Where readers came from
| Site | Count |
|---|
| google.com | 67 |
| reddit.com | 55 |
| com.reddit.frontpage | 34 |
| com.linkedin.android | 18 |
| linkedin.com | 16 |
| duckduckgo.com | 15 |
| out.reddit.com | 11 |
| bing.com | 9 |
| gemini.google.com | 5 |
| perplexity.ai | 4 |
| kagi.com | 3 |
What people searched for
| Search | Count |
|---|
| idaaaatarbell | 8 |
Links people followed
| Link | Count |
|---|
| feeds.aaddrick.com | 11 |
| github.com/aaddrick/claude-desktop-debian | 9 |
| github.com/aaddrick | 5 |
| aaddrick.com | 4 |
| aaddrick.github.io/le_prix_de_respirer | 3 |
| github.com/aaddrick/scripting-blender-scenes | 3 |
How this works
Each page sends one small request when it loads, plus one for a search, a click
on an outside link, or a scroll past a quarter of the page. There are no cookies.
An hourly job counts those requests from the server logs and keeps only the
totals: no addresses, no browser details, and only the name of a referring
site, never the page. Searches, referrers and links show up here once they've
been seen 3 times in the range.