Someone forked it

I built it.

Two weeks, mostly with an agent — not a detail I could have written eighteen months ago. A running application made from our real components and tokens. Mock data. Runs on anyone's laptop. No IIS.

Presented it on the twenty-fourth as a clean slate: something designers and product managers build in, rather than describing screens for someone else to make.

Today, two days later, someone on the team forked it into their own repository. Without being asked. No kickoff, no project, nobody chasing them.

That is the entry. That is the whole thing I wanted to record.

I have presented a lot of proposals over twenty years and I have learned to distrust a room. People nod at a good argument because nodding is polite and costs nothing. I have watched initiatives collect unanimous approval and then quietly not happen, and I no longer count approval as a signal at all.

Nobody forks a repository to be polite. They do it because they want to use the thing this afternoon.

Two days is also the right interval. Long enough that it was not a favour to me, short enough that nothing had gone stale. If it had taken six weeks I would have learned only that people are busy.

So the measure I am going to run this programme by, and the one I will teach the rest of my team to use, is not whether a plan gets approved. It is how quickly somebody takes a copy and changes something inside it. Everything else is a conversation about a document.

Where this goes: I want every designer and product manager in this company running their own copy by the end of the year, and I want the design system to spread the way this fork did — because it is the fastest way to get your work done, not because anyone mandated it. Systems that require enforcement are already failing. Systems people take without asking are the ones that last.