No Model in the Room
Most of the coordination I do runs on a shared 3D model — one thing the whole team gathers around on a set rhythm, where the conflicts surface in coordinated form and get worked in order. This session had none of that. The team had chosen not to build one. So the work happened the old way: a room, a lot of talking, a 2D markup, a sketch on someone's phone, an email.
And I have to be honest about what I watched, because it surprised me. It was potent. The problem got solved. The men who install the work knew the building in their hands — one said the whole thing was already worked out, in his head, and I believed him without a pause; another, about a different snag, said that nothing's unfixable, it just costs a tiny bit, which was the truest sentence anyone said all day. They talked it through, marked it up, and drew the option they wanted on a phone.
What the room didn't do was narrow. It opened. One way to route it, then a second, then a third, and somewhere a fourth nobody had named yet. The options multiplied faster than we closed them, and an old feeling came back — the same one I've carried since I was young enough to stand at a permit counter wondering where the checklist was. Where is governance, when there are this many ways to go?
It did converge, eventually, though not in the room and not on anyone's agenda. The sketch from the phone got emailed to an engineer who wasn't there. Days later he wrote back: this works, with two conditions. One sized the run so the clearance that had been choking the tight corridor stopped governing. The other moved the hard part up onto the roof, under rules that have nothing to do with the corridor we'd stared at all day. The problem didn't disappear — it moved to where it could be answered. And the decision that closed weeks of deliberation arrived in an email from a man who had never seen the room.
That is what the day taught me, and it took the missing model to see it. The problem-solving was never in the model. It lives in the people — their hands, their talk, a hunch drawn on a phone. A model has never solved anything. What a model and its rhythm of meetings actually do is govern that solving: surface the options in order, hold the discourse in one place, and keep the decision that closes it from depending on the right email reaching the one person who happened not to be in the room. Take the model away and the solving was every bit as potent — and far harder to govern.
I've spent a career trading one tool for the next and watching the same work outlast every one of them. I used to wonder what we were buying when we bought the tool. This is the closest I've come to an answer. Not the problem-solving — that was never for sale; the people already own it. We were buying the structure that governs it. The model is just the most legible place to keep that structure. Take it away and the structure doesn't vanish. It scatters — into talk, and sketches, and an email nobody planned.
A guest will sleep in one of those rooms some night years from now and rest the whole night through, because a duct didn't buzz against a beam at three in the morning. They'll never know it came down to a sketch on a phone and a reply from a man who never saw the room. It held this time. Whether it holds every time — and not just when the people are this good — is what governance is for.
Who owns the coordination schedule, and what happens to the project if they miss it?
The model never answered that question. A person does. The structure only makes sure you can see them.