4406 entries. 94 themes. Last updated December 26, 2016.

1. From Gutenberg’s Movable Type to the Digital Book, and Other Studies in the History of Media

1.A. Relating the Rapidly Changing Present to the Distant Past, as Far as Book History is Concerned

Those familiar with the early history of printing, and its impact on European society in the fifteenth and sixteenth centuries, may have occasionally entertained the possibility that the transition from print to digital we are presently experiencing is a kind of fast-forward replay on a multi-dimensional scale of what happened after Johannes Gutenberg invented printing by movable type over five hundred years ago. Will the Internet and digital books eventually make printed books obsolete the way that printing eventually made manuscript copying obsolete? Could studying what happened in the late Middle Ages and earlier somehow provide insight into the present rapid change in the form and function of the book? Since these questions represent the intersection of two of my core interests--computing and book history--they have definitely piqued my curiosity.

My 2005 book, From Gutenberg to the Internet, was the first anthology to reflect the origins of the various technologies that converged to form the Internet. Separate from the anthology which comprises most of the book, the introduction contains, among other things, a relatively brief introductory essay which compares and contrasts the transition from manuscript to print initiated by Gutenberg’s invention of printing by movable type in the mid-15th century with the transition that began in the mid-19th century from a print-centric world to the present world in which printing co-exists with the various electronic media that converged to form the Internet. Specifically it compares the transition from manuscript to print to the transition from print to the Internet and the World Wide Web.

While it was not difficult to write the brief 2005 essay comparing these developments in media separated by more than five centuries, the more I delve into the topics, the more I appreciate their complexity. Attempting to compare two multi-faceted transitions in media separated by more than five hundred years is, I suppose, one of those problems that some people might wonder about, but few would try to research in detail. As challenging as it may be, this is a problem that I continue to pursue. Since publication of From Gutenberg to the Internet my efforts to research this and other problems raised in that introduction more accurately and in more detail have resulted in this database, the freewheeling nature of which has, of course, led to much wider fields of investigation, from the earliest surviving elements of what we may call information, to the present.

Writing essays like these, while building a database that covers much wider ground, is a little like trying to solve discrete portions of a very large and changing jigsaw puzzle. By the time I think I have formulated arguments supported by sufficient evidence, further information comes along to alter the perspective. Still, as research progresses, certain elements of the overall puzzle are coming into clearer view. Having researched in and around this topic for years, I doubt that we can, with any accuracy, compare fifteenth century developments in media with those of our own time. What I used to characterize as an attempt at comparison I now characterize more as a study of analogies. There are many analogies between fifteenth century developments and our time, but whether these developments are truly comparable may be debatable, and building the arguments one way or the other is problematic. To review fifteenth century developments in the form and function of the book and those in our time you need to have a reasonable grasp of the numerous topics involved in the book history of both periods. This means an understanding on some level of medieval technology as it related to book culture, of current computer technology, and developments in between.

One of the Chester-Beatty papyrii, in roll form. http://www.timetrips.co.uk/hieratic.htm
You need some understanding of the medieval manuscript tradition, out of which printing by movable type evolved. To understand aspects of the history of manuscript books over the nine hundred years of the Middle Ages I think you need to learn about the earlier great transition in the form and function of the book--the transition from the roll to the codex that took place during the second through fifth centuries CE. This is a topic that I cover in Section C of the second of these essays on "Transitional Phases." And these are just some of the wider topics that come into play!

Looking at this challenging range of topics, from the earliest written records to digital technology, many factors make analysis difficult. One is the quantitative scale of evidence when we work between the recent and the distant past. Before electronic computing and the Internet, which began to have widely-recognized impact upon society as a whole from the 1980s, the history of information may often be studied through specific examples, or examples on a relatively small scale, while the numbers are often several magnitudes higher when we address issues relating to the Internet, especially since the development of the web, in the 1990s. This is the result of several factors. On the one hand so much of the record is lost as we move backward in time; on the other hand there is the growth of information, which occurred with the advance of societies and populations, but which became much more widely recognized as explosive since the Internet. The early history of information in places like ancient Mesopotamia, Egypt, Greece or Rome often means the study of individual items, or groups of items-- often, but not necessarily books-- or libraries with a finite, though not always quantifiable number of physical volumes.

The 31-line indulgence preserved in the Scheide Library at Princeton.
Both before and after the invention of printing in Europe about 1450 we may think in terms of books written by individuals, or libraries formed by individuals, or institutions directed by individuals or small teams. 1.A.1Though paper and printing on paper originated in China, and the process spread westward over many centuries, the earliest printing was stamped into soft clay, circa 2250 BCE, in Mesopotamia, long before the developments in China.

Printing, of course, signficantly increased both the quantity of information production, and the speed and scope of its distribution. These incremental increases, dramatic by the standards of the relatively slow-changing fifteenth century before printing, are difficult to relate to the instantaneous or nearly instantaneous speed of electronic distribution over the Internet. The most significant change was that fifteenth century printed editions resulted in multiple copies--usually hundreds of copies versus one or a handful at a time resulting from manuscript copying. By 1480, twenty-five years after the introduction of printing by movable type, the typical print run of a book is estimated to have been between one hundred and three hundred copies. By 1500 printing presses had been established in 282 cities in Europe. It has been estimated that at least 35,000 different editions were printed during the fifteenth century, of which approximately 28,000 survive in one copy or more. How these numbers might have correlated to the production of manuscripts is far more difficult to quantify, but we can review the estimated time involved in preparing a hypothetical or specific manuscript versus the time involved in producing a hypothetical or specific book. To do so means analysis of numerous production stages, some of which I will review later.

Relative to speed of early printed book production, one way we can follow advances is to consider the speed of the printing component, which took place after typesetting and make-ready-- time-consuming processes about which I have not seen any early time calculations. Speed of typesetting and make-ready would have depended upon the efficiency of individual workers, and would also have varied from project to project, while pulling a hand-press was primarily a mechanical process which remained fairly constant for the first two hundred years after its introduction. In the mid-18th century a competent printer could expect to print on average, about 200 sheets per hour. Enhancements to the technology by the end of the eighteenth century, such as using an iron Stanhope hand-press-- more rigid than wood-- allowed the forme to be printed by one pull of the press instead of two, but output increased to only around 250 sheets per hour.

With the growth of literacy in the nineteenth century, and the widening of newspaper circulation, there was strong demand for increased printing speed to produce more and more copies of daily newspapers, some of which produced morning and evening editions. 1.A.2James Moran's Printing Presses, History and Development from the Fifteenth Century to Modern Times (1973) provides useful analysis of the advances in printing press technology and press output. The first edition of Moran's book is also a superbly designed, printed and bound monograph. Invention of the steam-powered press by Friedrich Koenig in 1810-1813 provided the first truly significant increase in speed first exploited by The Times of London newspaper, which in November 29, 1814 published its first issue printed on a double steam-driven Koenig cylinder press. The output of the new machine was initially 1,100 sheets an hour—more than four times that of the manually operated press previously used by the newspaper.

The Cowper & Applegath printing press.
By 1827 Edward Cowper & August Applegath in England completed the design of a four cylinder steam-powered printing press with capacity of 4,000-5,000 impressions per hour. In 1868 The Times of London newspaper installed a Walter press, developed by the owner of the newspaper, John Walter, that printed on continuous paper, further increasing the speed of production. The next major advance occurred only seven years later, in 1878, when J.G.A. Eickhoff built a four-cylinder perfecting press, capable of printing two sides of paper simultaneously. Since then newspaper printing technology has made great strides with full color newspapers, some of which in Asia are printed in millions of copies per day. Today's fastest double-width offset newspaper press, tall as a four-story building, may be the Mitsubishi DIAMONDSTAR 90 with a printing speed of 90,000 full color, 96-page broadsheet copies per hour. 

Since the way I approach some of these problems is through referencing surviving examples of primary sources, there is a disconnect when we relate interpretations of the past, based on sometimes unique works, and relatively finite or paucity of evidence-- very small editions, quantities most often gradually increasing with the advance of time-- to issues of modern media and computing and the Internet, which require us to think on an almost unimaginably larger scale, explosive growth rates, and usually too much evidence: hundreds of millions or billions of people, millions of books, a few of which are printed in millions of copies, trillions of URLs, as of January 2011 over 10 billion apps downloaded from Apple's app store, which opened only on July 10, 2008. In May 2011 Skype, founded in 2003, was purchased by Microsoft for $8.5 billion. At the time of purchase Skype, which had yet to turn a profit, was growing at the rate of 500,000 new registered users per day, had 170 million connected users, with roughly 30 million users communicating on the Skype platform concurrently. Volume of communications over the platform totaled 209 billion voice and video minutes in 2010. As large as these numbers are, they have precedents: the development of electric and electronic media, beginning in the 1830s with the development of the electric telegraph, stimulated more and more communication. If we go back to the end of World War II in 1945, the year in which telegraphic use peaked in the United States, Americans sent 236 billion telegraph messages that year, seeming a huge number relative to U. S. population at the time. With respect to the amount of information transferred, numbers may be deceptive since telegraph messages were charged for by the word, and tended to be exceptionally brief, while the amount of text, audio and video information that can be transferred or exchanged in one minute on the Internet is incomparably greater than the amount of text that could be exchanged in the same time by telegraph. Because of the availability of increasingly rich and diverse information over wireless networks, the nature of telecommunication has changed. As of May 2010, cell phones, used by about 90% of American households, were used more for data, such as text messages, streaming video and music, than speech, and during 2008 to 2010 the average number of voice minutes per user in the United States fell. In his book, The Information. A Theory. A History. A Flood (2011, p. 395), James Gleick quotes Jaron Lanier dramatically describing the scale of the ever-accelerating flood of electronic information we are experiencing: "It's as if you kneel to plant the seed of a tree and it grows so fast that it swallows your whole town before you can even rise to your feet."

To attempt to compare the rapidly changing present with the late Middle Ages or the more recent or more distant past, we often have to move back and forth from limited evidence viewed through history's microscope, more accessible to individual interpretation, to quantities of information that may be most efficiently viewed through the telescope of the algorithm and the search engine. This exponential increase in quantity of information confirms what all of us feel: that the web has been the game-changer. And yet these problems of scale, though enormously increased, are not new, as Denis Diderot raised aspects of them in the mid-eighteenth century when the number of printed books seemed to be getting out of hand. Even when Diderot wrote the problem of information overload was by no means new to scholars. Long before the invention of printing, when manuscript books were comparatively, expensive, scarce and often difficult to obtain, scholars who demanded the most intimate understanding of long texts expressed concern that the number of books being produced was becoming excessive, and worried about their quality.

A statue of the Muslim polymath Ibn Khaldun in Tunis.
To cite only one example, which happens to come from medieval Arabic literature, in The Muqaddimah, written in 1377, the Arab polymath and historian Ibn Khaldun worried about the great number of scholarly books then available in every field, so many that they could not be read in a lifetime. He recognized that the existence so many books resulted in the need for summaries in textbooks, which he understood served useful purposes. Summaries and condensations, as necessary as they were, he considered detrimental to scholarship and the acquisition of good study habits. True scholarship, he believed, required painstaking study of long and detailed works over a considerable period of time.1.A.3 Rosenthal, Franz, " 'Of Making Many Books There is No End:' The Classical Muslim View," Atiyeh (ed) The Book in the Islamic World. The Written Word and Communication in the Middle East (1995) 45. Ibn Khaldun's specific words on these topics may be found in the one-volume condensation of Rosenthal's translation of The Muqaddimah, abridged and edited by N. J. Dawood (2005) Chapter 6, section 34, the title of which is "The great number of scholarly works available is an obstacle on the path to attaining scholarship," and section 35, which has the title "The great number of brief handbooks available on scholarly subjects is detrimental to the process of instruction" (pp. 414-416).

Until the development of analog recording devices such as photography, cinematography, microfilm, sound recording, and audio and video tape in the nineteenth and twentieth centuries, books and archival records remained the primary means of recording and distributing information. For that reason what we know about the early history of information, and related topics like early views on information overload, from the earliest writings, through the manuscript period, and even through the first half or more of the twentieth century, comes primarily from the physical books or documents that survived. These tend to be the work of individual authors or identifiable groups of writers. That for most of recorded history surviving historical records are primarily in book and archival form explains why book history holds such a central place in the history of information, especially before the growth of analog and digital telecommunication and storage systems in the twentieth centuries. Apart from my life-long focus on books and their history, the central role of the book in the early history of records, particularly up to the twentieth century, may help explain why I concentrate my narrative and analysis of this panoramic database in this way.

Methods of comparing the growth rates of electronic information versus physical or book or archival information were developed from social science techniques. Toward the end of the twentieth century, with the rapid expansion of television, for example, researchers suspected that the growth of electronic information was surpassing the growth of information recorded and distributed by print. To study very large information flows new statistical research methods had to be developed, focusing on large-scale quantitative measurement rather than individual works. Social scientist Ithiel de Sola Pool measured the rapid growth of electronic media, including the growth of television, relative to the slower growth of print media from the 1960s. His pioneering study published in 19831.A.4 Pool, I. de S."Tracking the Flow of Information," Science 221 (12 August 1983) 609-13. Reprinted in Etheridge (ed), Politics in Wired Nations. Selected Writings of Ithiel de Sola Pool (1998). confirmed that for at least the past five decades books and print media represented only a very small, though very significant, diminishing percentage of total information flow. I believe that it is reasonable to assume that the percentage of stored information recorded by books may remain small, yet very significant relative to content, whether books are printed or digital. In a study published in 2011 Hilbert and Lopez show that books represented only 1% of the world's stored information as early as 1986. 1.A.5 Hilbert & Lopez, "The World's Technological Capacity to Store, Communicate, and Compute Information," Science, 332, 60-64, figure 2. "Paper-based storage solutions captured a decreasing share of the total (0.33% in 1986 and 0.007% in 2007), even though their capacity was steadily increasing in absolute terms (from 8.7 to 19.4 optimally compressed petabytes)" p.62). Regarding electronic information the authors state that 2002 was the first year that worldwide digital storage capacity overtook total analog electronic capacity, and that by 2007, almost 94 percent of the world's stored information was digital form. Not surprisingly, perhaps, in May 2011 the bulk of Internet traffic (49.2%) in North America was real-time entertainment applications: chiefly film downloads with their very large files, and it was predicted that entertainment downloads might represent 60% of Internet traffic in North America by the end of 2011. In June 2011 it was reported that social networks were used by 90% of U.S. Internet users, for an average of more than four hours per month. This growth in Facebook and other social networking sites was taking web users away from the so-called "document web" which includes documents such as digital books, or essays like the one you are presently reading.

Is it ironic that indexing and searching technology for digital information assists research on analog records, including books? Data scanned from print into the ocean of bits and bytes becomes more quantifiable, more readily accessible, in the sense of an encyclopedia article, an image, or in some cases an online digital facsimile, assisting us far more than hindering research, making my project possible. But, as some scholars observed as long ago as the Middle Ages, access to increasing quantity inevitably raises issues of overload, of quality, requiring efforts in focus, and evaluation. What may be most difficult in comparing old and new technologies with respect to these very general topics is the necessity of thinking about the unique and special, such as a famous illuminated manuscript or papyrus roll, or small groups of people listening to a book being read aloud in the ancient world, and then thinking about the opposite of unique-- the socially pervasive, Internet social media issues, and back again, requiring frequent change of focus. We may also observe, however, that in spite of the exponentially inscreased quantity of data, and the greater number of people often involved in projects today, individuals still play key roles. We can still tell their personal story.

Levels of complexity also complicate comparisons between earlier and current information technologies. As technical and difficult to learn as professional writing must have been in the ancient world or the Middle Ages, or however complicated the process of medieval manuscript illumination, or hand-production of papyrus or parchment or paper, or the mechanical process of printing by movable type must have been, these were arts that individuals or small groups, with effort and practice, could eventually master from the ground up. The computers we use today, and the software we typically run on them are so complex, that their hardware and software require large teams of engineers for design and programming. In 2001 the operating system Windows XP contained 45 million source lines of code; I have not been able to find data regarding the size of more recent versions of Windows. Thus even though the end products, such as the HTML editor I have been using to write these essays, facilitate individual use, and an individual may still write a book that can be printed on paper, published on the web like this, or issued in ebook format for ebook readers, the development of software for electronic composition, for display on ebook readers, and the design and production of ebook readers themselves, require numbers of people and capital investment magnitudes greater than they would have prior to electronic computing, the Internet, and the development of digital books. Though the end result of programming is to simplify, or at least to facilitate operations, the size of code to attain those goals may be immense.

The Chevrolet Volt, whose complicated machinery requires approximately 10 million lines of code to run.
An article published in the New York Times on January 21, 2011 mentioned that it takes eight million lines of code to run Boeing's new 787 Dreamliner, and it takes ten million lines of code to run the new Chevrolet Volt plug-in hybrid automobile. Inevitably, as programs become ever larger and more complex, the debugging process becomes more and more difficult, often resulting in bug fixes or updates, adding further layers of impenetrability, especially as compared to mechanical processes. 1A.6Assuming these numbers are correct, considering the difference in scale and presumed difference in complexity between the two devices, one has to wonder if the code on which General Motors is initially running the Volt may eventually be simplified, or if the Volt is actually more complex from the programming point of view than a huge airliner.

Other elements of complexity are the multi-media aspects and interactive features of the web relative to individual broadcast media. By broadcast media I refer to the traditional media such as print, television and radio, that transmit information from its source to a wide range of receivers: one way transmission, as compared to social media which combine broadcasting with reader or viewer feedback and comment sharing. Since its establishment in January 1996 The New York Times interactive online version passed the limitations of print and images on paper to include audio and video in many articles. You can watch television from their website, and you can interact with the columnists by entering your comments online either on the newspaper website or social media sites. These video and audio features turn up in many places on the web. From the standpoint of book history, in 2010 The Grolier Club of New York, the leading American book collector's club, keeping up with current technology, began making videos of selected meetings and events available from their website to members who are unable to attend, using new technology to promote interests which are frequently antiquarian in nature. Using new technology for historical research is, of course, also what I am doing here. When we trace media back to their roots we follow the histories of individual media such as manuscripts, print, photography, tv, movies, or radio. From the mid-20th century we may follow the development of experimental projects in multimedia and virtual reality. While we may find partial precursors, there is a disconnect when we try to relate the history of different broadcast media, among which I include print, to the immense diversity and interactivity of the Internet media experience, which allows hundreds of millions of people to produce online content of all kinds, and to interact through social media.

There is also a disconnect with respect to networking and information transfer, both in speed and amount of data. In the ancient world messengers on horseback, riding in relays on the Royal Road, as described by Herodotus, could travel 1,677 miles in seven to nine days. At that time an army on foot might have taken three months to cover the same distance.

The Chappe optical telegraph.
Until the development of the Chappe optical telegraph at the end of the eighteenth century, relay riding, especially over improved roads, remained the fastest method of communication over extended distances. Chappe's semaphore technology could transmit a 36-symbol message between Paris and Lille (about 230 kilometers or 143 miles) in about 32 minutes, or a little less than a minute per character, significantly faster than relay riders could cover the distance. The system eventually expanded to 556 stations covering and 4,800 kilometers, and remained in use for military and national communications until the 1850s when it was superceded by the electric telegraph. The Morse code for the electric telegraph allowed another incremental advance in speed, but like the optical telegraph, remained a method of transmitting individual characters or words by code, except over electric wires rather than line of sight--a major advantage in speed and efficiency, since messages were not delayed by low visibility caused by overcast, could also be sent in the dark, and could go longer distances between relay stations. Like all information fields, many technical advances in theory and practice caused speed of communication to accelerate, to the present level of essentially instantaneous transmission of small files, high speed downloads of large files like the one you are reading, speed enhancements being one of the most frequent "new and improved" announcements we receive, as we approach light speed with optical transmission. With so much of the world networked, all of us have access to a quantity and diversity of digital information that is unprecedented and incalculably larger and more diverse than anytime before. Without the space limitations of print, media can publish enormous texts and keep them archived for easy access. In 2010-11 Wikileaks, founded as recently as December 2006, could expose hundreds of thousands of secret documents, and newspapers could publish their selections without concern to the restraints of size or weight of their publications or the cost of paper. 1A.6 "Whether the arrival of WikiLeaks has fundamentally changed the way journalism is made, I will leave to others and to history. Frankly, I think the impact of WikiLeaks on the culture has probably been overblown. Long before WikiLeaks was born, the Internet transformed the landscape of journalism, creating a wide-open and global market with easier access to audiences and sources, a quicker metabolism, a new infrastructure for sharing and vetting information and a diminished respect for notions of privacy and secrecy. Assange has claimed credit on several occasions for creating something he calls “scientific journalism,” meaning that readers are given the raw material to judge for themselves whether the journalistic write-ups are trustworthy. But newspapers have been publishing texts of documents almost as long as newspapers have existed — and ever since the Internet eliminated space restrictions, we have done so copiously" (Keller, "Dealing with Assange and the Secrets He Spilled", The New York Times Magazine 01-20-11).

There is also the matter of scale with respect to time. Change in media occurred slowly in the ancient world and the Middle Ages, sometimes requiring centuries for measurable change, beginning to accelerate with the introduction of printing by movable type in the mid-fifteenth century, advancing inexorably, but measurable during that so-called incunabula period of printing from 1455 to 1500, more in terms of a decade, or multiple decades, than in terms of sudden disruptive events. Inevitably, disruptive events did occur during the late Middle Ages, as they did throughout history, such as outbreaks of plague, or in the first decades of the sixteenth century, the Protestant Reformation. Epidemics impacted society one way, ideas another; the contagion of reform spread by Luther's prolific publications resulting in the Reformation, impacting religion and society by its occurence. Yet the rate of disruption, as dramatic as it must have been at the time, seems, at least through the lens of history, to have been less frequent than now. Because most of us are connected by the web, we learn about new developments in information technology and digital books often, experiencing along with the barrage of available information a glut of change disruptive by its almost constant introductions of "new and improved." But, if we step back and look for precedents we may find them.

The Google N-Gram viewer, which allows users to track cultural trends based on the frequency of appearance in the Google Books database.
For example, the existence of the searchable texts of millions of books is being exploited by scientists and scholars in ways that might not have been imaginable just a few years ago, leading to entirely new approaches to the history of books and libraries, including the emerging field of Culturomics emanating from the Cultural Observatory at Harvard. On December 16, 2010 a highly interdisciplinary group of scientists and scholars from Harvard, MIT, Google, Houghton Mifflin publishers and the Encyclopaedia Brittanica published "Quantitative Analysis of Culture Using Millions of Digitized Books" in the journal Science. These new explorations of astronomically large amounts of digital text are outgrowths, on massively increased scale, of automated textual researches originated by Father Roberto Busa with the support of IBM as early as 1949. To me there is a beautiful irony in having what was later called humanities computing-- at the time a radically new approach to scholarship using cutting-edge technology-- originated by an Italian Jesuit priest for creation of an index verborum of 11 million words of medieval Latin mainly written by the thirteenth century scholastic philosopher and theologian St. Thomas Aquinas. Father Busa used innovative technology to understand the past. Faced with the barrage of innovation and change that we experience, it has been my life-long habit to search for historical perspective, to understand the present more clearly by understanding the past.

Created from both physical and digital sources, From Cave Paintings to the Internet has become a kind of personal research laboratory for the history of the digital and physical approaches to information and media. Some of the themes focus on digital information, some on the development of broadcast media, some on physical information such as book history; some are otherwise characterized. You will find that the data is searchable by a virtually unlimited number of parameters. Creating this database is one thing; using it is another. As much as I have immersed myself on these topics for nearly the past ten years, and I have designed the database to be accessible from many different points of view, my overriding preoccupation remains the relationship between past and present transitions in media, and the development of information technology as it relates to the history of books, libraries, and digital information. In spite of the challenges, some of which I have outlined, I find this a fascinating way to approach history in general, and I hope you will too. From my ongoing research I will present my latest work on the comparison of the transition in media which we are presently experiencing with that which occurred in the fifteenth century. Then I will review transitions in media before Gutenberg (Chapter 2). After that I will discuss the history of information searching--without which I could not have undertaken these studies-- from its origins in ancient libraries and archives to the development of web search engines (Chapter 3). Please bear in mind that all of these essays are works in progress, and that the earlier essays in this sequence are better developed than the later ones. (This section was last revised on August 6, 2011.)

back to top