A deliberately short list. The bias throughout is toward primary sources — organisations that gather and publish their own data or hold their own records — rather than commentary about them. Each entry says what it is actually useful for, because a list of links without that is just decoration.
On enterprise and the ordinary firm
- SBA Office of Advocacy — the standing statistical profile of small firms in the United States: how many there are, how much of employment they account for, and how that has moved. The state-level profiles are the fastest way to get an accurate picture of a particular place.
- BLS Business Employment Dynamics — establishment births and deaths, and the survival curves for new businesses. Sobering, and the correct antidote to survivorship bias in business writing.
- Federal Reserve Small Business Credit Survey — an annual survey of what small firms actually experience when they seek credit. Notable for capturing the firms that were turned down, which most financing coverage never sees.
- SCORE — decades of volunteer business mentoring through an SBA resource partner. Useful here mainly as evidence that the interpersonal transfer of practical knowledge can be organised at scale without becoming a programme.
On the Southwest
- Arizona-Sonora Desert Museum — the best general introduction to Sonoran Desert ecology, including the nurse-plant relationships that recur throughout these essays.
- Saguaro National Park — the National Park Service's material on saguaro biology, growth rates and longevity, drawn from long-running monitoring of the same stands.
- USGS Water Resources — instrument records for surface and groundwater. The long series are the point: they show variation that no living person has experienced directly.
- National Weather Service — Southwest Monsoon — seasonal records and explanations of a rainfall pattern that arrives almost entirely as bursts.
On long records and long horizons
- Laboratory of Tree-Ring Research — the University of Arizona laboratory where dendrochronology was developed, and the source of climate records reaching back millennia. The single most convincing demonstration available that a century is a short sample.
- Long Now Foundation — a standing argument for extending the horizon of ordinary planning, and a catalogue of attempts to build institutions that operate on that timescale.
- "The Gospel of Wealth" — Carnegie's 1889 essay, worth reading in the original rather than in summary. Its argument for building capability instead of distributing relief has aged better than its tone.
On places and the people in them
- American Community Survey — the workhorse dataset for population, housing, employment and commuting at small geographies. It will reliably tell you what is happening to a town, and reliably fail to tell you why.
- NPS Historic Preservation Tax Incentives — the mechanics of rehabilitating older commercial buildings, and an implicit argument about why adaptable building stock is worth keeping.
- Library of Congress Digital Collections — photographic and documentary records of American towns across a century. The most efficient cure known for assuming that the present arrangement of any place is the natural one.
A note on how to use a list like this
Read the data sources directly rather than the articles about them. Almost every organisation above publishes its methodology alongside its findings, and reading the methodology first is usually more instructive than the headline number — it tells you what the measurement can and cannot see, which is the part that gets lost in every retelling.
It is also worth noticing what is missing from this list: nothing here is a source about giving, philanthropy or the sector that surrounds it. That absence is deliberate, and the reasoning behind it is in Quiet Generosity.