Daniel's been digging through the ISO catalogue and wants to know what's lurking in the genuinely weird corners. Not the standards everyone knows — paper sizes, date formats — but the ones where you read the title and think, someone spent years of their life on this. He wants the most ludicrously specific standards ever published, what they cover, how they came to exist, and whether they're actually sensible once you understand the context.
Oh, this is my kind of tour.
I knew it would be. He specifically asked for standards for bizarre objects, tiny procedural details, improbable workplace situations, oddly precise terminology, and things that appear to have been standardized purely because two countries once disagreed about them. So we're going to the point where rational coordination and administrative obsession become indistinguishable.
And that point is closer than most people think. Let me set the stage before we dive into the really good ones. ISO isn't a government body — it's a network of national standards organizations, one per country, and they've published over twenty-five thousand standards. Everything from screw threads to social responsibility guidelines. The way a standard happens is: a problem emerges, usually because some industry needs things to interoperate across borders. A technical committee forms. Drafts circulate. Countries vote. And sometimes — sometimes — years later, a document emerges that specifies exactly how to brew a cup of tea.
Which is where we're starting.
ISO 3103. And I want to be precise about what it actually says, because this is the one everyone misrepresents. Two grams of tea per hundred milliliters of water. Water at ninety degrees Celsius — not boiling, ninety exactly. Steep for six minutes. Then pour into a white porcelain bowl.
A white porcelain bowl.
White porcelain specifically. The standard says the bowl has to be white, or alternatively, the tasting room has to be lit such that the color of the bowl doesn't interfere with the taster's perception of the liquor.
The liquor.
That's the term they use for the brewed tea. I'm not making this up.
So someone, in a meeting, said "we need to specify the bowl color" and nobody in the room said "maybe we're overthinking this."
But here's the thing — nobody in the room said that because they weren't overthinking it. This is a test method standard. It's not telling you how to make tea at home. It exists so that tea tasters in Sri Lanka and tea buyers in London can compare results on a level playing field. If you change the water temperature, the steeping time, or even the vessel color, you change what compounds extract and how the taster perceives the result. The standard removes every variable except the tea itself.
So it's basically a lab protocol that happens to involve a beverage.
And once you see that, the whole category makes sense. These standards exist to turn subjective judgments — taste, aroma, mouthfeel — into something repeatable and comparable. The comedy is in the gap between the mundane object and the weight of international consensus behind it.
The gap between "make a cup of tea" and "the tasting room shall be lit such that."
Right. And that pattern repeats. ISO 3591 — the wine-tasting glass. This one specifies a tulip-shaped glass with a stem, a bowl that holds exactly two hundred fifteen milliliters, and you only pour fifty milliliters for the actual tasting. The shape concentrates aromas toward the nose. The stem keeps your hand from warming the wine. The fifty-milliliter pour means every taster is working with the same volume, the same surface area, the same evaporation rate.
So a wine in Bordeaux scores the same as a wine in Buenos Aires because the glass is identical.
In theory. And the glass itself is — I've held one of these, they're oddly satisfying objects. The standard emerged from the French wine industry, which had been using similar glasses for decades, and eventually the international community said, fine, we'll codify what you're already doing.
Which raises a question that's going to come up again. How much of this is genuine coordination and how much is one country saying "our way is now the world's way"?
We'll come back to that. But first I want to follow the thread from pleasure to safety, because that's where things get serious. ISO 3874 — handling and securing of Series 1 freight containers. This standard specifies exactly how to lash a shipping container onto a ship. The angles of the lashing rods, the tension, the number of lashings per container.
Down to the degree?
Down to the degree. And the reason isn't bureaucratic obsession — it's that in the nineteen sixties, containers were falling off ships. Regularly. Killing people. Costing billions. Different shipping lines used different methods, different countries had different rules, and the result was cargo at the bottom of the ocean and crew members who didn't come home.
So what looks like a committee specifying the angle of a rod is actually a safety standard written in blood.
That's the phrase. And once the standard exists, it becomes invisible infrastructure. Every container ship in the world lashes cargo the same way. You never think about it because it never goes wrong — and it never goes wrong because the standard exists.
The sign of a successful standard is that you forget it's there.
Yes. And the sign of a failed one is a container floating in the North Atlantic.
But tea and wine and shipping containers — these are at least things you can touch. Daniel asked for the obscure corners. What about standards for things that don't exist as physical objects at all?
ISO 233. This one standardizes how to transliterate Arabic characters into Latin script.
Wait. There's an international standard for how to spell Arabic words in English letters?
Not just English — Latin script generally. And yes. Because in the early days of international library cataloging, France and the United Kingdom had completely different systems for doing this. A book by an Arabic author would be cataloged one way in Paris and a different way in London, and if you were trying to find it across borders, good luck.
So two countries couldn't agree on how to write down sounds, and the solution was a multi-year committee process.
And the result is infrastructure you interact with constantly without knowing it. Every time an Arabic name appears consistently across databases, library systems, passport offices — that's ISO 233 doing its work. You only notice it when it fails. When your name gets mangled on a boarding pass, that's a transliteration standard failing in real time.
The boarding pass as ISO violation.
Unironically, yes. And this is where the "ludicrous" label starts to break down. To a librarian in the nineteen seventies trying to catalog a collection that spans five scripts and twelve languages, ISO 233 wasn't ludicrous. It was essential. The comedy is in the collision between the mundane output — a string of letters — and the enormous machinery required to produce it consistently.
But some standards really are just... I mean, there's one for the shape of a banana, isn't there?
There is. And this one's interesting because it's not actually an ISO standard — it's a Codex Alimentarius standard, which is a joint FAO and WHO body. But it gets cited in ISO contexts constantly. The standard specifies minimum length, minimum diameter, and acceptable curvature for bananas sold internationally.
Acceptable curvature.
The banana must not be excessively curved. And before you laugh —
Too late.
— the reason is packing efficiency. A banana that's too curved doesn't fit in a standard box. It gets bruised. It spoils. It spoils the bananas next to it. The entire supply chain for a perishable good depends on predictable geometry.
So the curvature standard is a bruise-prevention standard disguised as a shape rule.
And bruise prevention is a food-waste standard, and food waste is a climate standard, and suddenly the banana curvature committee looks less like bureaucrats and more like supply-chain engineers.
I'm still going to laugh at "acceptable curvature."
Oh, absolutely. The laughter and the respect coexist. That's the whole point.
What about workplace situations? Daniel asked about improbable workplace scenarios that somehow got standardized.
ISO 45001 is the big one for occupational health and safety, but that's broad. The specific weirdness lives in the older standards. There's one — I'm trying to remember the number — that specifies the maximum force a human can exert on a lever in different positions. Standing, sitting, arms extended, arms close to the body.
Someone measured this.
Someone measured hundreds of people pulling levers in different positions and published the results as an international standard. And the reason is that if you're designing a machine — an aircraft control, a factory floor panel, an emergency shutoff — you need to know whether the operator can actually pull the lever hard enough in an emergency.
So it's not "how hard can a person pull a lever," it's "how hard can a person pull a lever when they're seated, reaching left, with their non-dominant hand, while panicking."
The standard probably doesn't specify the panicking part, but the safety factors account for it. And this connects to a whole family of ergonomics standards. There's one for the height of a computer monitor relative to the desk. There's one for the force required to press a button. There's one for the acceptable noise level of a keyboard.
The keyboard one feels personal.
Mechanical keyboard enthusiasts would riot if it were enforced. But in an open-plan office, that standard exists because keyboard noise is a measurable productivity drain on the people who aren't typing.
I want to go back to something you mentioned earlier — standards that exist because two countries disagreed. What's the best example of that?
The pallet standard. ISO 6780. It's a diplomatic document that lists six incompatible pallet sizes and essentially says "these all exist, good luck."
So the standard didn't actually standardize anything.
It standardized the disagreement. The United States uses a forty-by-forty-eight-inch pallet. Europe uses a twelve-hundred-by-eight-hundred-millimeter pallet. Japan uses something else. And none of them fit efficiently in the others' trucks and warehouses. The committee couldn't pick a winner because picking a winner would mean billions in retrofitting costs for the losers.
So the standard is a treaty that says "we agree to disagree, and here's a document listing the ways we disagree."
And that's not a failure, exactly. It's an acknowledgment that sometimes the coordination problem can't be solved by fiat. The standard at least makes the incompatibility legible. Before ISO 6780, you might not even know why your pallets didn't fit in a foreign warehouse. Now you know exactly why, and you can plan around it.
"Your shipment is incompatible with our logistics infrastructure per section four, paragraph two of the standard that was supposed to prevent this."
It's a beautiful thing.
Let's talk about the banana problem at scale. What's the strangest terminology standard you've encountered?
Oh, there's a whole vocabulary of standards for standards. ISO has a standard that defines what a standard is. It's meta-standardization. And within that, there are terms that sound absurd until you need them. "Normative reference" — a document you have to follow to comply with this one. "Informative reference" — a document that's helpful but not mandatory. The distinction matters enormously when you're doing compliance work.
But to anyone outside that world, it sounds like a committee spent three years defining the word "should."
They did. "Shall" versus "should" versus "may" — these are precisely defined in ISO directives. "Shall" means a requirement. "Should" means a recommendation. "May" means permission. And if you get them wrong in a standard, you've potentially created a legal obligation where you meant to create a suggestion.
So the difference between "shall" and "should" could be a lawsuit.
In a regulated industry, absolutely. If a standard says a safety system "shall" be tested weekly and you test it monthly, you're in violation. If it says "should," you might have a defense. Three letters, enormous consequences.
This is the part where the comedy starts to feel vertiginous. We're laughing at the tea standard, but the same machinery that produced the tea standard also produced the bolt that holds your airplane wing on.
And that's exactly the arc I wanted to trace. The same committee structure, the same voting process, the same obsessive attention to detail that gave us the white porcelain bowl also gave us ISO 9001 for quality management and ISO 13485 for medical device manufacturing. The ludicrous standards aren't a separate category — they're the visible tip of an enormous iceberg of coordination.
The iceberg being: everything that doesn't kill you because someone standardized it.
Right. And the standards that seem most absurd are often the ones where the stakes are lowest, so the specificity feels gratuitous. But the specificity is the method. It's not that the tea committee was unusually pedantic — it's that pedantry is the entire discipline.
I'm now imagining a world where the tea committee was casual about it. "Eh, steep it for a while. Use a cup. Whatever color."
And then Sri Lanka sues a buyer in London because their tea failed a quality test, and the buyer says "well we used a blue mug and boiling water," and the case collapses because there's no agreed method.
The standard as lawsuit prevention.
A huge amount of standardization is lawsuit prevention. Or more charitably, dispute prevention. When two parties in different countries with different legal systems need to agree on whether something is acceptable, they need a shared reference point. The standard is that reference point.
So the tea standard is a pre-nup for international trade.
That's... actually a really good way to put it.
I have my moments. What about standards that never should have existed? Not the ones that seem silly but are secretly sensible — the ones that are indefensible.
The challenge is that even the worst standards usually had a rationale at the time. But there are definitely ones that outlived their usefulness and nobody's bothered to kill them. ISO has a standard withdrawal process, but it's slow, and committees disband, and sometimes a standard just sits there for decades because no one has the institutional will to say "this is obsolete."
Give me an example.
There are standards for floppy disk formats. Still on the books. There's a standard for the dimensions of a microfiche card. There's a standard for — and this one's my favorite — the acoustic requirements for a telephone handset, originally written when handsets were physical objects you held to your ear and are now mostly irrelevant because everyone uses speakerphone or earbuds.
But the standard persists because removing it would require someone to fill out a form.
Multiple forms. Circulated to multiple countries. With a comment period. And nobody cares enough to do it, so the standard remains, a fossil of the analog age.
A fossil preserved in bureaucratic amber.
And that's a different kind of comedy — not the gap between trivial surface and serious purpose, but the gap between the standard's original urgency and its current irrelevance.
The telephone handset standard as a monument to a problem nobody has anymore.
But someone had it. Someone in nineteen seventy-five was furious that they couldn't hear their calls properly, and that fury became a committee, and the committee became a standard, and now we're stuck with it forever.
I find that oddly moving.
Me too. It's like finding a Roman aqueduct still technically in service.
Hilbert: I've been listening to this whole thing, and I think you're being too charitable.
Here we go.
Hilbert: The thesis you've been building is that these standards are sensible once you understand the context. And for some of them — the container lashing standard, sure. That saves lives. But the tea standard? It's a nineteen-seventies relic that persists because no one has the will to kill it. The wine glass standard is a French power play to enshrine their tasting culture as the global norm, and everyone else went along because the alternative was a standards war nobody wanted to fight.
The French power play part is hard to deny.
Hilbert: And the pallet standard — you said it yourself, it's a document that lists incompatible sizes. That's not coordination. That's a failure dressed up as a success. The ISO process is slow, it's opaque, and it's often captured by incumbent industries who use it to lock in their own technology and lock out competitors.
So your argument is that the sensible-sounding justifications are post-hoc rationalizations for what's essentially political inertia.
Hilbert: Partly. And partly that the "actually sensible" framing lets standards bodies off the hook. If every ludicrous standard is secretly brilliant, there's no pressure to clean house. The floppy disk standards, the microfiche dimensions, the telephone handset acoustics — these aren't infrastructure, they're clutter. And clutter has a cost. Someone, somewhere, is still checking compliance against a standard that hasn't mattered in thirty years.
I'll concede the clutter point. The withdrawal process is broken. But I'll push back on the tea and wine standards. The fact that France benefited from the wine glass standard doesn't make it wrong — the shape really does concentrate aromas better than a straight-sided glass. The physics works regardless of the politics.
Hilbert: The physics works, but the committee chose the French physics over other physics. There are other glass shapes that also concentrate aromas. The standard didn't pick the best one — it picked the French one. And now if you want to sell wine-tasting equipment internationally, you're making a French glass.
That's a fair hit. The standard encodes a specific cultural tradition as a universal requirement.
Hilbert: And the tea standard — six minutes, ninety degrees, white porcelain. That's an Indian and British tea-tasting tradition. A Chinese tea taster would use different parameters for many teas. The standard isn't universal — it's a particular tradition that won the committee vote.
But that's how standards work. Someone's tradition wins. The alternative is no standard, and then you can't compare anything across borders.
Hilbert: Or the alternative is a standard that acknowledges regional variation instead of pretending one way is the way. The pallet standard, for all its flaws, at least does that — it says "these six sizes exist, here they are." Why can't the tea standard say "here are three accepted methods depending on tea type and regional practice"?
That would be a more honest document.
It would. And ISO has moved somewhat in that direction — newer standards are more likely to include regional annexes and alternative methods. But the old ones are sticky.
Hilbert: That's my point. The stickiness is the problem. Once a standard exists, it's nearly impossible to change, even when it's clearly encoding one country's preferences as global rules. The machinery you're admiring is also a machinery for entrenching power.
So the question isn't whether standards are good or bad — it's whose standards, and who gets to write them.
Hilbert: Who pays the compliance costs. Small producers in developing countries who want to export tea have to buy the standard, read the standard, and build their process around a six-minute steep in a white porcelain bowl that has nothing to do with their local tea culture. That's a real barrier dressed up as a neutral technical specification.
I don't disagree with any of that. But I still think the alternative — a world without these standards — is worse. The container lashing standard may have been written by the big shipping lines, but it also means fewer containers fall on people's heads. The transliteration standard may have been a Franco-British compromise, but it also means Arabic authors have stable names in international databases.
Hilbert: Sure. I'm not arguing for zero standards. I'm arguing that the "ludicrous" ones deserve more skepticism than you've been giving them. Some of them are sensible. Some of them are sensible but outdated. And some of them were never sensible — they were political victories that got a number and a cover page.
The floppy disk format standard as a monument to a battle nobody remembers.
Hilbert: The fact that nobody remembers the battle doesn't make the monument any less real. Someone is still paying for its upkeep.
Where does that leave us? The tea standard is a test method that encodes a specific cultural tradition. The wine glass is French physics enshrined as universal physics. The container lashing is life-saving engineering. The pallet standard is a diplomatic treaty that failed to solve the problem it was created to solve. And the floppy disk standard is a ghost.
They were all produced by the same process. That's what I keep coming back to. The same voting rules, the same comment periods, the same committee structures. The process doesn't know whether it's producing a safety standard or a fossil. It just produces.
Which means the absurdity isn't in any particular standard — it's in the fact that we've built a global machinery for generating consensus documents, and that machinery is indiscriminate. It will standardize a life-saving bolt with the same gravity it applies to a tea bowl.
That's the comedy, right there. Not that the standards are silly, but that the machinery is so powerful and so earnest and so completely unable to distinguish between the urgent and the trivial.
The committee as a force of nature.
The committee as a glacier. It moves slowly, it grinds everything in its path, and it leaves behind a landscape of polished stones — some of them diamonds, some of them just rocks.
We're all living in that landscape, whether we know it or not.
Every time you drink tea from a cup that isn't white porcelain, you're violating an international standard. And that's a comforting thought, somehow.
It is. The standards are there if you need them. And if you don't, they're just... there. Waiting. In case someone, somewhere, needs to compare two teas and be absolutely certain the only difference is the leaves.
What's the next standard that'll seem ludicrous in fifty years? AI safety testing? Drone traffic management?
The standard for how to prompt a language model.
Oh, that's coming. You know it's coming. Someone's going to specify the exact format for a system prompt, and in forty years people will laugh at it the way we laugh at the tea standard.
"Two hundred tokens, temperature zero point seven, and the output shall be evaluated in a white interface."
I'd serve on that committee.
I know you would. This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop for making us sound like we know what we're talking about. If you've got a weird prompt you want us to tackle, send it to show at my weird prompts dot com.
We'll be back soon.
