About
How we report
Parts of Franklin Frontline are gathered automatically from public calendars and records. This page says which parts, where they come from, and how what we publish is checked.
Gathered by script
The trackers read their sources every morning.
The listings and maps below are filled by code that reads each source on a schedule. A listing goes up as the source printed it, without a person reading each one first, so a venue's mistake can show up here until we catch it or a reader tells us. Each tracker names its sources on the page.
Live music
Read from each venue's own calendar every morning, with repeats and Nashville rooms taken out by rule. Our picks, marked with a star, are chosen by a person.
What's on
Read from organizers' own calendars every morning. An event an organizer sends us is read by an editor before it goes up.
The development map
Read from the city's meeting agendas and TDOT's project pages each morning. The project pages are written and updated by hand from those records.
Weather
The National Weather Service's forecast and alerts, live. The station record is computed from NOAA's daily data.
Openings and closings
Leads come from the city's Beer Board agendas and the businesses' own pages. A person checks each one and updates the ledger every month.
From the record
Every figure traces to a document.
Articles, explainers and guides are built from the public record: agendas, minutes, staff reports, budgets, rate sheets, census tables and map layers. Each piece lists its sources and links to them, so you can check the work.
The vote records are typed from the minutes, many of them scanned, and checked against the city's own roll-call record. Where the two disagree, the vote says so and the disagreement is listed on the corrections page.
The Thursday email is put together from the same calendars the trackers read, with a short note at the top.
Checked
Checked against the source, and fixed in the open.
Explainers carry a "last checked" date: the day every figure in them was last compared with its source. Many of those figures are also compared with the source data by automated tests, which fail when the two drift apart.
Every picture is credited. A stock photo says "Stock photo," a rendering says "Rendering," and a picture from another outlet or a public archive names its source and its licence.
When we get something wrong, we fix it and add a dated line saying what changed. Every one of those is on the corrections page, and the desk's address is on the About page.