\"Which parts of this are real?\" should be a question with an answer
TL;DR The most expensive bug I hit yesterday wasn't a crash. It was a stub that answered politely. Four shapes of the same problem: a column nobody writes to, a model nobody calls, an engine backed by a fake, and a resolver that falls back to a null implementation silently . The fix isn't "delete the fakes". Fakes are how you ship in slices. The fix is making which parts are real an enumerable…
The most expensive bug encountered yesterday was not a crash, but a stub that responded politely. This stub manifested in four distinct ways: a column without a producer, a model not called, an engine backed by a fake, and a resolver that silently defaulted to a null implementation. The solution to the issue lay not in eliminating the fakes, but in establishing a clear distinction between what was real and what wasn't.
A critical insight was that null signified uncertainty, not zero. Furthermore, an unreachable provider should not be misconstrued as a cluster that had collapsed to nothing. The day was filled with similar commits, each assumed to be unique but ultimately identical. Throughout the day, no exceptions were thrown and no tests failed, yet the system persistently provided answers it should not have.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written; read the original for the full account.
