Old bug in Microsoft Word

Open Microsoft Word, type =rand(200,99), and press Enter. The application fills the document with the sentence “The quick brown fox jumps over the lazy dog” repeated enough times to produce roughly 235 pages.

The sentence is a pangram—it contains every letter of the English alphabet—which makes it genuinely useful for testing fonts and print output. The =rand() function was a developer tool, not a product feature, and it shipped in Word for years before most users knew it existed.

Calling it a bug is technically arguable. It is an undocumented function that behaves deterministically and produces a predictable result. What makes it feel like a bug is that nothing in the Word interface suggests it exists, and nothing explains what it is doing when it fires. Users who encounter it by accident (or via a Chetan Bhagat novel, which is how I found it) experience it as surprising behavior rather than a feature they invoked.

That distinction—between behavior that is technically intentional and behavior that is experienced as broken—is one that product engineers spend a lot of time navigating. A feature no one knows about is not meaningfully different from a bug from the user’s perspective. The code path works; the mental model does not.

The more interesting observation is that this stayed in Word for years without being fixed, patched out, or documented. Which suggests that the teams maintaining the product either did not consider it worth touching, or actively wanted it to persist as a low-risk escape hatch for testers and developers who knew about it.

Every large, long-lived software product contains things like this. They accumulate in the dark corners of codebases that have been through multiple ownership changes and rewrites. Most of them are harmless. Some are not. The difference matters a great deal when you are doing security auditing rather than casual testing.