The reason internal codenames leak into the market

Products ship with a name chosen by committee and are discussed for years afterwards under the name engineering used during development. The official name appears on the box. The codename appears in conversation, in bug reports, in job adverts and eventually in the press, having accumulated more usage than the name that replaced it.

What everyone sees

Leaked codenames are treated as security failures — someone talked, a document escaped. Confidentiality does fail. But that explains how the word became public, not why it persists after an official name is available and actively promoted with a budget behind it.

What is actually happening

Clark describes language use as coordination on a shared lexicon within a community: expressions become entrenched because a group has used them successfully, and the entrenched term is retrieved faster than a newer synonym. Garrod and Anderson documented lexical entrainment, in which partners converge on a description and then reuse it in preference to alternatives. Both mechanisms favour the codename. It was in circulation earlier, was used more often, and is usually shorter, so it is more accessible at the moment of speaking. Official names win on documents, where retrieval is not under time pressure, and lose in speech, where it is.

Why it stays hidden

The persistence hides because organisations measure names in published material, where the official name is enforced. Speech is unlogged. By the time the codename surfaces in a public article it has been the working term for years, and it appears to have leaked recently when it had in fact never stopped being the primary word.