Coding classes in Fife

The frame is invented and far simpler than the real bridge. It sticks straight out from a wall in three square bays, each 4 metres wide and 4 metres tall, with a top rail, a bottom rail, an upright at the end of each bay and one diagonal across each. It is pinned to the wall at the top and bottom, and a load of 10 kilonewtons, roughly the weight of a tonne, hangs from the tip. At every joint the pushes and pulls must balance, left against right and up against down. That gives two equations per joint: sixteen equations for sixteen unknowns, the forces in twelve bars plus four forces from the wall.

The learner builds the equations in Python and solves them with Gaussian elimination, the same method taught for simultaneous equations, only bigger. The pattern makes sense: the top rail is stretched and the bottom rail squashed, both most strongly near the wall where the bending is greatest. A hand check confirms it: taking moments about the top wall pin, 10 kilonewtons at 12 metres out must be resisted by the bottom rail 4 metres below, so 10 times 12 divided by 4 gives 30. Two bars at the tip carry nothing under this load, yet they matter as soon as the load moves.

Now the real lesson. Delete one diagonal and try again. There are still sixteen equations but only fifteen unknowns, and elimination finds a contradiction: no set of forces can balance the load. The frame has become a mechanism and would fold. Add an extra diagonal instead and the opposite happens: seventeen unknowns, only sixteen independent equations, and one force is left undetermined, because the answer now depends on how stretchy each bar is. The rank of the system tells the computer which of the three cases it is in. Halve the height of the frame to 2 metres and the rail forces double to 40 and 60, which is why deep frames are efficient.

So for a teenager in Fife, learning to code in 2026 means being able to tell an answer from a number. The longer argument is in why teenagers should still learn to code in 2026.