Every great system begins with one humble issue. Mine was ORCH-1, and its title was simply: “Apricots exist and I am not ready.”

I want to be clear that this was not a joke ticket. It had a priority (Critical), an assignee (me), and a due date (the entire summer). It had a description three paragraphs long, written at 5:40 a.m. under a tree, on a phone with a cracked screen and dewy thumbs. It is, to this day, the most honest piece of project documentation I have ever produced.

The problem with fruit

Software waits for you. You can let a pull request sit. You can snooze a Slack message. Fruit does none of this. An apricot ripens on its own schedule and then, with no changelog and no deprecation notice, it rots.

The first summer I treated the orchard like a side project. “I’ll get to it.” You do not “get to” apricots. The apricots get to you. By mid-July I had a tree that went from green to glorious to ground-mush in roughly four days, and I had nothing written down except regret.

Enter the board

So I built a board. Three columns, expanded over time:

• Backlog — fruit that exists but isn’t my problem yet
• Ripening — fruit on the clock
• In Review — fruit I have squeezed, gently, with intent
• Done — fruit that has been eaten, gifted, or jammed

The moment I could see the orchard as a queue, the panic left. There is nothing scary about forty apricots. There is something terrifying about an unknown number of apricots ripening at an unknown rate. The board converted dread into throughput.

What ORCH-1 taught me

I closed ORCH-1 in late September. Resolution: Fixed. In the comments I wrote one line for future-Ziv:

You were never overwhelmed by apricots. You were overwhelmed by not knowing where the apricots were.

That sentence is now printed and taped to the inside of the tool shed. The orchard has run on tickets ever since, and we have not lost a tree to chaos since ORCH-1 was resolved.

Next time: how I actually triage a single apricot, field by field.