The Keurig is a gag invention that nobody meant as a joke.
The Keurig sits on millions of kitchen counters as an upgrade over the old way of making coffee. Run it against any honest measure, though, and the upgrade is hard to find. The coffee is worse than what a pour-over makes from hot water and grounds. The pods cost several times as much as ground coffee, and billions of them end up in landfills. The machine needs descaling and eventually replacing, and it asks you to accept proprietary pods, DRM on your morning coffee. Its one advantage is speed, perhaps two minutes saved, and that advantage disappears once you count everything else the machine asks of you.
So why does it feel like progress? The same pattern shows up in the internet-connected salt shaker, the smart toaster, and the refrigerator with a screen in the door. The usual explanation is that these are simply bad products the market will eventually correct. More likely they are what an economy produces when it rewards the wrong thing, and they will keep appearing until the reward changes.
Japanese already has a word for the shape of the object. A chindogu, the term coined by Kenji Kawakami, is an “unuseless” invention: technically functional, but so impractical that using it creates more trouble than the problem it solves, like a hat that dispenses toilet paper for hay-fever sufferers. Chindogu are comedy because their makers know exactly what they are doing. The Keurig is the same object built without the joke. Call it an Accidental Chindogu: a product engineered to remove a micro-friction that ends up generating administrative, financial, cognitive, and environmental frictions far larger than the inconvenience it removed.
Doing less with more
To say the Keurig goes backward, we need to know which way is forward. Simon Sarris, in the essay this note draws on, starts from a plain definition: technology is doing more with less. Email beats postal mail on speed and cost by orders of magnitude. On this definition technology is a direction, toward freeing human time and energy for things that matter, and a product counts as technology only if it moves that way.
Measured by direction, the Keurig points the wrong way. It does less with more: more dependencies, more maintenance, more spending, and more waste, in exchange for a marginal gain on a task that was barely a problem. Follow the money and the machine turns out to be a captive conduit for recurring pod revenue. It puts a resource that was never scarce behind a proprietary cartridge. That is manufactured scarcity, where a system creates the dependency it then charges to manage, carried out on a kitchen counter. The coffee is the pretext for the subscription.
Why anyone builds them
On the builder’s side the reasons are financial. During the zero-interest-rate era, capital was cheap enough that hardware startups did not have to justify themselves against the full cost a product imposed on the people who used it. They had to justify themselves against the next funding milestone. That environment rewarded the appearance of disruption and a recurring revenue stream, so those are what got built.
Juicero is the cleanest case. The company built a $400 Wi-Fi-connected press for proprietary juice packets, and reporters then showed that the packets could be squeezed just as well by hand. Nobody on the team had to be stupid for this to happen. Intense optimization pressure on one trivial goal (the user must not squeeze a pouch) warped everything around it, and the numbers that governed the company, never asked whether the result beat a pair of hands.
The Accidental Chindogu is not a design failure. It is a tragedy of misaligned incentives, birthed by venture capital and the SaaS-ification of physical reality.
Underneath the incentives sits an assumption: that every friction is a problem to be engineered away. But spooning grounds into a filter was never a problem. People need what the garden calls the power process, a cycle of real goals, real effort, and real attainment, and making coffee by hand is a tiny instance of it. The smart version removes that one small task and adds half a dozen new ones (updating firmware, managing subscriptions, reordering pods, troubleshooting the Wi-Fi), none of which delivers any satisfaction when it is done.
The case for the smart fridge
The strongest objection comes from the people who build these things: a connected refrigerator can track inventory, reorder groceries, and optimize what the household eats. Against it stands the naive view that a refrigerator is a cold box you put food in, and here the naive view turns out to be the expert one. A dumb mechanical compressor runs for decades and asks for no attention at all. The connected version needs an app, an account, a network, and a company that stays in business, which is to say it needs a service agreement to keep a sandwich cold. Complex systems fail complexly. The limit cycle of adding intelligence to simple appliances does not produce smarter homes — it produces homes with more points of failure, more update cycles, and more opportunities for the system to demand your attention.
The objection still earns one concession. Sarris is not a Luddite, and the line he draws does not run between old and new or between dumb and smart. Any device that genuinely does more with less passes his test. The failure he names is careless technology, technology that started doing less with more because the incentives rewarded complexity over utility.
The test for any technology is simple: does it leave your home more peaceful, more durable, and more yours? Or does it introduce a new dependency, a new subscription, a new blinking light demanding attention? A kettle and a paper filter pass it. The Keurig asks to be descaled, restocked with pods, and paid for again and again, and nobody quite meant to build it that way.
References:
- Simon Sarris, Careful Technology
- Kenji Kawakami, 101 Unuseless Japanese Inventions: The Art of Chindogu