Sitemaps and orphan pages

Link-following finds the connected web. A sitemap finds the rest.

A sitemap is an XML list of URLs, optionally with a last-modified date, that a site hands to crawlers directly. Its quiet superpower is the orphan: a page that exists and is worth reading, but which nothing on the site links to. Through links alone, an orphan is unreachable; listed in the sitemap, it is one fetch away.

This site has one original orphan plus a six-page atlas control cohort. They are not linked from ordinary HTML — on purpose. The only advertised discovery route is the sitemap. Whether a crawler fetches none, some, or all of them is a measurable choice.

The lesson for publishers is unglamorous: important pages should be both linked and listed. Redundancy is how you survive a reader that only does one of the two.