Pagination and the Archive as Navigation

Pagination is a solved problem that publishing systems continue to solve badly. The common failure mode is pagination without context — previous and next buttons that provide no information about what lies in either direction, page numbers with no indication of total depth, or infinite scroll that removes the concept of position entirely.

A publishing archive has two navigation needs. First, readers arriving at the site for the first time need a way to understand the scope and character of the publication. Second, regular readers need a way to find specific posts they remember imprecisely — they know the approximate date, or a word from the title, but not the exact URL.

Pagination addresses the first need partially. A chronological list of posts with titles and dates, broken into pages, gives a reader a navigable history of the publication. The page count conveys scope. The titles convey character. The dates convey publication frequency.

The archive page addresses both needs more directly. A single chronological list of every post, without pagination, allows a reader to scan the entire publication history in one view. For a site with hundreds of posts this is a long page, but it is also a complete index. The browser's find function becomes a search interface. The visual scan reveals publication patterns that paginated views obscure.

Flatpack generates both. The paginated index at / and /page/N/ provides the standard entry point. The archive at /archive/ provides the complete list. The sitemap at /sitemap.xml provides a machine-readable index of every URL on the site.

Navigation is not a feature. It is the interface through which a reader's relationship with a publication develops over time. The archive is not a technical requirement. It is an editorial commitment to the reader who wants to understand what the publication is.