Eo Ena — narrator
STOUTENBURGER ALMANACS.B.A.
Rebuilding The World — Edition 1
Knowledge
K Bread·Grey·Wren RTW-002 Basic

The Digital Library of Alexandria

The modern library of Alexandria is not a building — it is a constellation of servers you do not own, running software you do not control, storing knowledge you only rent. And it is already burning, in slow motion.

The Essence

  • The internet was designed as a decentralized network — no single point of failure. What we built on top of it is the opposite: a handful of companies control most of humanity's access to information. Google, Amazon, Microsoft, and Meta host the majority of the world's data on servers concentrated in a few dozen locations.
  • GeoCities contained 38 million user-built pages when Yahoo deleted it in 2009. An estimated 80% of the knowledge shared there — tutorials, hobbyist guides, community wisdom — was lost permanently. This was not an accident. It was a business decision.
  • Google Reader, Google Code, Google Plus, Google Health, Google Stadia, Google Domains — each contained user-generated content and data. Each was shut down when it stopped being profitable. The pattern is not a conspiracy. It is a business model.
  • Academic knowledge is locked behind paywalls. Elsevier, Springer, and Wiley control access to the majority of peer-reviewed research. A single journal subscription can cost a university tens of thousands of dollars per year. Individual access to a single paper: thirty to fifty dollars. The researchers who wrote the papers are not paid by the publishers.
  • Link rot is measurable. A 2024 Pew Research study found that 38% of web pages from 2013 were no longer accessible. The half-life of a URL is approximately two years. Knowledge that exists only as a link is knowledge with a countdown timer.
  • The Internet Archive — the single most important effort to preserve the digital commons — operates on donations, has no government mandate, and was successfully sued by publishers in 2023 for lending digital books. The only institution trying to save the digital library is under legal and financial siege.
  • Terms of service give platforms the right to modify, restrict, or delete user content at any time. You do not own what you post. You license it to a company that can revoke the license unilaterally.
  • Cloud storage is not backup. It is access to someone else's computer, contingent on an active account, a working internet connection, a solvent company, and compatible software. Remove any one of those four dependencies and the data is inaccessible.

Why It Matters After the Fall

You do not need a catastrophe to lose digital knowledge. You need only a quarterly earnings report. Every platform shutdown, every paywall increase, every terms-of-service change is a small fire in the digital library. The smoke is already visible: broken links, deleted archives, paywalled research, expired domains.

In a crisis — power grid failure, economic collapse, network disruption — the concentration of knowledge in cloud services becomes an existential problem. Communities that depend entirely on internet access for practical knowledge (how to purify water, grow food, treat injuries, build shelter) will find themselves functionally illiterate.

The remedy is the same as it was in antiquity: make copies, distribute them widely, and ensure they work without the network.

Minimum Starting Conditions

  • A device capable of storing files offline (USB stick, hard drive, SD card, or printed paper)
  • The ability to identify which knowledge is critical for your community
  • Basic understanding that "saved online" and "backed up" are not the same thing

The Process, In Plain Stages

  1. Recognize the dependency — List the knowledge you access online that you could not reproduce from memory. Recipes, medical information, repair manuals, growing guides, technical references. This is your vulnerability surface.
  2. Download and convert — Save critical resources in open, durable formats. HTML, plain text, PDF, or print. Avoid proprietary formats that require specific software to read. A .docx file is a ZIP archive of XML — it requires Microsoft's schema to interpret. A plain text file requires nothing.
  3. Distribute physically — A USB stick in a drawer is more durable than a cloud subscription. Give copies to friends, family, community members. The value of a backup is proportional to its distance from the original.
  4. Print the irreplaceable — Paper requires no electricity, no software, no internet connection, no account. A printed manual in a waterproof bag is the most robust knowledge storage technology ever invented.
  5. Maintain and update — Knowledge changes. Check your offline copies periodically. Replace outdated information. Add new knowledge as you acquire it. A preserved library is not a frozen library.

How To Verify Without Instruments

  • Search for a website you used five years ago. Is it still there? If not, do you have the information it contained?
  • Try accessing an old email account. Can you still log in? If the provider shut down (Hotmail to Outlook, Yahoo Mail changes, university email after graduation), is the content recoverable?
  • Open the oldest digital file you own. Can your current software read it? If it is a format from the 1990s or early 2000s, the answer may be no.
  • Check a URL you bookmarked more than two years ago. According to measured link rot rates, there is roughly a one-in-three chance it is broken.

Common Failures & Recovery

  • "But Wikipedia is free and permanent" — Wikipedia is maintained by volunteers and funded by donations. Its servers are centralized. A local copy of Wikipedia (available via Kiwix) fits on a 90GB drive and works completely offline. Download it.
  • "I saved it to Google Drive" — Google can suspend your account at any time, for any reason, including automated false-positive policy violations. When your account is suspended, all associated data — Drive, Gmail, Photos — becomes inaccessible simultaneously.
  • "The government archives will preserve it" — Government digital archives face the same format obsolescence as everyone else. The US National Archives has acknowledged that significant quantities of born-digital records are at risk of becoming unreadable.
  • "Blockchain will solve this" — Blockchain is a ledger, not a library. It can verify that a piece of data existed at a given time. It cannot store, index, search, or serve significant quantities of human knowledge. The costs per byte of storage are orders of magnitude higher than any alternative.
  • "AI knows everything" — Large language models are trained on snapshots of the internet. They do not have access to knowledge that was deleted before training, paywalled content they could not scrape, or information created after their training cutoff. They hallucinate with confidence. They are not an archive.

Hazards

  • Digital dark age — We are producing more data than any civilization in history and preserving less of it as a percentage. Future historians may know more about the 15th century than the early 21st century because print books survive and social media posts do not.
  • Platform lock-in — Migrating from one platform to another becomes harder over time. Data export tools are limited by design. The longer you stay, the harder it is to leave, and the more you lose when the platform leaves you.
  • Single sign-on fragility — If you use one account (Google, Apple, Facebook) to log into other services, losing that account means losing access to everything connected to it. A single credential failure cascades across your entire digital life.
  • Subscription fatigue — Knowledge behind subscriptions (news, research, tutorials, software documentation) disappears the moment you stop paying. You are not building a library. You are renting shelf space that can be reclaimed at any time.

Sources / Lineage

  • Zittrain, Jonathan — The Future of the Internet and How to Stop It (2008), ISBN 978-0-300-15124-5
  • Rumsey, Abby Smith — When We Are No More: How Digital Memory Is Shaping Our Future (2016), ISBN 978-1-63286-259-4
  • Kahle, Brewster — Universal Access to All Knowledge (Internet Archive, ongoing)
Sources
  • Zittrain, Jonathan — The Future of the Internet and How to Stop It (2008), ISBN 978-0-300-15124-5
  • Rumsey, Abby Smith — When We Are No More: How Digital Memory Is Shaping Our Future (2016), ISBN 978-1-63286-259-4
  • Kahle, Brewster — Universal Access to All Knowledge (Internet Archive, ongoing)
↔ See also — related cards
K Why Knowledge Disappears
Print Verify Index