Visitors' browsers keep files from operator nodes and fetch the next pages' files early

When a visitor’s browser fetches your site’s files from operator nodes, as described in the operator nodes entry, it builds each file from pieces it checks one by one. Until now, a file built that way was thrown away after one use, so the next page asked for it again. Now the browser keeps it, and while the visitor reads, it also fetches the files the pages you link to will need.

What changes for your visitors

• Files are kept for later pages. A file the browser built from verified operator-node pieces is kept and loaded from the browser, even before it has reconnected to any node.
• Kept files are never stale. They belong to one deploy. When you deploy again, the old copies are never served and are deleted.
• Files for the next pages are fetched in advance. After the page has loaded and the browser is idle, it reads up to four pages your page links to, making the same requests following those links would. It then fetches the images, stylesheets, scripts, fonts and icons those pages use from operator nodes, and checks them like any other piece.

Limits on fetching in advance

• At most 2 MB per visitor every 30 minutes.
• Never pages, video, audio, or files over 1 MiB.
• Nothing when the visitor has Data Saver on or is on a 2G-class connection.
• It stops when the visitor leaves the page.

Trade-offs

• Each visit can use up to 2 MB of the visitor’s data for files they may never open.
• Your site gets up to four extra page requests per visit, for the linked pages that are read.
• It only happens on sites placed on operator nodes, and only once the browser has connected to a node. Pages themselves still come from the Regional Gateway.

What you need to do

Nothing.