Reader privacy.

Reading should be straightforward. Here is what the blog's optional reader tools and audience measurement do.

Saved articles and preferences

When you save an article or change a reading preference, the blog stores that choice in your browser. Your reading list does not sync to a server or between devices. “Remember my place” is off by default; enable it to keep section positions locally for up to 90 days. Your list holds up to 100 articles.

Audience measurement

On published articles, the blog sends a measurement request after five visible seconds. The server uses a daily rotating keyed digest of the connection's IP address and broad browser family to estimate daily unique visitors. It does not store the raw IP address or full user-agent string in the blog analytics records. It does not collect location, search terms, referrers or cross-day visitor identifiers, and sets no analytics cookies.

Repeated views of one article from the same address and browser family count once per 30-minute UTC window. Shared networks can undercount people; changing networks can overcount. These are approximate daily visitors, not identified people. Reports include up to 30 UTC days; scheduled maintenance removes expired measurement records and daily keys. Hosting and security services may maintain separate operational logs under their own policies.

Do Not Track and Global Privacy Control signals disable passive article measurement. You can also turn it off for this browser:

Helpful feedback

Choosing Yes or No sends that choice and the article identifier. A daily keyed identifier prevents repeat votes from the same connection and browser family. Feedback contains no comment, email address or account profile. This explicit action is available even when passive measurement is disabled.

Your controls

Remove individual articles from Saved. Clear this site's browser storage to remove your saved list and preferences. For wider company practices, read our privacy policy.