Home | Intro | Background | Participants | Links | Discussion | Proceedings
Responses to How Much Information Is There In the World? by Michael Lesk:
- Philip Morrison, Physicist, Massachussetts Institute of Technology:
There is a good record of the first effort I know of to collect all human information in print anyway. It was begun in October 1945 by James B. Conant, president of Harvard.
Fearful of the eventual destruction of most world cities in the aftermath of the WWII nuclear war, he thought it worthwhile for the Widener librarian to study very privately what that would mean. That scholar came up with an estimate and a caution; it would take 2.5 billion printed pages to do the job for all languages in print. That is only tens of terabytes, plus an equal allowance for images? That was fifty years back, so extrapolation might give a tenth or more of a petabyte.
But he cautioned that it could not fail to alarm the entire scholarly world. Was that risk worth it? Conant dropped his idea, suppressed the story from all his books, until it came out only in late 1993 in a fine biography, James B. Conant: ...,written by James G. Hershberg, A.A.Knopf,1993. (NYTimes review by me in its Sunday review section, probably early Jan 94.) So much for precedent!
My own reading of the estimate (by Michael Lesk) in the Getty meeting papers would increase it to include the info represented by all the craft skills in the world. That is mainly not now symbolized at all, but present and testable by trials, of cooks and cowherds and ceramists and carpenters and musicians and sailors and rice farmers and every mechanical and chemical technician in the world. I do not think it would change the overall estimate by a factor ten, but maybe by that much.
History will rest as always on search of old MS and old books, but now go on to old DRAM and old discs forever, with all sorts of new scholarly tasks. Why not? The archives of well-sold storage devices are the best things to keep, in addition to whatever the libraries choose to save. Their best help is a good guide to as many as may be of storage inventions and the data left by those who apply and sell them in major countries. I will bet that most of all the petabytes are network check and currency and ATM transactions, plus human face recognition, plus those petabytes a year of talk and gestures etc.
Our truly wizard posterity will be quite able to restore the reading devices of anything worthwhile that they may find, or I would be much surprised. The whole tale is an archivist's dream, and probably oxymoronic. But great fun !
* * *
Home | Intro | Background | Participants | Links | Discussion | Proceedings
