Our own numbers

This site's whole claim is that a directory should publish what it measured and when. That standard applies to its own traffic too.

All-time, since 96 days ago. 361247 crawler hit(s) recorded and excluded from every count above. 2002 request(s) from our own deploy checks and tests, also excluded.

Last 14 days

Where visitors came from

What was asked for

How this is counted

• A visitor is one salted hash of address and user agent, on one UTC day. The hash input includes the date, so it cannot be followed from one day to the next, and the address itself is never written down.
• Requests to this page and to /api/stats are never counted. Reading the scoreboard must not move it.
• Crawlers are recorded and excluded. Deleting them would hide a traffic source; counting them would turn a search engine into an audience.
• Our own requests are recorded and excluded too — deploy checks, browser tests, the machine this is built on. These numbers decide whether the project continues, and counting our own work as demand is the most comfortable way to get that decision wrong.
• Counts are written from a bounded in-memory buffer, so a restart can lose up to two seconds of events.

Same numbers as JSON: curl https://stillworks.supercapybara.com/api/stats