This observation comes from Tony Hoare’s inaugural lecture at Oxford, “The Mathematics of Programming”. It notes the sheer size of modern computer programs, which can run to many thousands of pages of dense text, far larger than any expression a mathematician would normally handle.

The point identifies a genuine difficulty for Hoare’s own argument. He believed programming to be a mathematical activity, and that programs could be proved correct by mathematical means. Yet mathematics has traditionally worked with compact formulae and short proofs, not objects on the scale of large software systems.

The remark therefore frames a challenge as much as a claim. If mathematical methods are to guarantee the correctness of real programs, they must be developed to cope with a scale unlike anything in mathematical tradition. It reflects Hoare’s awareness of the gap between the promise of formal methods and the practical size of the systems they must address.