We are now in the middle of an information-technology upgrade. This is rare in human history. The one we're currently experiencing, the change from analog books to digitally encoded text is the first such upgrade since Gutenberg. There was another major upgrade 1600 years before Gutenberg, and that was the switch from the ancient scroll to the codex. Starting a couple centuries before Christ, important books began to be copied from papyrus scrolls over onto sheets of parchment. By the first century AD such sheets were routinely bound together into what we'd recognize as a "book" form. This form was called a "codex". This form wasn't adopted for convenience. No one had a problem with the scroll form, and there was no conception of it as cumbersome. The Codex was a necessary response to Egypt's decision to stop export of its papyrus. Egypt decided to prevent the use of papyrus outside Egypt. This was no attempt to drive the price up, more like an attempt to impede the cultural development of the rest of the world. People turned to parchment. Parchment is made of animal skin. Goats and sheep were ubiquitous, safe from being monopolized. There was that advantage, and the fact, too, that parchment was more durable than papyrus. Still, it was very expensive: one codex might require 25 goat hides. Goats and sheep aren't rectangular and there are only so many good rectangular pieces to be cut from one hide. A long scroll of animal hide with pieces stitched together turned out not to be feasible. The resultant compromise was the form of the codex; folded and bound rectangular pieces of parchment, held between two flat wooden "covers". What we perceive as obviously more convenient a form was actually just the ancients doing the best they could when prevented from doing it the way they'd have preferred. The letters of Archimedes to his mathematical colleagues had already been copied many times over on papyrus and disseminated throughout the Mediterranean world. He was already famous and revered. It was natural such important books would be recast as codices when the technology upgrade was necessary. As the old papyrus scrolls wore out and needed to be recopied, they were recopied in codex form on parchment. Archimedes jumped that hurdle with ease. Other ancient authors did not. We probably lost scores of names and huge amounts of information about the ancient world in the transition from scroll to codex. It was not worth it to just copy everything. A codex was expensive and labor intensive. Here, though, after the transition of his works to codex form, Archimedes' luck runs out. The codices containing his works suffer centuries of attack from every conceivable angle, the worst being pillaging hordes of barbarians (who always seem to vandalize great libraries when they visit). One by one copies of the works of Archimedes are destroyed or lost. In some cases the parchment on which they were written was deemed vastly more valuable than the text, and the parchment was put through the recycling process known as "palimpsesting", meaning "re-scraping". At the time of the Gutenberg upgrade finding an Archimedes on papyrus was absolutely out of the question and only two Archimedes codices remain: Codex A, and Codex B. They don't exactly match, but, between the two of them there's enough agreement to get Archimedes over the most important hurdle of all: the transition to mass print. An extremely happy ending. Except then Codex A and Codex B disappear. The oldest known sources of Archimedes' ideas can no longer be checked and scrutinized in light of new finds, new information from other sources. Greek historians, math historians, literary historians, historians in general, are thwarted from ever getting deeper into the subject. All this I digested from the book The Archimedes Codex by Reviel Netz & William Noel. I'm pretty sure by now everyone knows what happened next. I just finished the book and I was completely fascinated. There was lots in it I wasn't expecting, as well as things you would expect: it takes you on a tour of the difficulties of translating ancient Greek, of turning a goat hide to parchment, of tracing out the provenance of a book no one knew existed, of putting together a team of experts to manage such a find, and, above all, of trying every possible modern technology to coax the ghost of a mathematical treatise into manifesting itself from a moldy, centuries old piece of animal skin, a treatise that had been dissolved in weak acid, scraped off, and then written over. The Archimedes Codex, (which refers to newly discovered Codex C, now the oldest known extant source for the works of Archimedes) is currently on display at the Walters Museum in Baltimore, which was the hub of all the efforts of the last 13 years to image the treatise. The Codex was bought at auction in 1998 by the mysterious Mr. B., who turned it over to that museum for analysis. No one knows anything about Mr. B except that he's "as rich as Croesus". (My main suspect was Steve Jobs, but I suppose they'd have revealed him as the guy who popped 2 million for the book now that he's passed away.) As of this time, the identity of the filthy rich private owner of the Archimedes Codex still remains a mystery. Anyway, I haven't spoiled anything and strongly recommend the book. I was highly entertained during the four days it took me to read it, and disappointed when it was over.