svara

The vibe coding maximalist position can be stated in information theory terms: That there exists a decoder that can decode the space of useful programs from a much smaller prompt space.

The compression ratio is the vibe coding gain.

I think that way of phrasing it makes it easier to think about boundaries of vibe coding.

"A class that represents (A) concept, using the (B) data structure and (C) algorithms for methods (D), in programming language (E)."

That's decodeable, at least to a narrow enough distribution.

"A commercially successful team communication app built around the concept of channels, like in IRC."

Without already knowing Slack, that's not decodable.

Thinking about what is missing is very helpful. Obviously, the business strategic positioning, non technical stakeholder inputs, UX design.

But I think it goes beyond that: In sufficiently complex apps, even purely technical "software engineering" decisions are to some degree learnt from experiment.

This also makes it more clear how to use AI coding effectively:

* Prompt in increments of components that can be encoded in a short prompt.

* If possible, add pre-existing information to the prompt (documentation, prior attempts at implementation).

The funny thing about this is that even if the output is bad, it's actually good.

Could you say more on how the tasks where it works vs. doesn't work differ? Just the fact that it's both small and greenfield in the one case and presumably neither in the other?

Hey, thanks, that was quite interesting!

I'd be curious to hear your thoughts on how the "fixer", who sounds rather ineffective as an executive, came into this position, in what sounds like overall a rather effective organization.

I've been personally thinking quite a bit about what makes organizations work or not work recently, and your story is quite interesting to me as a glimpse into a kind of organization that I've never seen from the inside myself.

How large a demand for cars does the Chinese government have do you think?

> I really don't see a solid economic future for Germany when enough other countries implement more progressive economic policies.

People do change their minds when the pain becomes too intense to ignore, but that is what it takes.

Yes, throughput is determined by the bottleneck and above a certain organization size, the bottleneck often is coordination costs.

Yes, it's the lump of labor fallacy.

Doesn't exclude the possibility of short term distribution, though.

Can you give some examples of this? Maybe have something online? I would love to learn more about how to do proof driven AI assisted development.

My understanding is that all recent gains are from post training and no one (publicly) knows how much scaling pretraining will still help at this point.

Happy to learn more about this if anyone has more information.

Do you keep a collection of these? What for?

I think that mostly depends on how good a writer you are. A lot of people aren't, and the AI legitimately writes better. As in, the prose is easier to understand, free of obvious errors or ambiguities.

But then, the writing is also never great. I've tried a couple of times to get it to write in the style of a famous author, sometimes pasting in some example text to model the output on, but it never sounds right.

Yes, it's been odd to observe the parallels with the web3 craze.

You asked people what their project was for and you'd get a response that made sense to no one outside of that bubble, and if you pressed on people would get mad.

The bizarre thing is that this time around, these tools do have a bunch of real utility, but it's become almost impossible online to discuss how to use the tech properly, because that would require acknowledging some limitations.

Opus 4.6:

Walk! At 50 meters, you'll get there in under a minute on foot. Driving such a short distance wastes fuel, and you'd spend more time starting the car and parking than actually traveling. Plus, you'll need to be at the car wash anyway to pick up your car once it's done.

You're right, and I enjoy using coding agents too. I've built some things with them I wouldn't have otherwise.

However, it's been a full quarter now since November 2025.

Based on facts on the ground, i.e. the rate and quality of new software and features we observe, change has been nowhere as dramatic as your comment would suggest.

It seems to me that a possible explanation is that people get very excited about massive speedups in specific tasks, but the bottleneck of the system shifts somewhere else immediately (e.g, human capacity for learning, team coordination costs, communication delays).

> So getting AI to do bits, really means getting AI to do the really easy bits.

As someone who gets quickly bored with repetitive work, this is big though.

> The wording here is fascinating, mainly because they're effectively acting as arbiters of "vibes".

A very common tension in law everywhere.

In the US you now have a 'major questions doctrine'. What the hell is a major question?

One of the things about this story that don't sit right with me is how Scott and others in the GitHub comments seem to assign agency to the bot and engage with it.

It's a bot! The person running it is responsible. They did that, no matter how little or how much manual prompting went into this.

As long as you don't know who that is, ban it and get on with your day.

If past patterns are anything to go by, the complexity moves up to a different level of abstraction.

Don't take this as a concrete prediction - I don't know what will happen - but rather an example of the type of thing that might happen:

We might get much better tooling around rigorously proving program properties, and the best jobs in the industry will be around using them to design, specify and test critical systems, while the actual code that's executing is auto-generated. These will continue to be great jobs that require deep expertise and command excellent salaries.

At the same, a huge population of technically-interested-but-not-that-technical workers build casual no-code apps and the stereotypical CRUD developer just goes extinct.

More likely you would just train for emitting svg for some description of a scene and create training data from raster images.

Software engineers (US based particularly) were more than happy about software eating the economy when it meant they'd make 10x the yearly salary of someone doing almost any other job; now that AI is eating software it's the end of the world.

Just saying, what you're describing is entirely unsurprising.

It's really unclear unfortunately.

The correlative effect is quite clear, i.e people who have high omega 3 levels (eat a lot of fish) have health benefits.

But in random controlled trials Omega 3 supplements have not had convincing effects.

It might be because the supplements aren't very good, or because there's actually something completely different going on, like fish displaces less healthy foods from the diet.

I mean, introducing a technique from one field in another is innovative.

You don't get to claim you invented it, but a lot of progress happens by finding connections between things that are individually well known.

I hear you, and that's where my mind goes first on this issue too.

But with social media, many of the people most into it, when asked, will say they wish it didn't exist.

A lot of kids feel they have to be on it, but wish it didn't exist.

People sound and behave more like actual drug addicts than just mere fans of a medium around social media.

The writing on this website is giving strong web3 vibes to me / doesn't smell right.

The only reason I'm not dismissing it out of hand is basically because you said this team was worth taking a look at.

I'm not looking for a huge amount of statistical ceremony, but some detail would go a long way here.

What exactly was achieved for what effort and how?

Everything's on a spectrum, but there's a point where you're so far along on the spectrum that it makes sense to call it something else.

See, "quantity has a quality of its own".

Sometimes you have to leave the theoretical view aside and just look out the window. How are people using this? Is it hurting them? What can we do about it?

I don't like blanket bans, but putting TikTok and, say, a publishing company marketing novels, in the same category because they strive for an audience, doesn't clarify anything. It just confuses the discussion.

> (...) you get these two camps 'for and against' whereas the best way forward is to insist on data rather than anecdotes.

I think that's an issue with online discussions. It barely happens to me in the real world, but it's huge on HN.

I'm overall very positive about AI, but I also try to be measured and balanced and learn how to use it properly. Yet here on HN, I always get the feeling people responding to me have decided I am a "true believer" and respond to the true believer persona in their head.

That's one way to look at it, as just the next iteration of subredditsimulator.

The qualitatively new step leading to emergent behavior will be when the agents start being able to interact with the real world through some interface and update their behavior based on real world feedback.

Think of an autonomous, distributed worm that updates its knowledge base of exploit techniques based on trial and error and based on information it discovers as it propagates.

It might start doing things that no human security researcher had foreseen, and that doesn't require great leaps of the imagination based on today's tech.

That's when you close the evolutionary loop.

I think this isn't quite that yet, but it points in that direction.

You're right it isn't fair to reject someone's scientific argument just because they seem unfamiliar with how professional science works.

We shouldn't have believed the study more if it actually had been in Nature.

I don't think that's what I was saying, though.

The issue in this thread was about taking a step back and looking at the overall plausibility of the conclusions, taking together multiple studies.

I agree with the GP that the argument doesn't really pass the smell test.

That's still the main issue, and it is something that people who don't understand scientific publishing struggle understanding/doing, because they lack the intuition for how certain results came about.

It does not require more explanation: publication bias means null results aren't in the literature; do enough small low quality trials and you'll find a big effect sooner or later.

Then the supposed big effect attracts attention and ultimately properly designed studies which show no effect.