W27 - Infrastructure Standards Working Group

Last week we launched a new working mechanism — the Infrastructure Standards Working Group. This is not a spur-of-the-moment idea or a whimsical initiative; it’s the result of a period of observation, reflection, and shared consensus building.

In fact, during the engineering governance initiative in the second half of last year, I already sensed that the task-based approach wouldn't suit the team’s next-stage pace and challenges. Task forces are a good tool for solving problems, especially when addressing acute bottlenecks or technical debt, because they allow rapid focus and concentrated effort. The problem is that task forces are essentially short-term operations: they depend heavily on individuals and intense rhythms, and without institutional safeguards their results often regress over time or quietly vanish as personnel change.

Over the past period I exchanged ideas with many people; there is consensus on the importance of infrastructure, but also awareness of practical issues. For example, shared understanding fades as new people join; progress depends on a few drivers; there is no mechanism for sustained investment; motivated colleagues can’t find ways to contribute; and active contributions can even trigger negative feedback loops.

I roughly divide the team’s infrastructure maturity into three levels:

Level One — Chaos and Emergency: infrastructure is fragmented, problems are patched reactively, and the team lacks a sense of overall responsibility.

Level Two — Orderly but Fragile: there is some accumulation, with relatively clear standards and processes, but they are prone to decline and lack protective mechanisms.

Level Three — Systematic and Evolving: stable mechanisms provide a safety net; engineering practices accumulate and evolve over time, steadily approaching the level of outstanding teams.

We are currently at the second level: orderly but facing hidden decline, slipping backward “without noticing.” If we keep doing things the old way, we will likely end up cycling through another round of people tearing things down and rebuilding, perpetually stuck in the same band.

We need to find a new organizational approach to shift infrastructure work from episodic efforts to sustained campaigns. The working-group mechanism is the solution we’re proposing at this stage.

What would I consider success? Being able to track the infrastructure problem map at any time and to form a clear evolution path — a mechanism the team deems worth preserving and passing on.