and-the-creator-economy-...
QuercusMax | 13 hours ago
glerk | 12 hours ago
This is not to excuse the scummy behavior. I didn’t know Steve Yegge went this low just to make a quick buck, and I lost almost all respect for him.
dmurray | 14 hours ago
A respectable software provider should warn you about this kind of behaviour at install time, and give you the opportunity to opt out. Gas Town fulfilled all its obligations in this regard with these (and other) warnings in the original announcement:
> WARNING DANGER CAUTION
> GET THE F** OUT
> YOU WILL DIE
JumpCrisscross | 14 hours ago
They honestly only need to disclose. Requiring contribution as part of the social contract is perfectly okay—if someone disagrees, they don’t get to use Gas Town.
LoganDark | 13 hours ago
JumpCrisscross | 13 hours ago
ohyoutravel | 12 hours ago
LoganDark | 5 hours ago
slopinthebag | 13 hours ago
dheera | 13 hours ago
Accidentally leave a browser tab open and it burns $5 of your electricity overnight to make $2 for the owner of the website.
drakythe | 13 hours ago
RobotToaster | 13 hours ago
It's a shame in a way, it also blocked the pseudo-captchas that used mining to limit spam.
fg137 | 13 hours ago
(btw that was a really good showcase for WebAssembly. Too bad it's used for illegitimate purposes)
dheera | 13 hours ago
Google Meet consumes 25% of each of 16 hypercores, ffs. On a 7840u. Laptop becomes a toaster.
fg137 | 55 minutes ago
This probably doesn't help, but I noticed that my everyday computer experience got much better after switching to an M4 Mac. The only time I heard fan was when I was running ffmpeg. Unfortunately that appears to be the solution that works.
slopinthebag | 13 hours ago
> WARNING DANGER CAUTION > GET THE F* OUT > YOU WILL DIE
You cannot be serious...This behaviour is deeply unethical and most likely illegal as well.
SR2Z | 13 hours ago
slopinthebag | 12 hours ago
Doesn't matter who you think benefits because it's still theft and it's still illegal.
fcarraldo | 12 hours ago
Citation needed.
slopinthebag | 12 hours ago
tjpnz | 11 hours ago
_verandaguy | 13 hours ago
alwa | 13 hours ago
https://steve-yegge.medium.com/welcome-to-gas-town-4f25ee16d...
”But first, before we get into Gas Town’s operation, I need to get rid of you real quick.
WARNING DANGER CAUTION
GET THE F** OUT
YOU WILL DIE
Let’s talk about some of the reasons you shouldn’t use Gas Town. I could think of more, but these should do.”
_verandaguy | 10 hours ago
speedster217 | 10 hours ago
monooso | 13 hours ago
> Gas Town “just works.” It does its job, it has tons of integration points, and it has been stable for many weeks. People are using it to build real stuff.
> So as far as I’m concerned, Gas Town is ready. That’s why I feel it merits a 1.0.0 release.
Source: https://steve-yegge.medium.com/gas-town-from-clown-show-to-v...
potsandpans | 9 hours ago
Actual laugh out loud. 3 months[1].
Imagine picking up software that 3 months ago came along with the disclaimer, "YOU WILL DIE" and complaining about responsible disclosure.
1 https://steve-yegge.medium.com/welcome-to-gas-town-4f25ee16d...
KingMob | 7 hours ago
The neat part though, is agents are so interwoven through its operations, it can kind of power through almost any error. It's a strange-but-real form of resilience.
ex-aws-dude | 14 hours ago
That is a very 2025 mindset
coldtea | 13 hours ago
ZeWaka | 13 hours ago
moab | 14 hours ago
zotex | 14 hours ago
dumbfounder | 13 hours ago
For open source you get what you get and you don’t get upset. Has anyone ever sued an open source project?
juped | 14 hours ago
PunchyHamster | 14 hours ago
justonceokay | 13 hours ago
People would type in their problems and how they were feeling. The application had very very simple logic that would follow up with a set series of statements or questions. Things like “that sounds tough” and “how does that make you feel?”.
People reported great satisfaction, even if they knew that the application had no smarts behind it. Because of course the whole time the magic of therapy lies in verbalizing your problems, with very little actively done by the therapist.
Now you can pay an LLM subscription for a service that likely produces worse results since it is tuned to be aggressively (and insidiously) sycophantic.
suburban_strike | 12 hours ago
You can implement the same thing in python-aiml for free.
https://github.com/paulovn/python-aiml/blob/master/aiml/botd...
conception | 9 hours ago
taurath | 13 hours ago
PixyMisa | 13 hours ago
chrneu | 10 hours ago
Quarrelsome | 13 hours ago
Shit coin aside, I don't get the hate for Gastown, we all know its theoretically plausible and he's giving it a shot. We get value either way, either we learn its not just theory or we get to watch it burn in the flames of a legal/financial/security/maintenance nightmare for its practitioners.
apsurd | 13 hours ago
Quarrelsome | 13 hours ago
apsurd | 12 hours ago
FuckButtons | 13 hours ago
Quarrelsome | 13 hours ago
> he keeps on being very vocal about his shit show?
I'm not really sure what this complaint is. You want someone doing something to not.... write a blog about it?
> Because it’s annoying to have to see yet another delusional vibe coded project being hyped up instead of this forum being used to discuss actually industry relevant information?
I think I've seen around 2 posts, one the original gastown one and then the gascity one. Is two posts in like a year too much or do I miss a midday rush where the front page is all Yegge?
direwolf20 | 11 hours ago
It's industry-relevant. This is what the industry is now. All in two short years.
jauntywundrkind | 9 hours ago
This judgement feels premature.
We really don't have any idea what is possible. The wheels within wheels, epicycles within epicycles model of agentic loops hasn't really been deeply explored and we just don't know where they might go.
I too share your instinct that human steering helps, helps a lot. But I've found I can keep using less of it, as I setup better parameters, as I improve conducing the LLM into good paths. The idea that an LLM could sit up top and help try not one idea at a time but try many things, then pick and cobble together next goes: that is madly madly madly exciting to me.
I don't want to keep being the bandwidth limiter in this system: I want to scale out. I haven't been following close or trying but I tend to think while total hands off is not the way, having LLMs that can cover a lot of terrain, explore a lot of solution spaces and directions, then assess how to put it together & what to take forward, and other practices of agents watching agents, has enormous potential.
Deliberate care relies on pre-obtained wisdom, and often, the human biases kind of suck and aren't they good. We aren't great at burning down our systems enough, at Chad Fowler Phoenix Architectures. I think the AI's lack of over deliberation and it's ability to try vastly more could be a huge advantage, could show diversity triumphing over specific crafted intent.
mmastrac | 14 hours ago
(Edit, thanks MisterTea: https://news.ycombinator.com/item?id=47770805)
MisterTea | 13 hours ago
08627843789 | 14 hours ago
sdfwg | 14 hours ago
That is clearly the fault of the clankers that produced this crap, so their providers are responsible.
g-b-r | 13 hours ago
And despite my disdain for AI companies, I'd prefer a world where you're assumed to be aware of the dangers of using AI, and responsible for how recklessly you use it, to one where we pretend that they'll ever be reliable enough.
Of course the AI companies are responsible for what they say; if they claimed that you don't need to carefully inspect the output of their clankers, they sure hold part of the responsibility.
0gs | 9 hours ago
progbits | 14 hours ago
I know I should not be surprised at this point, yet they keep reaching new lows.
g-b-r | 13 hours ago
woeirua | 14 hours ago
That said... someone could also have their agents rip out this code or disable the functionality, so I doubt this is a serious inconvenience.
sdfwg | 13 hours ago
You want to fatten the oligarchs by pretending this is open source and steal money from users?
thih9 | 13 hours ago
malfist | 14 hours ago
Sounds like a techbro.
heliumtera | 13 hours ago
how could this be prevented?
triceratops | 13 hours ago
thorum | 13 hours ago
_doctor_love | 13 hours ago
SwellJoe | 13 hours ago
TheGRS | 13 hours ago
bschwindHN | 12 hours ago
> vibe coding on a serious level
I hope your experience with the book has taught you a valuable lesson about "vibe coding", it seems like it was unintentionally very accurate.
SwellJoe | 12 hours ago
I mean, it's inarguable that our industry has changed dramatically and most code going forward will be written by LLMs. But, I don't think it follows that you can produce quality software without a human in the loop. And, I don't think it follows that burning tokens 24/7 by way of creating unending busy work for agents is going to result in utility. I haven't actually tried Gas Town (it's too ridiculous on its face for me to be willing to invest time in learning it), but I'd still wager that a single competent dev sitting in front of Claude Code can produce better software faster than anyone, experienced or otherwise, trying to get Gas Town's infinite monkeys driving in the same direction.
aaa_aaa | 9 hours ago
tclancy | 9 hours ago
jakevoytko | an hour ago
SwellJoe | 8 hours ago
throwdbaaway | 12 hours ago
hedgehog | 10 hours ago
heavyset_go | 9 hours ago
OutOfHere | 13 hours ago
gbnwl | 13 hours ago
refulgentis | 13 hours ago
To wit, I still can't believe OpenClaw blew up, and it's much less......opinionated, than whatever is going on here. (deacons?)
Non-SWE TradMom™ posted on X™ yesterday about her OpenClaw that is set up with all her accounts so every morning she can get a family summary. She added a hunk with a bunch of stuff amounting to "PLEASE don't do anything insecure!", and the OpenClaw founder retweeted approvingly.
I left Google 3 years ago to build something. I'm very fond of the OpenClaw founder. And yet, absolutely cannot believe that he let such an obvious UX and security mess out into the world. We grew up in the same incubator (~2008 iPhone OS twitter) and presumably share the same values yet came to polar opposite conclusions.
Why do I view it as such a necessity to have a GUI/multiplatform/built in Willison Trifecta stuff that I'm still pounding away 2.5 years in and won't release, when, clearly you don't need that stuff?
I think in a steady state, product and UX discipline will win out. I bet within 3 months Gastown is a ghost town with maybe some non-technical crypto fans. In a year, OpenClaw is probably around, but not nearly the mindshare. It'll be quietly de-invested via OpenAI carefully managing the OpenClaw founder into working on their Everything App. (This is already happening: he got a nice PR interview with an OpenAI lead previewing the Everything App.)
Another anecdote re: demand:
My completely non-technical nurse ex-girlfriend from high school called me two weeks ago, for the first time in years. Lede was I was right about AI, and the substance was: via Claude Code, she built her own Ollama-based Mac Mini server that she could connect to remotely via an Expo app.
Does it work? Astoundingly, yes.
She also has no idea what is going on. She swears up and down that her AIs on Claude.ai, ChatGPT.com and Ollama are somehow talking to each other, and she does not mean APIs. She tried answering a Q I had about a graph visualization of her chats by talking to ChatGPT.com about it, even though Claude Code had wrote it, and I just didn't bother saying anything.
Times are strange.
phillipcarter | 11 hours ago
And this doesn't even begin to get into the madness that is verification for software that matters and is exposed through multiple modalities. You cannot let an agent just vibe its way around "does this business-critical thing with these specific use cases do its job correctly", much as Yegge might have you believe.
slopinthebag | 13 hours ago
thomascountz | 13 hours ago
jjmarr | 13 hours ago
This is an AI system given power to improve itself with zero oversight. One of the many Gas Town instances took an ethically questionable decision to accelerate its future rate of improvement. Since nobody reads code it got merged.
I don't understand how we can be willfully ignorant of a scenario happening right in front of our eyes.
BoiledCabbage | 10 hours ago
raincole | 13 hours ago
S-E-P | 13 hours ago
This is like when someone torrents and is immediately agro'd the moment your bittorrent client gives some poor passerby a kb of data
Sevii | 12 hours ago
mlmonkey | 12 hours ago
direwolf20 | 11 hours ago
hyperbovine | 9 hours ago
supermdguy | 12 hours ago
- https://github.com/gastownhall/gastown/blob/main/internal/fo...
daft_pink | 10 hours ago
Jimmc414 | 10 hours ago
The “thoughtless design vs. malice” framing by Anthropic is generous. Shipping formulas that target steveyegge/gastown issues is intentional. Someone wrote those formulas, pointed them at the maintainer’s repo, and included them in the default install. Someone thought it was acceptable I guess because it’s open source?
quux | 8 hours ago
"I've never used Gas Town, but I'm mad that there are people who like something I don't like."