Why Standards Are the Quiet Inheritance of Civilization

If you assemble a piece of furniture from a different country, the screws will fit. If you plug a phone charger into an unfamiliar wall, it will charge. If you load a shipping container in Shanghai, it will fit on a chassis in Long Beach. None of these compatibilities is accidental, and almost all of them were contested before they were boring. The history of standards is the history of fights that ended so completely that we forgot they had happened.

The screw thread war

For most of the nineteenth century, every machine shop made its own screws, and they did not interchange. A bolt from Manchester would not thread into a nut from Birmingham, let alone one from Philadelphia. The first attempt to fix this was Joseph Whitworth's 1841 paper proposing a uniform thread angle of 55 degrees, with a defined pitch for every diameter. British industry adopted it within a decade. The Americans, however, did not.

William Sellers proposed a different standard in 1864, using a 60-degree thread angle, easier to manufacture with then-current cutters. The United States adopted Sellers; the British Empire kept Whitworth; the rest of the industrialized world picked one or the other based on whose machine tools they had imported. The result was that for most of the twentieth century, an American mechanic and a British mechanic working on the same engine could not share fasteners.

The metric ISO thread standard, finalized in 1947, was the third option that eventually won. It took until the 1970s for the British military to formally retire Whitworth, and aerospace and railroad systems still contain Whitworth threads today, because legacy infrastructure is too expensive to retrofit. The fight was settled, but the fossil record remains.

The gauge war that built two countries

Railway track gauges (the distance between rails) became a national identity question in the nineteenth century. Britain settled on 4 feet 8.5 inches, an oddly precise number that traces back to George Stephenson's preference, which itself traces back to the wheel-ruts of horse-drawn coal carts in Northumberland. Most of Europe and North America followed.

Spain, fearing a French invasion, deliberately chose a different gauge so that French troop trains could not roll across the border. Russia did the same. Australia adopted three different gauges in three different colonies, because each colonial government chose its preferred British exporter, and the resulting "gauge break" at colonial boundaries cost the country decades and substantial money to unify. The choice of standard, in other words, was a strategic decision about who you were willing to be interoperable with.

The container that ate the docks

Until the 1950s, cargo was loaded by hand. Longshoremen swung crates, sacks, and barrels off ships, into warehouses, onto trucks. A 10,000-ton ship took a week to unload, employing hundreds of stevedores. Cargo theft was a major operating expense; spoilage was routine; the cost of moving goods across an ocean was dominated by the cost of getting them on and off the ships at either end.

Malcom McLean, an American trucker, had the now-obvious idea: standardize the box. He converted a tanker, the Ideal X, to carry 58 metal boxes, and sailed it from Newark to Houston in April 1956. The unloading time at Houston was less than eight hours.

The hard part was not the engineering; it was the standardization. McLean's boxes were 33 feet long. Other operators built 35-foot boxes, 27-foot boxes, 20-foot boxes. Cranes had to be redesigned for each. Trucks had to be redesigned for each. For most of the 1960s the industry was paralyzed by the prospect of converging on a single size, because whoever's existing fleet was on the wrong side of the decision would be writing off enormous capital.

The ISO standardized container, in two variants of 20 and 40 feet, was finalized in 1968 after years of negotiation. The decision changed the geography of the world economy. Within twenty years, a journey that had once cost roughly half the value of the cargo cost less than one percent of the value. Factory locations, port cities, and entire industries reorganized around the new economics. Today's globalized supply chain is what happened after the box won.

The plug that did not

Walk through a hardware store in twenty different countries and you will find roughly fifteen mutually incompatible electrical plug shapes. The British BS 1363 (three rectangular pins), the European Schuko (two round pins with side grounding), the American NEMA (two flat blades), the Australian I-pattern, the Italian L-pattern, the Israeli H-pattern, the Swiss J-pattern. Each was settled in a regional process during the early twentieth century, when the cost of converging was high and the benefit looked low because international travel was rare.

By the time the cost of not converging became obvious (every traveler in the world owns adapters), the standards were entrenched in housing stock measured in centuries. There is no realistic path to a single global plug. The IEC has tried multiple times. The most they could agree on was IEC 60906-1 (a Brazilian-influenced N-pattern) which only Brazil ever adopted. The plugs are an example of a standard fight that never finished, and probably never will.

The internet's improbable coherence

The most surprising standardization story of the late twentieth century is that the internet has any standards at all. Telephone networks were standardized through state monopolies and treaty organizations like the ITU, with formal, slow, hierarchical processes. The internet was standardized by the IETF, which has no formal authority and operates by "rough consensus and running code." Anyone can submit a draft. Standards (the RFC series) become standards because people implement them, not because anyone decreed it.

The fact that this produced TCP/IP, HTTP, DNS, SMTP, and TLS, all of which actually interoperate at internet scale, is one of the more remarkable institutional achievements of the era. The contrast with the OSI standards, designed by ITU committees as the "official" successor to TCP/IP and widely deployed nowhere, is instructive. The slow, official process produced a beautifully complete specification that nobody implemented. The chaotic, code-first process produced messy specifications that everyone implemented. The standards that win are the ones that are easier to use than to ignore.

The shape of the inheritance

Standards are not the absence of disagreement; they are disagreements that have been settled long enough that the settlement looks natural. Every shipping container, screw thread, electrical socket, and protocol you use today is a fossil of an argument someone fought hard to win. The strange thing is how invisible this becomes once it is settled. We notice the wall plugs only when we travel; we notice screw threads only when something breaks; we notice the protocol stack only when it fails.

The quiet, accumulated, often-unloved work of standardization is one of the largest gifts each generation of engineers leaves the next. It is a very long-running form of cooperation, and it is the reason the modern world fits together at all. Most of it was decided by people whose names are not in any history book, in committee rooms whose minutes are forgotten. It is worth the occasional moment of recognition.