Expansion Library — All of Human Knowledge, Offline
This guide is the portable core — under 600 MB, fits any pen drive. When you have internet, you can pull down the rest of recorded knowledge to read forever offline: the whole of Wikipedia, all of Project Gutenberg, medical encyclopedias, repair manuals, dictionaries and more — using free, open ZIM files and the Kiwix reader.
How it works — 3 steps
Build a Survival Drive — what fits
| Drive | What you can carry |
|---|---|
| This guide only | ~0.6 GB — the curated survival core (any USB stick ever made) |
| 32 GB stick | Guide + Medical Wikipedia + Simple Wikipedia + Wikivoyage + iFixit + a Gutenberg subset |
| 128 GB stick | Add full Project Gutenberg (every public-domain book) + Wikipedia "nopic" |
| 256 GB+ / SSD | Add full Wikipedia with images — essentially the whole library of humanity in your pocket |
Pen-Drive Essentials small & high-value
The highest survival-value archives that fit on a modest drive.
The Big Archives needs a large drive / SSD
The complete sets. Sizes are approximate and grow over time — pick the flavour that fits your drive.
Offline Maps & Navigation
🗺️ Open the built-in offline map — renders right in this browser, no app needed. Prototype region: Northern England & Yorkshire. Add any region with extract-map.sh.
| UK map, by detail | Approx size |
|---|---|
| One region / county (e.g. your local area) | ~100–400 MB |
| Whole UK — navigation (Organic Maps) | under 1 GB |
| Whole UK + topographic contour lines (OsmAnd) | ~2–2.5 GB |
Raw OSM data — Great Britain .osm.pbf | ~2.0 GB |
Scripted download (optional, recommended for the big files)
Prefer the command line? get-knowledge.sh in the repo is a one-button downloader for these archives — no need to hunt for the exact dated filename yourself. Built for connections that will drop mid-transfer over a multi-hour or multi-day download: it checkpoints progress, resumes exactly where it stopped, retries with backoff, and verifies the finished file against the SHA-256 Kiwix publishes for it (not just its size).
# the one-button case: full Wikipedia, with images (~115 GB) -> ./zim/
sh get-knowledge.sh wikipedia-maxi
# any other archive on this page — see the exact names:
sh get-knowledge.sh list
# check a download already on disk is intact and readable, no re-download:
sh get-knowledge.sh --recheck
# grab a whole curated bundle in one run (skips what's already verified if
# it's interrupted and re-run): medical + travel + textbooks + repair +
# dictionary + primary texts + the four survival-specific archives above
sh get-library.sh core
# or hand-pick several: sh get-library.sh wikimed gutenberg wikibooks
Every archive lands in ./zim/. If kiwix-tools is installed (brew install kiwix-tools on macOS, apt install kiwix-tools on Debian/Raspberry Pi), the scripts also build zim/library.xml automatically — a combined catalogue so kiwix-serve --library zim/library.xml serves every downloaded archive from one page with search across all of them, instead of opening each .zim individually in the Kiwix desktop app.
ZIM files and Kiwix are produced by the Kiwix non-profit and are free to use and share. Wikipedia content is CC BY-SA; Project Gutenberg texts are public domain. Browse the full catalogue at library.kiwix.org.