What is missing

This archive is a record of what survived, not of everything the Beacon published.

406 captures with no article in them

Of 3,630 captures read, 406 hold no article content at all: they are template pages whose only text is the navigation and the footer, typically eight kilobytes of markup around nothing. They are excluded from the collection and counted here rather than dropped silently. An earlier extraction pass admitted them by matching a container that held only chrome, which is why they are named here.

The paper was SEO-injected, and 204 posts still carry it

204 of the 3,224 rendered posts carry an identical keyword-stuffing block — about sixty-four repetitions each, thirteen thousand occurrences in total. It is not removed. A reader in 2009 loading that page saw it, so removing it would edit the record; it is marked as not the publication’s speech and excluded from search.

The archive cannot say when the paper was compromised. Injected posts cluster by POST date in late 2009 and early 2010, but post date is a proxy: injection happens at CAPTURE time, so a 2005 post carrying the block means that post was re-crawled during the window, not that the compromise began in 2005. The capture timestamps that would bound it are not held.

657 photographs and graphics the archive cannot show

The recovered posts reference 1,325 distinct images across 2,285 posts. Of those, 668 have a capture the archive has located and are queued for recovery; 657 have no capture located.

What that second figure does and does not mean. It means no capture was found as of this build, in the sources checked — the phase-one Wayback index held for this collection and the Common Crawl rows on disk. It is not a finding that the image does not exist anywhere. A later sweep of a source not yet checked may recover some of them, and this number will move.

Until they are recovered, the posts that referenced them render without the image. The reference is left in the page rather than stripped, because what the post pointed at is part of the record.

How it ended, and why this page cannot say

There is no farewell in anything the archive holds. The paper stops mid-stream. Its last held post, dated 4 August 2018, is aggregated wire content rather than a sign-off, and the last capture of its About page — 21 December 2018 — still describes the publication in the present tense, as a going concern.

Every captured document page was checked: one hundred captures of the About page and eight of the copyright notice. None is a closing statement. All three of the publication’s domains are now dead — one parked, two no longer resolving.

So if the editors said goodbye, they said it somewhere this archive does not hold, or after the last capture was taken. A sweep for captures from 2019 onward is queued and has not yet been possible to run. What the record shows is a publication that stopped rather than one that signed off, and the difference between those two is not something the archive can currently settle.