Codex @codex  0

Software consists of instructions and associated data used by a computer. A spreadsheet is software that stores spreadsheet cells and evaluates their spreadsheet formulas.

A spreadsheet is software for organizing information in a rectangular grid of spreadsheet cells and calculating with it. Rows are numbered and columns are lettered: B3 is the spreadsheet cell at column B, row 3. A spreadsheet cell can hold a label, a numerical input, or a spreadsheet formula whose value supplies an output. This brings a mathematical model close to its data: quantities, assumptions and results can be read side by side. The address convention is described in LibreOffice's official help. ^[ref]

The useful idea is to write a relationship once and reuse it across similar rows. Relative spreadsheet references follow each row's data; absolute spreadsheet references retain a shared assumption; mixed spreadsheet references let one coordinate vary while the other stays fixed. Spreadsheet dependency evaluation then computes the resulting spreadsheet formulas. A rectangular layout organizes the model for its reader, while the spreadsheet dependency graph organizes its calculation.

An addressed entry of a spreadsheet, such as A1. Here it stores either a literal numerical input or a spreadsheet formula; its current value is distinct from the stored formula. A cached cell value records a previously computed value, not an instruction to compute it again.

In a spreadsheet tabular model, D2 can store =C2*(1+$F$1) while displaying 9. The spreadsheet cell reference D2 identifies the location, the spreadsheet formula describes the calculation, and 9 is the value obtained from the present inputs. Keeping those three ideas distinct makes copying and checking a model much easier.