Eighteen Years of Agile: Strange Phenomena and the Quest for Software Engineering Fundamentals in China

The author reflects on a discussion in a fan group that reminded him of an old 2018 article titled "Strange Phenomena of Eighteen Years of Agile" by Xiong Jie, noting that the piece is long and unlikely to be read fully by most readers.

He recounts how China's IT industry was spurred by the 2000 State Council "Document 18", which offered generous subsidies for software and IC development, leading to rapid growth from tens of thousands to millions of practitioners within a decade.

This rapid expansion brought classic software crisis issues—poor quality and delivery delays—prompting the adoption of CMM (Capability Maturity Model) as a remedy, with substantial government subsidies for certification levels.

Despite the influx of CMM, the industry still suffers from superficial practices: requirement gathering reduced to perfunctory interviews, project management reduced to arbitrary deadline negotiations, and configuration management hampered by massive, poorly managed codebases.

The author shares vivid anecdotes illustrating these problems, such as a client demanding a three‑hour requirement session only to receive a two‑minute summary, and developers struggling with a 5 GB repository containing thousands of branches.

He argues that the root cause is a lack of solid software engineering fundamentals. Extreme Programming (XP) and Test‑Driven Development (TDD) are presented as the only approaches that truly address the core skills gap, providing reliable unit tests, continuous integration, and disciplined refactoring.

The piece critiques the degeneration of Scrum into "zombie Scrum" where teams work in isolation without shared code ownership or automated testing, leading to stagnant productivity.

Later sections discuss newer trends—micro‑services, DevOps, cloud‑native development—showing that each wave merely masks the underlying deficiency: insufficient basic skills and the absence of effective TDD.

Concluding, the author emphasizes that improving software quality and speed requires deliberate practice of fundamentals, not just tools or processes, and describes his own training sessions where participants repeatedly practice TDD and refactoring to build lasting competence.