A standing team with named desks

A desk on an org chart is a rectangle with a label in it until three more things are true. It fires on something other than a person remembering to click go. It is bounded — there's a specific list of what it's allowed to touch and what it isn't. And it leaves a timestamp: a real answer to "what did this desk actually do last," checkable without finding the person who set it up and asking them.

Here are four, kept generic on purpose — the pattern is what matters, not whose project they run against. Click through them. Each one has a different trigger and a different boundary, and each one has something real it shipped, not a placeholder.

The desk I trust least on any chart is the one whose status a person types by hand. It's accurate the day it's written and a guess every day after. A status board that rewrites its own table on a timer is a different object entirely — not because the underlying information is more sophisticated, but because the timestamp on it means something. Stale is a property of who's holding the pen, not of the information itself.

Not every desk creates. The watcher desk above does the opposite: its entire job is noticing that another desk's output went missing or wrong, fast enough that it doesn't matter. That's why its cadence is the shortest of the four — thirty seconds, not fifteen minutes, not a push hook. A desk that only corrects has to check more often than a desk that produces, because the cost of noticing late is exactly the cost of not noticing at all.

What none of the four desks can tell you is whether they're still running. A cron job that silently stops doesn't announce it — the timestamp just stops updating, and nothing reads a timestamp that never changes as an event. The failure mode of a standing team isn't a desk doing the wrong thing. It's a desk that quietly stopped doing anything, weeks before someone notices the gap it left.