Built to Last, Not to Last Long Enough
My car has no computer in it. Not "an old computer" — none. Pop the hood on my Lada Niva and you get a carburetor, a distributor, a battery, and a bunch of things you can point at and say "that part does that job." When something goes wrong, the car tells you where to look, because there's almost nowhere to look. I have fixed that Niva in a parking lot, on the side of a trail, and once in my own driveway with a socket set and a flashlight held in my teeth. No dealership software, no proprietary scan tool, no error code that only means something to a technician with a subscription.
I spend my working life around some of the most complex software humans have built. Maybe that's exactly why I find something clarifying about a machine that can't lie to you, can't hide its logic behind a locked module, and can't be rendered a brick by a firmware update it never asked for.
The Default Has Flipped
Somewhere in the last few decades, the default response to "this is broken" quietly changed from "let's fix it" to "let's replace it." It happened gradually enough that most people never noticed the flip. A toaster used to be worth opening up. Now the estimate for fixing a lot of things — headphones, washing machines, increasingly cars — routinely comes back higher than replacing them outright, and that isn't always an accident of economics. Sometimes it's the design intent.
That's the part that should bother us more. It's not that manufacturing got so efficient that new is genuinely cheaper than repair. It's that products are often engineered, deliberately, to make repair harder: glued-shut enclosures, batteries that aren't meant to come out, parts that only talk to other parts from the same manufacturer, diagnostics that require proprietary tools nobody outside an authorized shop can own. That's planned obsolescence with better PR. It doesn't announce itself as "throw this away and buy another one" — it just makes the alternative annoying enough that most people give up.
A Generation That Never Learned
The knock-on effect is generational. A lot of people my age and younger never watched a parent fix a carburetor or re-wire a lamp, because there wasn't much left worth teaching — the object was too sealed to open, or cheaper to bin than to bother with. Skills that used to pass down almost by osmosis, just from growing up around people who fixed things, have quietly stopped transmitting. It's not that people got lazier. The objects around them stopped inviting the attempt.
That matters beyond the garage. Fixing something teaches you that systems are knowable — that a problem has a cause, the cause can be found, and the machine will make sense if you look hard enough. That's a useful disposition, and it's one "buy a new one" quietly trains out of you.
Cars Are the Clearest Case
Cars show this shift starkly, because it happened within living memory and the stakes are higher than a broken toaster. A car from the '80s or earlier is a set of mechanical systems you can trace by hand: fuel goes here, spark happens there, this cable pulls that lever. A modern car is a rolling network of dozens of computers, many locked down so that even independent mechanics, not just weekend tinkerers, need manufacturer-approved tools and licenses to do what used to be a Sunday afternoon job. The right-to-repair movement exists because this stopped being hypothetical: people who own their cars outright are still barred from fully diagnosing them without paying the manufacturer for the privilege.
None of this is nostalgia for discomfort. Old cars rusted, guzzled fuel, and would happily kill you in a crash a modern car would shrug off — I'm not pretending otherwise every time I'm out in the Niva on a trail with mud past the axles. The argument isn't that less technology is automatically better. It's that a specific kind of complexity — added for lock-in rather than function, the sensor whose main job is justifying a dealership visit rather than improving the drive — has costs we don't usually count: landfills quietly filling with electronics that had years of life left, skills nobody bothers learning anymore, things discarded not because they're worn out but because understanding them was engineered to cost more than replacing them.
What the Alternative Looks Like
The Niva isn't a rebellion, it's just older than the shift. It represents a philosophy that used to be the default and now has to be actively chosen: build the thing so a normal person, with basic tools and patience, can find what's wrong and fix that specific thing — instead of replacing the whole unit because nobody, including the owner, is allowed to know what's inside. That philosophy shows up anywhere someone keeps something alive past its expected life: a cast-iron pan you re-season instead of tossing, a fermentation crock that's decades old and improves with use rather than degrading. None of it is efficient by the metrics that got us here. All of it assumes the object is yours to understand, not just to own.
That's really the whole argument. Repairability isn't a nice-to-have bolted onto a product — it's a statement about who the object belongs to. Something you can fix belongs to you. Something you can only replace belongs, in every way that matters, to whoever decided you couldn't.