Tribes2Wiki

The Community Tribes 2 Wiki — a rebuild of tribes2wiki.com, recovered from the Internet Archive and maintained in the open.

The original wiki ran on MediaWiki 1.16 from 2008 until it went offline. This project recovers what the Internet Archive preserved, converts it to Markdown, and publishes it as a static site so it can't quietly disappear again — the content lives in git, and anyone can fix an article with a pull request.

Status

	Original wiki	Recovered
Content pages	121	75
Uploaded files	87	2

Roughly half the wiki is gone, including Jericho mobile point base — its second most-read page. MISSING_PAGES.md lists every known gap, ordered by how many surviving articles link to it. Rewriting those is the most useful thing a contributor can do.

See NOTICE.md for provenance, CONTENT-LICENSING.md for how content is licensed (there is no single site-wide licence), and AUTHORS.md for credit to the original contributors.

Local development

npm install
npm run dev

Then open http://localhost:4321. npm run build produces the static site in dist/.

Repository layout

src/content/docs/     the wiki itself, as Markdown  <- edit this
src/components/       Starlight component overrides
src/styles/           theme
scripts/              the archive import pipeline
archive/              committed record of what was imported, and from where

The import pipeline

The import was a one-time seeding step. The Markdown in src/content/docs/ is now the source of truth — re-running npm run convert regenerates it from the archive and will overwrite any hand edits. The scripts are kept in the repository so the import is reproducible and auditable, not because they should be run routinely.

Command	What it does
npm run fetch-archive	Queries the Wayback CDX index, caches every surviving capture, writes archive/manifest.json
npm run fetch-history	Recovers per-revision authorship from archived history pages into archive/history.json
npm run convert	Converts cached captures to Markdown, writes archive/import-report.json
npm run report-missing	Regenerates MISSING_PAGES.md from the import report
npm run replay-history	Replays the recovered content as git commits attributed to their original authors

Raw captures are cached in archive-cache/, which is gitignored — archive/manifest.json records the URL, timestamp and SHA-256 of every capture, so the cache can be rebuilt exactly.

Notes for anyone re-running the import

• Wayback rate-limits aggressively; the fetcher is deliberately slow and resumable.
• tribes2wiki.com lapsed and became a domain-parking page, which the Wayback Machine kept crawling under the old /wiki/ URLs. Captures without MediaWiki markers are rejected.
• The wiki answered on both tribes2wiki.com and www.tribes2wiki.com, so a single crawl appears as two CDX rows; captures are deduplicated by timestamp.
• Distinct titles can slug to the same path (Targeting laser vs Targeting Laser). Both are kept and reported under "Possible duplicates" in MISSING_PAGES.md rather than one silently overwriting the other.

Contributing

See CONTRIBUTING.md. The short version: edit the Markdown, open a pull request.