You On AI Field Guide · Maintenance Knowledge The You On AI Field Guide Home
TxtLowMedHigh
CONCEPT

Maintenance Knowledge

The accumulated, contextual, often undocumented understanding of how a system actually works—why this server crashes on the third Tuesday, which function was written non-obviously to avoid a race condition, where the seams are and why they leak—that lives in the maintainer's body and departs when they do.
Seventy percent of all software engineering labor is devoted not to building new things but to keeping existing ones running. David Edgerton's use-centered history of technology reveals that this maintenance majority is not an anomaly but the structural norm across every technological domain: ten maintenance workers for every combat soldier in the Second World War, the bicycle mechanic as the unsung engine of healthcare delivery in the developing world, the railroad worker who kept the infrastructure of industrialization running long after the era of locomotive invention had ended. Maintenance knowledge is the cognitive resource that makes the maintenance majority possible—and it is the resource that large language models handle least well. It is not the knowledge that lives in documentation, which describes the system as it is supposed to work. It is the knowledge of the gap between specification and reality: the understanding of why a particular implementation was
← Home0%
CONCEPTBook →

Keep reading with YOU ON AI

Unlock the full book, 10,000+ field-guide entries, and a 1000+ thinker library. If you have a book code, register now — it takes a minute.

Register with book code Sign in