Coding classes in Herefordshire

The learner starts with a shelf of 229 books, the number of medieval manuscripts the cathedral says now occupy two bays of the Chained Library, kept strictly in order. Fifty new books arrive, each belonging at a random place in the order, and the learner writes three versions of the shelf. The first is a packed rod: books sit on consecutive hooks, so putting one in means sliding every later chain along by one. In programming, that is an array, and the slide is the cost of inserting into the middle of one.

The second version is the computer scientist's answer, a linked list, where each book simply holds a pointer to the next. Inserting is then cheap, just two links to change. But there is a catch that surprises most learners: to find where a new book goes, you have to walk along the chain from the start, one book at a time, and that walking costs almost as much as the sliding did. The work has not disappeared; it has moved from moving things to finding things.

The third version is what librarians do anyway: leave spaces. With a couple of spare hooks in every run of ten, most new books need only a few neighbours moved, and the whole shelf needs re-spacing only occasionally. Two spares per ten cut the work to 1,039 slides; five spares cut it to 264. Databases, file systems and text editors all use versions of this trick. The price is shelf space, and choosing how much to leave is a real design decision.

So a Herefordshire teenager should learn to code in 2026 to make the design choice, not just to type the structure. The longer argument is in why teenagers should still learn to code in 2026.