jt2190

> How do you not see the difference between a machine that will hallucinate something random if it doesn’t know the answer vs a human...

Your claim here is that humans can't hallucinate something random. Clearly they can and do.

> ... that will logic through things and find the correct answer.

But humans do not find the correct answer 100% of the time.

The way that we address human fallibility is to create a system that does not accept the input of a single human as "truth". Even these systems only achieve "very high probability" but not 100% correctness. We can employ these same systems with AI.

The market currently values your reading of HN comments at $0.

Having an app that tells you that you have a thing is not the same as physically having the thing. Owning gold in that sense is no different than owning bitcoin: They both rely on a societal fiction that you can actually get the value of the asset when you expect that you can. In other words: The presence of a reliable financial and legal system is what makes the myth real for both assets.

We forget that the idea that the gold could be “further away from”/“not actually in the physical possession of” the human who “owns” was not universally accepted at all points in history. In a weird way, bitcoin is more forthright in that it says out loud “trust me this is valuable because we all agree it just is” where non-custodial ownership of gold tries to imply that there’s less need for this trust because there’s an actual shiny rock on the planet somewhere and humans have a history of valuing shiny rocks.

Ultimately it seems like it’s the “trust” part that mostly matters. Bitcoin is certainly harder to trust at the moment. But gold is rallying because there is less trust overall. That’s not good for either asset.

Every successful software project reaches an equilibrium between utility for its operators and bugs, and that point very rarely settles at 0% bugs [1].

When software operators tolerate bugs they’re signaling that they’re willing to forego the fix in exchange for other parts of the feature that work and that they need.

The idea that consumers will somehow not need the features that they rely on anymore is completely wrong.

That leaves the tolerable bugs, but those were always part of the negotiation: Coding agents doesn’t change that one bit. Perhaps all it does it allow more competitors to peel away those minority groups of users who are blocked by certain unaddressed bugs. Or maybe it gets those bugs fixed because it’s cheaper to do so.

> I also do not see it happening at scale while competition is considered the default operating mode of society at large.

You don’t even need competition between people and orgs, just between solutions that work more-or-less equally but come with different second-order tradeoffs. Consider two approaches that solve a company’s problem equally but create different amounts of work for different people in the organization. Which solution to choose? Who gets to decide, based on what criteria? As soon as even a little scale creeps in this is inescapable.

> The French government… announced last week that 2.5 million civil servants would stop using video conference tools from U.S. providers — including Zoom, Microsoft Teams, Webex and GoTo Meeting — by 2027 and switch to Visio, a homegrown service.

> I think the reality is that… people are willing to take on MUCH greater risk today for reward than they were prior to the pandemic.

I’ll quibble that people have no idea of the risks they’re taking. I read somewhere that amateur stock traders spent something like 4 minutes researching their purchase. Balanced portfolios are just too boring and tedious.

Last paragraph is informative:

> Anthropic relies heavily on a combination of chips designed by Amazon Web Services known as Trainium, as well as Google’s in-house designed TPU processors, to train its AI models. Google largely uses its TPUs to train Gemini. Both chips represent major competitive threats to Nvidia’s best-selling products, known as graphics processing units, or GPUs.

So which leading AI company is going to build on Nvidia, if not OpenAI?

> [Seth is using AI to try] to take over the job of the farmer. Planting, harvesting, etc. is the job of a farmhand (or custom operator).

Ok then Seth is missing the point of the challenge: Take over the role of the farmhand.

> Everyone is working to try to automate the farmhand out of a job, but the novelty here is the thinking that it is actually the farmer who is easiest to automate away.

Everyone knows this. There is nothing novel here. Desk jockeys who just drive computers all day (the Farmer in this example) are _far_ easier to automate away than the hands-on workers (the farmhand). That’s why it would be truly revolutionary to replace the farmhand.

Or, said another way: Anything about growing corn that is “hands on” is hard to automate, all the easy to automate stuff has already been done. And no, driving a mouse or a web browser doesn’t count as “hands on”.

> Employees get options at common stock prices.

More specifically, when employees are granted options contracts the strike price of those contracts is based on the last valuation of the company prior to the grant. If all is going well and the valuation is increasing those options are also increasing in value. Here we have a sale which values the company lower than the prior valuation. Recent option grants will likely be underwater, earlier grants would still be profitable.

> The valuations you see, like $12bn, are for preferred stock.

No, the valuation is for the whole company, all of its shares, preferred and common. How this value is distributed among shareholders depends on the deal, but generally there is a “seniority”, roughly: creditors (debt holders) are paid first, preferred shares next, then common shareholders last. This order can be negotiated as part of the sale.

> So no employees got stock priced at $12bn, but all of them get paid at a $5.15bn valuation.

It’s just not possible to know what each individual employee’s outcome is. We don’t know how much of that 5.5 billion will be left over for common shareholders including the employees. Note that employees have received salaries so their overall outcome is greater than zero dollars, but perhaps their total compensation outcome is lower than they hoped for the time they put in.

> Not saying they did well, but depending on the 409a valuations, they still might have made money.

Yes, some might have and some might have not. We just don’t know without more details.

Edit: singron’s answer (sibling comment) attempts to model the employee outcome in a rough but reasonable way.

The tariffs haven’t really started hitting consumers yet, believe it or not. Major retailers rushed to get their Christmas goods into the U.S. before the deadlines. In 2026 we’ll start to see more of the impacts.

It's a bit more complex and nothing new.

Roughly: An individual lawyer creates a "national brand" law firm and advertises it widely to find new clients. Of course, an individual lawyer can only take so many clients a year, so advertising nationally doesn't make sense since they'd quickly have more clients than they could ever manage. Additionally, a lawyer advertising nationally would need to be licensed in many jurisdictions and possibly also be able to handle many different legal specialties. This is essentially impossible for one person to do themselves.

Instead, the lawyer sends clients to their "associates" in the client's jurisdiction with the specialty that the client needs. For this referral, the lawyer gets a fee, agreed to beforehand.

Probably because the bidder is an aggregator sending the referral to many law firms, not just one. Individual law firms thus get outbid for an exclusive referral click, but can still pay the aggregator for a non-exclusive referral.

That is not what Eric S. Raymond (esr) was describing.

GNUnix was developed using the Cathedral-style, Linux was developed using the bazaar-style. How Linux development was coordinated was thought to be impossible for something that had to be as solid as an operating system. The essay is a deep dive, exploring the conditions that the Linux project needed to ship an OS.

> In large companies, speaking up about what you see as a “bad project” is a good thing. But only in moderation. Sometimes the mark of seniority is realizing that arguing with people who won’t listen isn’t worth it; it’s better to save your counsel.

To be clear, you’re literally saying:

Legacy == vibe coded

And:

Others can not depend on vibe coded software

Thus you seem to mean:

Legacy code can not be depended on

I presume that through some process one can exorcise the legacy/vibe-codiness away. Perhaps code review of every line? (This would imply that the bottleneck to LLM output is human code review.) Or would having the LLM demonstrate correctness via generated tests be sufficient?

You’re “pushing back” against the term “legacy code” with an argument that someone “needs to take responsibility”.

> Teams don't just work together magically and "organically".

This is the “employees are resources” mentality, and is common in mature companies. In a startup, however, you need to hire for individuals that will “make it rain”, and not wait around for some process or manager to tell them to communicate or do work. If your employees are not coming together as a team and figuring things out without your handholding you’ve hired the wrong people.

Where did you get the idea that “legacy code” equals “abandonware”? The world runs on massive legacy codebases that have been maintained for decades.

> I doubt that the LLM drivers deliver something better…

That is your definition of “better”. If we’re going to trade our expertise for coin, we must ask ourselves if the cost of “better” is worth it to the buyer. Can they see the difference? Do they care?

LLMs remove the familiarity of “I wrote this and deeply understand this”. In other words, everything is “legacy code” now ;-)

For those who are less experienced with the constant surprises that legacy code bases can provide, LLMs are deeply unsettling.

> … not all programmers program for the same reason, for some of us, LLMs helps a lot, and makes things even more fun. For others, LLMs remove the core part of what makes programming fun for them. Hence we get this constant back and forth of "Can't believe others can work like this!" vs "I can't believe others aren't working like this!", but both sides seems to completely miss the other side.

Unfortunately the job market does not demand both types of programmer equally: Those who drive LLMs to deliver more/better/faster/cheaper are in far greater demand right now. (My observation is that a decade of ZIRP-driven easy hiring paused the natural business cycle of trying to do more with fewer employees, and we’ve been seeing an outsized correction for the past few years, accelerated by LLM uptake.)

> But they promise [the creation of wealth] to the society.

It feels like you're trying to describe the "social contract" between private banks and the rest of society, but putting the full responsibility of "wealth creation" only on one party in the contract: The bank.

The other party, Society, is given access to capital when they borrow. The rate they're charged should be competitive since there is presumably more than one lender. By borrowing money they try to "create wealth", then to repay the loan principal along with a bit of the new wealth in the form of interest.

Very interesting. I would have intuitively hand-waved the variability of each laborer’s output for those as “within a tolerable amount of time, most of the time” but I see your point that at the busiest times an order for a single, more complex item could cause a significant slowdown in the whole pipeline, especially, as you point out, there is a limited number of staff and workstations.

> ... there's a good deal of work contained in the "head cache" of the workers (lack of transparency)...

Is this true in a Starbucks? Every order goes into the computer before it's made, whether via the app or at the register. Or are you referring to something else more specific, like performing and recording the individual steps in making a speciality coffee?

I still don’t understand why a rate-limiting approach is not preferred. Why should I care if the abuse is coming from a bot or the world’s fastest human? Is there a “if you need to rate limit you’ve already lost” issue I’m not thinking of?

The source report

“How cheap is battery storage?“ https://ember-energy.org/latest-insights/how-cheap-is-batter...

In short: Cheaper batteries plus already cheap solar means that solar is now a cheap source of “anytime electricity”.

The assertion was that switching vendors would save $10M. I asked why the new vendor would forego $10M that the old vendor was able to collect. Are you saying that the new vendor has to offer this discount otherwise there’s no incentive to migrate? (I agree that migrating is very difficult politically.)

Why would any Enterprise Software vendor leave $10M on the table?

It feels like you’ve mixed up cause and effect: Getting access to mental health resources that can result in a diagnosis does not bestow a higher social rank upon people, those people’s existing higher social rank allows them easier access to mental health resources and thus they are more frequently diagnosed/treated.

You also seem to be implying that these diagnoses are frequently not merited and just used to increase social rank somehow, which is extremely cynical.