Chapter 1 promised that the Marquee's spreadsheet chaos has a fix; this chapter builds it. Five topics turn the cinema's four kinds of facts into four real tables — with declared columns, types that refuse nonsense, keys that name every row, and pointers that connect the tables to each other. By the last page you can read the complete Marquee schema the way you read a map, and "relational" finally means something exact.
5 topics
This is the chapter where the database stops being an idea and becomes a structure you can draw. Everything a relational database holds lives in tables, and a table is a stricter, more honest thing than the spreadsheet it resembles: its columns are declared up front, each column has a type that rejects values that don't fit, and each row is one complete fact. The strictness is not bureaucracy — it is exactly the protection Chapter 1 showed the spreadsheet lacking.
The five topics build in a straight line. First the table itself: rows, columns, and the declared shape. Then types — what each column promises — and NULL, the strange marker for "no value here" that trips up professionals. Then primary keys, which give every row a name nothing can confuse; foreign keys, which let one table point at another and put the "relational" in relational databases; and finally the junction table, the pattern that handles "many customers attend many screenings" — and turns out to be where the Marquee's most important data lives.
The Marquee schema — four tables, three pointers, the whole book's home