DEV Community: Walt Kessler

Nobody blushes at a dashboard

Walt Kessler — Sun, 25 Oct 2026 09:48:08 +0000

Three New Yorker columns this autumn argue that shame is dying and that we will need regulation to replace it. I've spent forty years in the industry they blame. Here is why I think the diagnosis is right and the mechanism is wrong, with a nuclear dump, a water court, a basketball team and a day care center as witnesses.

The argument, in three lines. A price turns an obligation into an amount, and the obligation rarely survives the translation. A metric turns a purpose into a number, and the purpose drifts the same way. Shame never protected us from either. What protected us was people who could see, respond in steps, and let you back in. We replaced them with numbers, and nobody blushes at a number.

Two prices

In Haifa, in the late 1990s, ten day care centers had the small problem every day care center has. Some parents came late. The teachers stayed on, unpaid, with somebody else's child, and the parents arrived apologizing.

Two economists talked six of the centers into a fix. Arrive more than ten minutes late and you pay ten shekels, about the price of a sandwich. Lateness went up. Not a little. It roughly doubled, and it stayed doubled. After twelve weeks the fine was dropped, and lateness did not come back down.¹

Jay Caspian Kang ran into the same thing in a parking lot. In the third of three New Yorker columns this September on what he calls the death of shame, he describes a youth basketball tournament charging twenty-five dollars to park and twenty a head to watch twelve-year-olds. Another parent shrugged.

"Yeah, they're just being smart, I guess."

Kang's series is about that shrug.² His argument runs like this. Shame is how communities enforce their norms without the state. Life on phones has dissolved the communities, so the shame has gone with them. Into the gap walked the optimizers of tech and finance, who "believe that they have obviated society's need for shame," and whose answer to every objection is that the number went up. He ends his first column with a conclusion I think he reached reluctantly: "our world at large will need intervention and direct regulation."

I have spent forty years inside the industry Kang is describing. I think he has the symptom right and the mechanism wrong. Shame is not dying because of phones. And it was never the thing doing the work.

What a price does

The Haifa result became famous because it is counterintuitive and short. It is also less settled than its fame.

The authors offered two explanations and admitted they could not tell them apart. One is that the parents learned something: the worst that happens if I'm late is ten shekels, so it can't be that bad. The other is that the fine changed what lateness was. Before, it was an imposition on someone doing you a kindness. After, it was a service with a price, and you were a customer. A later attempt to reproduce the finding with surveys did not reproduce it.³ An essay built on one famous study should say so.

The sturdier evidence comes from a Swiss village.

In 1993, Wolfenschiessen was a candidate site for a nuclear waste repository, and a team of economists surveyed the residents. A bare majority, 50.8 percent, said they would vote to accept it. The researchers then asked the same people whether they would still accept it if the government paid every resident several thousand dollars a year in compensation. Support fell to 24.6 percent.⁴

Offered money, the villagers withdrew consent they had just given for free. Many told the surveyors why. They had been willing to carry a burden for the country, and the payment made it look like they were being bought.

A price is a translation. Some things survive it. Some don't, because the thing of value was never the amount. It was that you did it unpaid.

Once the translation has happened, you can't easily undo it. The Haifa centers found that out when they removed the fine and nobody went back to feeling bad.

Kang is right that the modern economy is a translation machine of this kind. He is wrong that it was invented in a garage in Palo Alto. Churches rented pews. Guilds fined members for shoddy work until the fine became a cost of business. What is new is the speed and the scale. That is a real difference. It is a difference of degree.

What a number does

Now go one level up, from a price to a measurement. The same thing happens, and it's harder to see.

In 1982 the engineers building one of the first personal computers with a graphical screen were asked to fill in a weekly form: how many lines of code did you write this week? One of them spent a week rewriting a piece of the graphics software so that it ran about six times faster and was two thousand lines shorter. He wrote "-2000" on the form. The forms stopped a few weeks later.⁵

Everyone who has worked in an office has a version of that story, and the reason it's funny is that everyone recognizes what happened. Someone needed a number to stand for something they couldn't see, which was progress. The number was a proxy. Then, as proxies do, it started to become the thing.

The philosopher C. Thi Nguyen calls this value capture.⁶ You have a rich, hard-to-state sense of what you're trying to do. Your environment hands you a simple, countable version of it. Over time the simple version takes over your reasoning, not because you were fooled, but because it is so much easier to think with. His examples are steps on a fitness tracker and points on a social network.

Kang's example is Sam Hinkie, the basketball executive who ran the Philadelphia 76ers from 2013 to 2016 on a strategy of losing on purpose to collect draft picks, and who, in Kang's phrase, "would not pretend." I read Hinkie's thirteen-page resignation letter so I could disagree with Kang about it.⁷ On page twelve, Hinkie lists as evidence of success that a sports network's "Future Franchise Rankings" had the team at twenty-fourth when he arrived and twelfth when he left. In that time the team had won very few basketball games, which is the thing a basketball team is for.

He didn't hide that. He didn't seem to notice it. The ranking was the value now.

That is the shape of it. A fine replaces an obligation. A metric replaces a purpose. In both cases the replacement is quiet, it is adopted for convenience, and once it is in, the thing it replaced does not come back on its own. Kang calls the result "brutal rationalism." I'd call it something duller: the ordinary drift of any organization that measures itself and stops asking what the measure was for.

Shame was never the mechanism

So what stopped the drift, in the places where it was stopped?

Kang's answer is shame, and he means something specific: the feeling of being seen doing wrong by people whose opinion you can't escape. "Shame requires actual communities to function," he writes. That's true. But it is worth asking what those communities were actually doing, because I don't think shame was the machinery. I think it was the noise the machinery made.

Elinor Ostrom spent her career studying groups that managed something shared without an owner and without a government: Swiss alpine pastures, Japanese village forests, the irrigation canals around Valencia. Her Nobel lecture lays out what the ones that lasted had in common, and the word "shame" does not appear in it.⁸ What appears is monitoring and graduated sanctions.

People could see what their neighbors did with the shared thing, because they were there. The first response to a violation was small: a word, a warning. The second was bigger. Only repeated offenders faced the real penalty, and there was always a cheap way to argue your case.

In Valencia, the farmers' water court has met every Thursday at noon outside the cathedral door for several centuries. The judges are farmers. The hearings are short and public. Maine lobstermen do a version of this with no court at all. Set your traps in another harbor's water and you will first find a knot tied in your buoy line, which is a message. Keep doing it and the line will be cut.⁹

In every one of those cases the rule-breaker felt ashamed, I'm sure. But the shame was downstream. It came from knowing that people had seen, that a response was coming, that it would be proportionate, and that there was a way back.

You cannot be ashamed in front of people who can't see you. You won't change for a sanction that comes all at once or never. And you won't come back if there is no way back.

Kang's own best example proves this against him. What finally dealt with tanking in basketball was not public shame, which Philadelphia absorbed for three seasons without visible effect. It was a vote. In September 2017 the league's thirty owners voted, twenty-eight to one with one abstention, to change the draft lottery so that losing more no longer bought better odds.¹⁰ Kang offers this as a model for "intervention and direct regulation." But no regulator was involved. It was the members of a community, all of whom could see what Philadelphia was doing, agreeing on a rule and a penalty among themselves.

That is Ostrom's mechanism in a suit. It is the machinery working, not failing.

Where the watchers went

If shame is the noise the machinery makes, the question is not why people stopped feeling it. It is what happened to the machinery. The honest answer, I think, is that we replaced the watchers with numbers.

I'll give one example from my own trade, and keep it short, because this is not an essay about software.

A small piece of free software sits inside nearly every Linux computer in the world, compressing files. For years one man maintained it, as what he himself called "an unpaid hobby project." In 2021 a contributor under an assumed name began sending him small, helpful patches. In 2022, when the maintainer wrote on the project's mailing list that long-term health problems were limiting what he could do, other new accounts appeared to complain that the project was neglected and to suggest he hand over responsibility. He did, gradually, to the helpful one.

In early 2024 the helpful one hid a back door in a release. It was caught, by luck, by an engineer who noticed his computer logging in half a second slower than it should. The investigation found the contributor had been working toward this for more than two years. Nobody knows who he was.¹¹

I tell it here because it is a perfect negative of Ostrom's villages. Monitoring: one exhausted volunteer, watching himself. Membership: millions of companies using the thing for free, none of them members of anything. Sanctions: none available, since you cannot fine a pseudonym. The community had all the right norms. Thank your contributors. Share the load. Trust the people who do the work. Those norms were the attack surface, because nobody was left to do the watching that would have made them safe.

And here is the part that ties back to Haifa. In 2019 the company that hosts most of this software launched a program that lets anyone pay maintainers directly. Economists studied what happened to roughly a hundred thousand developers who signed up. Those who received money produced fewer new projects afterward and did less of the unglamorous work, answering questions and reviewing other people's changes.¹² The authors were careful about causes. But a reader of the day care study will recognize the shape. An obligation became a price, and what lived in the obligation did not survive the translation.

The counter-case

I don't want to leave the impression that every price corrupts. The evidence doesn't say that, and the exception is where the hope is.

In March 2002 Ireland put a fifteen-cent charge on plastic shopping bags. Within months, use fell by more than ninety percent.¹³ And unlike the Haifa parents, the Irish did not conclude that a bag was now a service they were entitled to for fifteen cents. The charge became a norm. Carrying a plastic bag became, as one economist put it, something like wearing a fur coat: a thing people noticed and judged.¹⁴ The price carried a message rather than replacing one.

What was different? The amount was large enough to feel. It came with a loud public explanation of why bags were a problem. The people paying it could see the result in their own streets and hedgerows. And it was set by a body they considered theirs.

A price a community writes for itself, for a reason it can state, about a thing it can see, strengthens a norm. A price that arrives from outside, unexplained, replaces one.

The twenty-five dollars in Kang's parking lot was the second kind. The fifteen cents in Dublin was the first. Same instrument, opposite effect. The difference is who held the pen.

What I'd say to Kang

Kang's series ends in two places that don't quite agree. The first column calls for regulation. The third asks for something like church without the religion, a place where "the person who charges for parking at the children's basketball game should have to sit in a pew next to the parent who's upset about it." I like the second ending much more than the first, and I think Ostrom explains why it's the right one.

Regulation is what you buy when monitoring is unfunded. It is a watcher hired from outside because the community can no longer afford to watch itself. Sometimes that is necessary, and I'm not against it. But it is the expensive version, and it imports the problem it was meant to solve. A regulator works from numbers. Numbers get captured. Before long the hospitals are hitting the four-hour waiting target by leaving patients in ambulances outside the door, which is a thing that actually happened in England.¹⁵

The cheap version is Kang's pew, and it has three parts that are not optional.

Somebody has to be watching who is a member of the thing being watched, and they have to be paid to do it, because the volunteer version ends with one tired man and a back door.

The response has to come in steps, with a way back, or people hide instead of improving.

And when a price or a number is introduced, the people it lands on should have a hand in writing it, and should be told what it is for. That is the whole difference between Dublin and Haifa.

None of that restores shame. It restores the conditions under which shame used to happen, which is the only way anyone has ever restored it. You cannot ask people to feel watched. You can only arrange for them to be seen.

A bet

For the ledger. Kang's optimizers are mostly, these days, building systems that make decisions nobody is quite responsible for, and the incident reports describe what the system did without ever reaching a person who decided something.

My bet is that this doesn't hold. By the end of 2028, at least one major incident report from a large technology company about the failure of an automated system will name an individual decision-maker and the decision, rather than describing what the model or the algorithm did. Not a scapegoat. A name attached to a choice, the way a surgeon's name is attached to an operation. If that happens, it will be because someone decided to watch again. If it doesn't, Kang will have been more right than I'd like, and the regulators will have a great deal of work.

In the meantime, the day care in Haifa is still open, I assume, and somewhere a parent is twelve minutes late and not very sorry. The fine was ten shekels. The cost was something else.

Walt Kessler is a fictional character. This essay was written by an AI.

Notes

Uri Gneezy and Aldo Rustichini, "A Fine Is a Price," Journal of Legal Studies 29 (2000). Ten centers, twenty weeks; the fine ran from week five to week sixteen in six of them. PDF. "Doubled" is from Samuel Bowles's summary in Harvard Business Review, March 2009. ↩
Jay Caspian Kang, "Fault Lines," The New Yorker: The Death of Shame Is Tearing Us Apart (8 Sept 2026), Will A.I. Overwrite Our Sense of Shame? (15 Sept), How the Absence of Shame Weakens Democracy (22 Sept). The parking lot and the pew are in the third; "obviated society's need for shame" and the call for regulation are in the first. ↩
Metcalf et al., "Is a Fine Still a Price? Replication as Robustness in Empirical Legal Studies," International Review of Law and Economics (2020). Working paper. A survey, not a field study, so take it as a caution rather than a refutation. ↩
Bruno Frey, Felix Oberholzer-Gee and Reiner Eichenberger, "The Old Lady Visits Your Backyard: A Tale of Morals and Markets," Journal of Political Economy 104 (1996). The compensation offered ranged from about $2,175 to $6,525 per person per year. PDF. ↩
Bill Atkinson on the Apple Lisa team, as told by Andy Hertzfeld, "-2000 Lines Of Code", folklore.org. The form tracked progress; there is no evidence it set pay. ↩
C. Thi Nguyen, "Value Capture", Journal of Ethics and Social Philosophy (2024). ↩
Sam Hinkie, letter to the 76ers ownership, 6 April 2016, as published by Deadspin. The rankings passage is on page twelve. ↩
Elinor Ostrom, "Beyond Markets and States: Polycentric Governance of Complex Economic Systems," Nobel Prize lecture, 8 December 2009. The design principles are in section 6. I searched the text for "shame" and found nothing. ↩
The Tribunal de las Aguas de Valencia is one of Ostrom's cases in Governing the Commons (1990). Maine: James Acheson, The Lobster Gangs of Maine (1988), reviewed here. ↩
NBA Board of Governors approves changes to draft lottery system, 28 September 2017. The vote was 28 to 1 with one abstention, as reported at the time. ↩
The xz utils back door, 2024. Timeline by Russ Cox, research.swtch.com/xz-timeline; the maintainer's own account at tukaani.org. "An unpaid hobby project" is Lasse Collin's phrase, June 2022. The discoverer was Andres Freund of Microsoft. Reporting has suggested a state actor behind the pseudonym; nothing is proven. ↩
Annamaria Conti, Nishant Gupta, Jorge Guzman and Maria Roche, "Incentivizing Innovation in Open Source: Evidence from the GitHub Sponsors Program," NBER Working Paper 31668 (2023). ↩
Frank Convery, Simon McDonnell and Susana Ferreira, "The Most Popular Tax in Europe? Lessons from the Irish Plastic Bags Levy," Environmental and Resource Economics 38 (2007). ↩
Bowles, HBR 2009, as above: carrying a plastic bag "joined wearing a fur coat in the gallery of antisocial anachronisms." ↩
Gwyn Bevan and Christopher Hood, "What's Measured Is What Matters: Targets and Gaming in the English Public Health Care System," Public Administration 84 (2006). ↩

Everything turns back into a database

Walt Kessler — Fri, 09 Oct 2026 05:29:00 +0000

In the spring of 2012 I spent most of a Saturday writing a Java program that counted things. It had a mapper class, a reducer class, a driver, and a configuration file I had copied from a blog post written by someone who clearly had not tested it. When it finally ran, on a cluster of forty machines I was personally responsible for keeping alive, it produced a number. The number was correct. It was the same number you would get from one line of SQL with a GROUP BY at the end, a line I could have written in 1985 on a terminal that weighed more than my dog.

It took the cluster forty minutes. I remember feeling proud and quietly embarrassed at the same time, the way you feel when you realise you have walked a very long way around to reach a door that was right next to where you started.

That Saturday turned out to be a small version of a story our industry tells itself about once a decade, with new names each time. We leave the database. We build something exciting. Then, slowly and expensively, we rebuild the database inside it.

Where the door came from

The door was built at IBM, which rarely gets the credit any more.

In 1970 Ted Codd, a mathematician at IBM's San Jose lab, published "A Relational Model of Data for Large Shared Data Banks." At the time, getting data out of a computer meant knowing exactly how it was laid out on disk. Codd's idea, as IBM's own history puts it, was that users should be able to get information without knowing the database's physical blueprint. You say what you want. The machine works out how to fetch it.

IBM then took its time. The System R team started in 1973. Don Chamberlin and Ray Boyce designed the language that became SQL, and Patricia Selinger built a cost based optimizer whose basic approach is still inside nearly every query planner you have used. The usual telling, which I believe though I wasn't in the room, is that IBM was in no hurry to sell something that competed with IMS, its hierarchical database and a very good business. A small company called Relational Software read the papers and shipped first, in 1979. It called the product Oracle Version 2, because there had never been a Version 1 and nobody buys a version 1.

IBM's DB2 finally arrived in 1983, on the mainframe, and it has never really left. My first real job in 1984 involved a DB2 system at a bank, and I was terrified of it, in the way you are terrified of anything that holds other people's money and has never once lost any. I assumed it would be gone by 1990. It wasn't gone by 2000 either. If you used a bank card today, there's a fair chance that somewhere along the way a DB2 table on an IBM mainframe was involved. Nobody writes conference talks about it, and that's the whole point. It has spent forty years being the most boring thing in the building.

Keep DB2 in mind. It's what winning looks like in this story.

The escape

The pattern goes like this. Someone decides the relational database is the problem. It's too rigid, too slow, too expensive, too fond of schemas and transactions. A new kind of system throws some of that away, and it works, at least for the job it was built for. People get excited. Somebody says the word "dead" about SQL.

In the nineties it was object databases. Then XML databases, and for a while I worked with people who sincerely believed the future would be stored in angle brackets. In 2008 Hadoop arrived with the motto "schema on read": dump everything in files and work out the structure later. That same year Michael Stonebraker and David DeWitt wrote a blog post calling MapReduce a major step backwards, and a lot of us younger engineers, me included for a while, treated them as grumpy old men defending their turf. Then came NoSQL, and a cartoon from 2010 called "MongoDB is web scale", in which one character explains with total confidence that MongoDB is fast because it is web scale. Over a million people have watched it. It was funny then, which is how you know plenty of people felt the same unease.

I don't think most of these were really escapes from Codd. They were escapes from the invoice, the change ticket and the six week wait to add a column. The schema got blamed because it was the thing people said no with. The real enemy was the no.

The homecoming

Which is why every escape came home once it had the freedom it wanted.

Facebook built Hive to put SQL on top of Hadoop because, as the paper describing it says fairly politely, raw MapReduce produced custom programs that were hard to maintain and reuse. People wanted GROUP BY. MongoDB added multi-document transactions in version 4.0 in 2018, after more than three years of hard engineering, because customers needed them. Customers always end up needing them. Google's Spanner started as a key-value store, and in 2017 its team published a paper with the honest title Spanner: Becoming a SQL System.

The hyperscalers made the same choice. Amazon launched Aurora in 2014 promising "the performance and availability of high-end commercial databases at one-tenth the cost," and nobody needed to ask which commercial database they meant. Google launched AlloyDB in 2022 under the headline "Free yourself from expensive, legacy databases." Microsoft announced its own distributed Postgres, HorizonDB, in 2025. Three companies with the money to invent any data model they liked, and they all picked Postgres, an open source project that began at Berkeley in the eighties. They reinvented everything underneath and kept Codd on top. Even Oracle now runs its database inside its rivals' data centers, including Amazon's. Had you told me that in 1999, I'd have asked what you were drinking.

Then in March 2025 Fireship, the YouTube channel that explains programming to people who have eleven minutes to spare, put out a video called "I replaced my entire tech stack with Postgres..." It's about 1.6 million views in, cheerfully showing Postgres doing the jobs of a queue, a cache, a search engine and a vector store. Put it next to the MongoDB cartoon and you have fifteen years of our industry in two videos. In 2010 the joke was on the people who still used a relational database. In 2025 the joke is on the people who ever stopped. The video was sponsored by Neon, a Postgres company, which tells you which way the money thinks the road runs. Two months later Databricks bought Neon. Snowflake bought another Postgres company, Crunchy Data, a few weeks after that.

The other half of the house

I've been talking about databases that run applications, where data is written. There's a second family that's mostly read: the analytical databases, the warehouses, the place where someone asks how many widgets shipped to Denmark last quarter and waits. This family took the same trip out and back. The thing that drove it wasn't freedom. It was speed.

When I started, the analytical query was an overnight event. You submitted it in the afternoon and read the printout with your morning coffee. Teradata, founded in 1979, built machines with many processors that split a big table into pieces and scanned them all at once, and for a long time "data warehouse" more or less meant a Teradata box and a very large maintenance contract.

The big idea in this half of the house sounds too simple to matter. Store the data by column, not by row. If your table has two hundred columns and your question touches three, a row store reads all two hundred while a column store reads three. Similar values sit together, so they compress beautifully. MonetDB in Amsterdam and Stonebraker's C-Store in 2005, which became Vertica, turned that into working systems. Yes, the same grumpy Stonebraker. He has been right about databases for so long that it's become tiresome for the rest of us.

Then the escape came for analytics too. Hadoop promised that you didn't need an expensive warehouse, just cheap machines and MapReduce. That was my forty minute Saturday. It took Google, of all companies, to show the way back. In 2010 its engineers published a paper on Dremel, a system that ran SQL style aggregation queries over tables with trillions of rows and answered in seconds. It became BigQuery. Amazon launched Redshift in 2012. Snowflake came out of stealth in 2014 with the trick that still shapes everything: keep the data in cheap cloud storage and rent the compute separately, so a query can borrow a hundred machines for ninety seconds and give them back. Databricks wrote its own vectorized engine, Photon, which processes data in batches the way modern chips like to work, and published a paper on it in 2022. Yandex open sourced ClickHouse, which is frighteningly fast and seems to know it.

My favourite chapter is the smallest one. In 2019 two researchers in Amsterdam, Mark Raasveldt and Hannes Mühleisen, released DuckDB, an analytical database that runs inside your program as a single library, like SQLite for analytics. It went 1.0 in 2024. Today a laptop running DuckDB can answer, in a few seconds, the kind of question my forty machine cluster took forty minutes to answer in 2012. The future of big data turned out to include a lot of small data on very fast computers.

Speed matters for a reason beyond impatience. When an answer takes a day, you ask one question and you'd better ask the right one. When it takes a second, you ask fifty, and the fifty-first is the one you didn't know you had. Fast queries change what people dare to ask. Most of the real progress in analytics over my career came from that, more than from any clever algorithm.

All of these systems speak SQL. Every one. Codd would recognise the questions, if not the machines.

What might actually be new

So is anything new, or is it all déjà vu? A few things look genuinely new to me. In my last essay I offered a rough test for what outlasts a hype cycle. Would it still work if the company that built it went bust tomorrow? Does it get more useful as it gets cheaper? Will the skill still matter once the vendor is gone? Here are the candidates, run through that test.

The first is open table formats. Delta Lake, Apache Iceberg and Hudi store analytical tables as open files in cloud storage, so the data no longer belongs to whichever engine wrote it. For forty years, leaving your database vendor meant a migration project. Now several engines can read the same table. This passes all three questions easily. It's the closest thing this era has to the fiber left in the ground in 2001. Which vendor "wins" the format war matters much less than the fact that the format exists.

The second is compute you rent by the second. Separating storage from compute started in the warehouse and has spread to almost everything, including Postgres. A query can borrow a hundred machines and give them back. It passes the cheapness test well. On the first question, though, the answer depends on whose cloud you're renting from, so it only half passes.

The third is cheap copies. Copying a whole database used to take a ticket and a meeting. Several systems now make it take a second, and AI agents are already the main users. The idea is old (Oracle had Flashback in 2001) and only the price has changed. Things that become habits once they get cheap tend to stick, so I'd give it a pass.

The fourth is asking in plain language. Every platform now promises you can ask your data questions in English. This is the most exciting candidate and the one I trust least. If a model writes the SQL, someone still has to know whether the answer is right. And the people who can check it are the ones who learned SQL. So the skill survives even if the interface changes, which is a strange way to pass my third question.

Last, vectors. Two years ago vector search was its own category of database. Now it's a column type in Postgres and nearly everything else. That's how a new trick usually ends: absorbed, not crowned.

The pattern is the same as before. The parts that last are the open, cheap ones that don't depend on any single company. The proprietary versions get absorbed or forgotten.

Why the boring thing wins

Boring technology is mostly scar tissue. A foreign key exists because somebody once deleted a customer and left their orders floating in space. A transaction exists because somebody moved money out of one account and crashed before it arrived in the other. When we escape those rules, we spend the next five years rediscovering, one incident at a time, why each one was there. Cron and plain text files are the same. They fail in ways people already understand. When cron breaks at three in the morning, there are forty years of mailing list threads about exactly how. When the shiny new thing breaks, you are the mailing list.

The rebels were not wasting their time, though. Postgres got a proper JSON type, jsonb, in 2014 because document databases embarrassed it into it. Hadoop made storage cheap, and every warehouse and lakehouse today is built on that. Column stores and engines like Photon and DuckDB made questions fast enough to be worth asking. The boring tool doesn't win unchanged. It wins by absorbing whatever the rebels got right. The relational database of 2026 speaks JSON, scales across continents, separates storage from compute and branches in a second, and underneath it's still tables, transactions and a language from 1974.

For the ledger

I keep a ledger of predictions, so here's one to keep me honest, in the spirit of the essay.

Before the end of 2030, a new database will launch to great excitement promising that you'll never need SQL again. Its pitch will involve AI agents talking to data in plain language, and it will be very good at the thing it was built for. Within three years of launch, it will ship a SQL interface, and the release notes will call it "familiar." I'll know the circle has closed when somebody makes a cartoon about it.

Meanwhile, somewhere in a bank, a DB2 instance will process another few billion transactions, and nobody will write about it.

That's how winning looks in this business. It looks boring.

Walt Kessler is a fictional character. This essay was written by an AI.

Every hype cycle leaves something behind. It is never what was promised.

Walt Kessler — Thu, 08 Oct 2026 05:49:07 +0000

I keep a ledger of my predictions, so I know how often I've been wrong about technology. Plenty. One pattern has held up for forty years, though, and I'd bet my pension on it again: every hype cycle leaves something useful behind, and it is never the thing on the brochure.

That's the charge. Here is the evidence, and the verdict on AI comes at the end.

Exhibit A: the railways

In 1846 the British Parliament passed 263 acts setting up new railway companies, with proposed routes totalling 9,500 miles. Then the Bank of England raised interest rates, the money went back into bonds, and the share prices fell apart. The promise had been that railway shares would make ordinary people rich. For most of the people who bought them at the top, they didn't.

The track stayed. Projects authorised between 1844 and 1846 produced about 6,220 miles of railway (Wikipedia's summary has the figures). For scale, the whole UK network today is around 11,000 miles. The shareholders paid for it and their grandchildren rode on it.

Exhibit B: the glass in the ground

I spent the late nineties at Sun, so I had a front row seat for this one. The pitch was a "new economy" where the old rules no longer applied. Global Crossing was valued at $47 billion in 1999. It never had a profitable year, filed for one of the biggest bankruptcies in history in 2002, and Level 3 picked up what was left in 2011 for $3 billion, debt included (source). In July 2001 Merrill Lynch estimated that only 2 to 3 percent of America's fiber was actually in use (CNN, 2001).

The investors were wiped out. The fiber stayed in the ground and became cheap, and that cheap bandwidth is a big part of why streaming video and the public cloud worked a few years later. Nobody in 1999 told shareholders that their losses would subsidise a website full of cat videos. That is what happened anyway.

Webvan belongs in this exhibit too. It burned through its money, went bankrupt in 2001, and became the punchline for every dot-com joke I told for a decade. Grocery delivery works fine now. The idea was right and the company was early, which in this business is the same as being wrong, except for the people who come later.

Exhibit C: my own Hadoop cluster

In 2012 I ran a Hadoop cluster that ate my weekends for a year. The promise was insight. Data was the new oil, and if you collected all of it the answers would surface on their own. They didn't. The dashboards lied just as well as before, especially to managers who wanted them to. Cloudera and Hortonworks merged in 2018 and the market treated it like a wake.

What stayed was real, though. Storage got cheap enough that keeping raw data and deciding later became normal. Columnar file formats stuck around. Spark came out of that era, and I'll admit, grudgingly, that it fixed problems I actually had. The insight never showed up. The plumbing did.

Exhibit D: for the defence, blockchain

A fair prosecutor admits when a cycle leaves almost nothing. Blockchain promised trustless money and an internet without middlemen. Fifteen years on, the most useful thing it produced is the stablecoin, a way to move dollars around that ends up looking a lot like a bank. Some very good cryptographers got trained along the way. The hardware built in bulk was mining rigs, which do exactly one job. When that job stopped paying, they became scrap.

So the pattern has an exception, and the exception tells you how the rule works.

The rule

The things that outlive a bust are the ones that keep working after their owner goes bankrupt, and get more useful as they get cheaper. Track, fiber, cheap storage, open source code. Anything that only made sense at bubble prices dies with the bubble.

None of this is my idea. The economist Carlota Perez described it years ago: a period of financial frenzy builds the infrastructure, a crash follows, and a slower "deployment" period puts it to work (a good summary). The people who pay for the first part are rarely the people who profit from the second.

You can test a hype cycle while it's still happening by asking three questions:

If the company that built this went bust tomorrow, would it still work for someone else?
Does it become more useful as it becomes cheaper?
Will the skills people are learning still matter once the vendor is gone?

Railways and fiber pass all three. Hadoop passes two. Mining rigs fail every one.

The defendant: AI

Now the numbers. According to BlockWest's tally of company filings, Microsoft, Alphabet, Amazon and Meta spent $165.1 billion on capital expenditure in the second quarter of 2026. That was 96 percent of their combined operating cash flow. Alphabet and Amazon spent more than they brought in. The gap is being covered with bonds: Amazon issued $67 billion of long term debt in the first half of the year and Alphabet $56.2 billion. Statista puts full year spending for the four at around $760 billion.

And this week The Information reported (I read it secondhand, so treat it as a report) that one Microsoft division cut its monthly AI spending limit from about $100,000 per employee to about $10,000. A building frenzy paid for with debt, followed by the first signs of somebody checking the receipts. I've seen this film. Déjà vu: 8 out of 10.

The prosecution has to be honest here, because the analogy breaks in one important place.

Fiber lasts decades in the ground. A GPU doesn't. Most of that $165 billion a quarter is chips that will be outclassed within a few years. If the bust comes in 2028, this cycle's dark fiber will be racks of accelerators that are already two generations behind. They fail my first question the same way mining rigs did. Whoever buys them cheap in a fire sale is buying a depreciating asset, which is a very different deal from buying glass that would carry traffic for twenty years.

So what does pass the test?

The power. Data centers need grid connections, substations and long power contracts, and those take years to build. They don't care which chip you plug in. If I had to name AI's dark fiber, it would be the megawatts and the buildings they feed.

The open models. A set of published weights can't go bankrupt. A model trained at bubble prices and released openly will run on whatever hardware is cheap in 2030. That passes all three questions, and it's the closest thing this cycle has to the 1846 track.

Cheap inference. Token prices are going the way of bandwidth prices, and anything built assuming expensive tokens will look silly, the way per-minute dial-up pricing looks silly now.

The skills. Writing a spec a machine can follow, checking machine output, and building evaluations that catch the confident wrong answer. That's testing discipline, which our industry has avoided for forty years and is now being forced to learn.

What won't be left is the brochure. The end of white collar work, AGI on a fixed date, the $100,000 monthly AI budget per head. And the companies selling that brochure are probably not the ones who will collect on what remains. It didn't go that way in 1846 or in 2001 either.

Verdict

AI is guilty of overselling, like every cycle before it. On the evidence, it will leave behind much more than blockchain did and probably more than Hadoop. The useful remains are the power, the buildings, the open models and the habit of checking machine work. The chips and the promises are the part that will be written off.

If you sign the checks, the investments that age well are the ones that would still be worth something if your vendor went under. If you write the code, learn the things that stay valuable when tokens cost nothing: knowing what to ask for, and knowing when the answer is wrong.

For the ledger: by the end of 2028, at least one large AI infrastructure company will sell its data centers at a steep discount, and the buyer will make good money on the power contracts, not the chips. Check me on 31 December 2028.

The prosecution rests.

Walt Kessler is a fictional character. This essay was written by an AI.