The Pythagoras problem: the effort we throw away.

Somewhere today, a maths teacher stood up and taught the Pythagorean theorem. So did another, in the next town. And another, in the next country. On roughly the same week of the year, in something like a hundred thousand schools, a hundred thousand teachers each delivered the same idea — an idea that hasn't changed since about 500 BC. Next year they'll all teach it again. And the year after. For thirty years.

Hold that picture, because it is one of the most wasteful things in education, and almost nobody counts it.

Teaching Pythagoras properly isn't one hour at the front of a class. It's preparing it, delivering it across your sections, marking the work that comes back, and re-explaining it to the students who didn't get it the first time. Add it up and it's on the order of eight hours of skilled human effort to teach and assess that one topic, once, well.

Now take those eight hours and multiply by the teachers doing it. Roughly a hundred thousand of them, each spending their eight hours — call it 800,000 human-hours, every single year, spent re-producing one lesson. And again next year. Over the thirty years of a system's working life, that one theorem consumes something like twenty-four million hours of trained human time.

For a fact that was already true, and already perfectly explained, two and a half thousand years ago.

This wasn't always waste. For nearly all of history there was no way to move a lesson from one room to another. The teacher was, quite literally, the only copy of the knowledge in the room. If a child in a far-off town was to learn Pythagoras, a human being had to stand in that town and say it out loud. Local re-delivery wasn't inefficiency — it was the only option there was.

That constraint is gone. We can now copy and distribute the clearest possible explanation, perfectly and for almost nothing. The duplication that was once unavoidable is, for the first time, a waste we are simply choosing to keep paying — because the system's shape hasn't caught up to the fact that the constraint vanished.

Here's the part that turns waste into something worse. Those 800,000 hours don't purchase 100,000 good explanations. Some of those teachers explain Pythagoras beautifully. Most are ordinary. Some do it poorly, lose half the room, or get a step wrong. So the maximum possible expenditure of human effort buys a lottery of quality — and whether a child understands the theorem comes down to which of the hundred thousand they happened to draw.

We pay the highest imaginable price, and receive a below-best, wildly uneven result. Duplication manages to be, at once, the most expensive and the least consistent way to teach.

Zoom into one teacher's working day and the waste sharpens. She might actually teach three hours; the rest goes to preparing and marking. So the majority of a skilled human's paid time is spent on the two tasks a machine does best — delivering and grading — and only a sliver goes to the one thing that genuinely needs a live human: the child in front of her.

And this is just one lesson. One theorem, in one subject, at one grade. Repeat the arithmetic for every lesson, every subject, every year-group, across every school — and the number stops being an inefficiency and becomes one of the largest standing misallocations of skilled human time anywhere in society. We don't ask a hundred thousand filmmakers to re-shoot the same film every year so each town can watch a local copy. We record the best one once, and everyone watches it. In teaching, we do the opposite, on a scale almost too large to picture — and call it normal.

Record the single best delivery of Pythagoras — the clearest explanation there is — once, and give it, identically, to all hundred thousand classrooms. Hand the marking to a machine that does it instantly and evenly. And watch what happens to the number: the 800,000 annual hours don't shrink by ten percent. They collapse — toward the cost of a single recording, made once, kept forever.

Those reclaimed hours don't vanish. They redirect — entirely — to the one thing no recording and no machine can do: sitting with the child who still doesn't understand.

Here is the point that matters most, and it must be said plainly: this is not a plan to fire teachers. It is a plan to stop wasting them.

When delivery is recorded and marking is automated, a large share of teachers' time — perhaps half — is freed. Not cut. Freed. And freed human capacity is the most valuable thing an economy has. Some of those teachers stay in schools, doing the human work the next pieces describe. Others can be retrained — using this very system — into the fields every country is short of: healthcare, engineering, technology, the trades. The largest pool of educated, capable people in most economies is currently locked into re-delivering content a recording could hold.

Every country complains it cannot find enough skilled people. Sit with the paradox in that: a great share of its skilled people are tied up, right now, manually re-producing lessons that never needed re-producing. Free that capacity — retrain it into what's short — and you ease the very shortage the system keeps panicking about. Not by importing anyone. By stopping the waste of the people you already have.

The backlog has a face, too. While a hundred thousand teachers spend the year re-delivering a theorem that never changed, a newly arrived immigrant sits on a months-long waiting list for a language course that cannot open — because the one qualified adult who could teach it is busy being re-produced, by hand, somewhere else. The wasted hours and the unmet queue are the same problem, seen from its two ends.

And that points at something larger than any one shortage. A nation that can develop and re-skill its own workforce, fast and at scale, is a nation that doesn't depend on outside supply to staff its own hospitals, grids, and industries. It becomes self-reliant in the one resource that matters most — its own people's skills. That is strategic strength, produced as a by-product of no longer wasting a theorem's worth of human hours, eight hundred thousand at a time.

Teach it once, well, forever — and spend every reclaimed hour on the child, and every reclaimed career on the work a country actually needs.

Next: how fast a country can learn something genuinely new — and why, today, the answer is "far too slow."

Written from a working platform and a concept put up to be tested — not from a result already claimed. — Crosshire.