My own machine, counted on 6 September 2026. Every square below is a folder. The white ones I have opened this year. The grey ones I have not, and most of them I will not.
In January 2026 I built a directory of 131 open source tools for ministries and churches, on the premise that subscription software quietly eats budgets that should go elsewhere and that a ministry ought to own its own data. It is the argument of this whole site in miniature.
It is easier to write that ministries should own their infrastructure than to keep maintaining the thing that would let them. I am not sure whether abandoning it was a failure of faithfulness or an accurate read that it was not mine to carry. I have not resolved that, and I distrust anyone in this field who says the question is easy.
My archive is not unusual. It is the norm rendered small.
What changed in the last twelve months is the cost of the first working version. Agentic coding tools have collapsed it, and ministry conferences are full of people discovering this at the same moment. Building was never the expensive part. Distribution, trust and maintenance were, and none of them got cheaper. Cheap creation against unchanged maintenance will fill this sector with more tools that work on the day they ship and nowhere after.
Three people, three directions, one claim.
A farmer, 1987
Wendell Berry, explaining why he would not buy a computer. His sixth standard for a new tool: it should be repairable by a person of ordinary intelligence, provided that he or she has the necessary tools.
He was asking whether a thing can be kept alive by the people who depend on it.
A maintainer, 2025
Henry Zhu maintains Babel, a piece of infrastructure a large part of the JavaScript world depends on without knowing it. He titled a talk about that work Open Source of Anxiety
. The words he uses are burden, overwork, and continuing to say yes. He calls open source roads and bridges. His text is Matthew 10:8.
An archive, 2026
Mine. Fifty-seven folders. One of them is the case for owning your tools, unowned.
Whoever can repair it owns it, and the people holding up shared infrastructure are mostly doing it unpaid, in the margins of other work, invisible precisely because nothing breaks while they keep going.
What the field needs is not slower innovation but more owners. The artefact does not compound. The capability does. A team that builds and maintains a mediocre thing can build a better second one; a team handed an excellent thing learns nothing and is stranded the day the builder leaves.
Counts from ls -d */ | wc -l on 6 September 2026. Berry, Wendell, "Why I Am Not Going to Buy a Computer", 1987, reprinted in What Are People For?, North Point Press, 1990. Zhu, Henry, "Open Source of Anxiety", talk, 28 October 2025. This page is itself a folder, and will be counted on one side or the other.