Universal Access to All Human Knowledge

Brewster Kahle speaking at the Library of CongressA link found from Matt Haughey’s a.wholelottanothing.org lead me to a talk by Brewster Kahle of the Internet Archive. His organization is working a variety of projects to make public domain content available in an “internet library”. Among these projects is the WayBack Machine, which archives the web.

The talk is part of a series at the Library of Congress and runs 1 hour and 26 minutes in RealVideo format. It is worth watching: Brewster Kahle: Public Access to Digital Materials (1hr 26min RealVideo).

Kahle’s basic idea is universal access to all human knowledge. Every book, speech, TV show, website, concert, etc. should be available to all of us. He looks at three main questions:

  • Should we? Yes!
  • Can we? Is it logistically possible from a technical and financial perspective? His answer: Yes.
  • May we? Will we be allowed to make all knowledge available under law? His answer: Yes.
  • Will we? He leaves this as an open question.

His numbers on the cost to digitize (scanning, etc.), store (disk space), and make available (bandwidth) all human knowledge are fascinating. According to Kahle, the hardware and labour costs required to make all book and all television and all music ever created available are not that difficult (within the hundreds of millions of dollars).

Taking books for example:

  • There are roughly 100,000,000 books ever created
  • The Library of Congress has about 26,000,000 books (I was impressed and amazed that the Library of Congress has 26% of all books ever created)
  • A book costs between $10 and $100 to acquire and digitize
  • A book takes up about 1Mb of space
  • 26,000,000 books would take 26 TeraBytes.
  • 1 TeraByte costs about $60,000
  • The entire Library of Congress could be stored for about $1.5 million dollars
  • Books can be printed, cut, and bound for $1/book from a mobile book printer (~$15,000)

If anyone has the right to make these claims – it would be Kahle – who’s organization is storing massive amounts of data as part of their WayBack project and other projects.


5 thoughts on “Universal Access to All Human Knowledge

  1. Much of what exists in books is already catalogues, scanned and available. Dal or UNB has, for example, microfilms of every document printed (including pamphlets) before 1800. I almost did an MA in English based on the collection. I suspect that the same is true for the major languages. I learned through font-man that there are still languages for which fonts do not exist and that there is a project to remedy this before deaths or digitization destroy the record. I would suspect that digitization and internet presentation of these sorts of collections is entirely within reality.

  2. What about translation of all of this content? Does it get digitized in it’s original form, or made available in multiple languages?

  3. I think translation could be a separate project. If you digitized all currently published material, you’d already have a lot of the translation done for you (having already been done for print).

  4. Universal access to all human knowledge would be wonderful, and I think we should strive for it but… It would be almost impossible to achieve. Even if you managed to store ‘all human knowledge’ it would never be accessible to all people unless there was world equality and everyone could afford a means of accessing it. It also raises questions about privacy and security: “Do I want you to know where I live?” and “should I tell the world how to make a 100megaton bomb out of a lump of coal, a paperclip and 6 litres of Sprite?”.

Comments are closed.