Positive Liberty

I was a luddite in the last post; now I’m going to be a technophile.

This week I attended the Digital Library Federation’s 2004 Fall Forum, where I gave a presentation of the user’s perspective on the electronic media, focusing on some online work I’ve done in history. I spoke before a group of programmers, administrators, and electronic librarians. What follow are my remarks to the conference, but first I’ve got to clear up a very serious mistake I made in my previous post.

Marcela asked in the comments below for the source to my claim that SMS and other instant messages now surpassed all other media in data content; as I recall it, the claim came from the keynote address by John Unsworth, although none of the slides of his presentation state it explicitly. It may well be that both my notes and the speaker himself were mistaken: His cited source, “How Much Information,” by Peter Lyman and Hal R. Varian, gives more detailed numbers which do not bear out the claim. Instead, they indicate that telephone messages are still by far the largest source of new information. Here are the numbers from the executive summary of their report:

Telephone calls worldwide – on both landlines and mobile phones – contained 17.3 exabytes of new information if stored in digital form; this represents 98% of the total of all information transmitted in electronic information flows, most of it person to person.

Most radio and TV broadcast content is not new information. About 70 million hours (3,500 terabytes) of the 320 million hours of radio broadcasting is original programming. TV worldwide produces about 31 million hours of original programming (70,000 terabytes) out of 123 million total hours of broadcasting.

The World Wide Web contains about 170 terabytes of information on its surface; in volume this is seventeen times the size of the Library of Congress print collections.

Instant messaging generates five billion messages a day (750GB), or 274 Terabytes a year.

Email generates about 400,000 terabytes of new information each year worldwide.
P2P file exchange on the Internet is growing rapidly. Seven percent of users provide files for sharing, while 93% of P2P users only download files. The largest files exchanged are video files larger than 100 MB, but the most frequently exchanged files contain music (MP3 files).

In other words, it’s not even close, and someone clearly erred somewhere. My apologies for either originating or passing on the mistake.

In any event, here are my remarks from the conference. The title of the talk was “Engaging the User: The “Encyclopedia of Diderot and d’Alembert: Collaborative Translation Project” and New Scholarly Paradigms;” it was co-presented with Kevin Hawkins, who introduced me and outlined the future plans for the project:

I am not a computer scientist, or a programmer, or a systems administrator. I’m not even a librarian. Compared to all of you, I am probably the least computer-literate person in the room, if not in the entire building. I am here to give a user’s perspective on the digital media in the humanities, focusing on the Encyclopédie Collaborative Translation Project, on which I’ve done some work in the past. I would like to begin my remarks with a quotation that is very familiar in my own field and may well be in yours as well. In 1979, the great French historian Emmanuel Le Roy Ladurie declared that, “Tomorrow’s historian will have to be able to program a computer in order to survive.”

Twenty-five years later, I am tomorrow’s historian, and I still cannot program a computer. In one sense, it’s not surprising: Nowadays we distinguish between programmers and content providers, a distinction that was by no means clear in 1979. But as the gap between the two has widened, however, it has become more and more important to allow for effective communication between the programmers and the content providers. In the big picture, I fear that historians have been quite reluctant to become content providers in the digital media, and I believe that their reluctance is partly because of this growing divide. Only very recently and tentatively have historians embraced the digital media, and they have done so quite slowly even when compared to other disciplines in the social sciences and the humanities.

The issues and subject matter that historians examine have changed tremendously since the nineteenth century, but our methods of production and dissemination have more or less stagnated. In the last hundred years, we have created entirely new fields of history, including demographic history, labor history, the new economic and cultural histories, and the histories of gender and sexuality. But production and dissemination remain very similar to the model perfected with the rise of the modern research university in the nineteenth century: Historians gather information at physical archives; they digest information at universities; they lecture on it to students, and they write up texts with formats that are substantially similar to those produced in the nineteenth century. Their texts are almost always articles or books that are published in paper-based, peer-reviewed outlets.

Where computers have entered the discipline, they certainly have not revolutionized it–at least not yet. Structurally, the computer has replaced the typewriter, leaving the rest of the process more or less intact. Historians may take laptops into the archives, but notetaking and information transfer methods are frequently just digital re-creations of the old-fashioned paper notecard. I can recall that as an entering graduate student, I asked how I might better use computers in gathering my information. And–at a university that is not Johns Hopkins–I was told that I shouldn’t bother. A tenured faculty member and a well-known authority in the field declared in no uncertain terms that notecards were the best and really the only way to do research: Write the cards in pencil while you’re at the archives. Then come home, set the cards out on your floor, and arrange them into the story. To this day, it remains the way that many of us work.

As a tool for dissemination, we view the Internet with suspicion. Do a search for the Holocaust, and you will soon understand why this is. Generally speaking, what history can be found on the web often ranges from the mediocre to the mendacious. The peer review process has created very high standards in most published historical journals. But it’s powerless to weed out the Internet cranks, and let’s face it, there are a terrible lot of cranks on the Internet. Conspiracy theorists, promoters of racial hatred, plagiarists, and advocates of thinly evidenced “alternate” histories abound. In one term, a student actually submitted a paper that had been plagiarized directly from a Holocaust-denial website; the following term, I entirely forbade my students from using the Internet, and promised to fail them if they did. Even before a group such as this, I can’t honestly say that I regret the decision.

For those of us who know how to find them, however, there are some bright spots out there, and the Encyclopédie Collaborative Translation Project is undoubtedly one of them. The original Encyclopédie is arguably the most important document produced in eighteenth-century France; if anything, its importance is surpassed only by the Declaration of the Rights of Man and Citizen. While it was not the first encyclopedia, it was remarkable in its systematic approach, its wide scope, and its enunciation of the Enlightenment idea that all things are in principle knowable.

Generally speaking, the eighteenth century was the first age to share that ideal; it may well have been the last age where human knowledge was still small enough to render the project of compiling it all possible. This compilation is exactly what Diderot envisioned, and to that end, he engaged some of the brightest minds of his era: Besides himself and d’Alembert, the Encyclopédie featured articles by Rousseau, Montesquieu, d’Holbach, Turgot, and Voltaire. In today’s terms, this would be a little like arranging a collaboration among Stephen Hawking, Madeleine Albright, Richard Rorty, J. D. Salinger, and Andrew Lloyd Webber.

Historically, the Encyclopédie is the starting point for virtually all questions touching on the Enlightenment, as it set the tone for how an entire generation of philosophers approached the world. The work’s direct descendants, from Britannica all the way to Wikipedia, owe the Encyclopédie an enormous debt. And to speak still more generally, the project of the Encyclopédie is also the project of the modern library itself: The systematic classification of human knowledge is an idea that we now take for granted in large part because of the efforts of Diderot and d’Alembert. Thus the Encyclopédie’s importance to the historian is absolutely paramount.

But how are we as historians to teach this source? At 70,000 articles, very few have ever read the entire thing, and I’m certainly no exception. Making it accessible to a non-francophone general reading public virtually requires any presentation of the Encyclopédie to be rapidly browsed, scanned for interesting content, and parsed according the needs of the user. Only with the rise of the computer has such a project truly become feasible. The original Encyclopédie was arranged not alphabetically, but by the branches of general knowledge as understood in the 18th century. It is cumbersome and difficult even for specialists like me to use. Within the past few years, most of us have come to rely on French-language electronic versions of this text in part for this very reason.

The Collaborative Translation Project now features nearly two hundred articles in translation, and these offer an excellent starting point for non-francophone students to learn about the state of 18th-century knowledge in many—though certainly not all—different fields.

One of the most important factors determining the relative success of the project so far is that the Encyclopédie Collaborative Translation Project does not exist in a vacuum. One reason that historians trust the veracity of the translations so far is simply the good reputation of the contributors, a reputation that has mostly been built in more conventional channels. For example, Dena Goodman, was already a leading published historian of the French Enlightenment long before she took over supervising this project. The same is true of many others who have contributed articles; indeed, as an advanced graduate student rather than a full-time professor, I am a rather junior partner in the whole enterprise. The Collaborative Translation Project, then, represents an expansion of traditional academic activities, but it is certainly not yet a fundamental revisioning of how historians do their work.

Besides the sources of the project’s authority, much of its material also comes from more conventional channels. Most articles that have been translated so far have come at the prompting of requests made on the Internet mailing list H-France, which is a private, highly-selective list open only to graduate students and faculty; that same list is a frequent discussion site for articles in progress and those that have already been translated. Indeed, it is hard to imagine the project ever succeeding without the H-France mailing list to supplement it.

While individuals not associated with the list may certainly submit translations, my impression is that they seldom do. To my knowledge, the production of new translations has not been assigned in history classes; for most undergraduates, the work is beyond their abilities. French language classes would probably be a better venue for the assignment. I understand that it has already been offered on several occasions, and I look forward to this type of practical and often quite fascinating assignment becoming more common in the future.

Still, the extant translations were for the most part done by specialists in their own subfields of history. For example, I have translated several articles from the area of religious history, where I could be expected to give a more competent translation than most. On the other hand, I have left the mathematics entirely alone, because I must confess that I simply don’t follow it. Thus, one weakness of the Collaborative Translation Project to date is that many of the translated articles reflect the interests and areas of relative expertise of those who have done the translations.

This bias shows in the selection of articles so far, which tends to favor articles on economics, anthropology, religion, languages, and other disciplines in the humanities and social sciences. The sections on engineering and mathematics, both extensive in the original document, have not been so well-represented in our efforts to date. One way of correcting this bias would be to enlist bilingual people with competence in these fields to do some translations as well.

One last way that the project might be improved is ambitious, but I thought I would offer it anyway, simply to generate discussion and to perhaps learn a bit more about the technical limitations of the project itself. Just as the Collaborative Translation Project does not exist in a vacuum, so too one text does not make history, not even a tremendously important text like the Encyclopédie. In this light, I would like to propose some changes.

Even before the Encyclopédie had fully appeared, contemporaries began critiquing, correcting, and in some cases flatly condemning its content. Voltaire devoted much of the last period of his life to a series of “Questions about the Encyclopédie,” often writing point-by-point refutations of others’ articles. And the debate has continued ever since.

Indeed, one of the best definitions of history that I have ever heard is that history is an argument about the past. For the Collaborative Translation Project to really do the work of history, besides merely being an excellent resource, it would have to have a much greater measure of user feedback than it currently possesses. Besides the translated article, could contributors perhaps be invited to submit commentaries on their translations, lists for further reading, notes on difficult words or phrases, and relevant historical responses? Could we perhaps implement a discussion forum under each article, much as weblogs now have?

I do not know whether these steps are technically feasible or even whether the principal architects of the project would approve of them. From my perspective, though, I could see how all of these steps could make the Encyclopédie Collaborative Translation Project a more useful teaching tool. In a larger sense, they might also bring more of the work of history to be done online, and we can only hope it might day replace the cranks.

So to start the discussion, the biggest question I have might actually be too general: I’d like to know how much of what I have just said was new or surprising to you, how much of it was familiar, and how the technology of both this project and more generally the digital approach to history can be made better from our interactions. While historians are adept at providing content, very often we don’t understand very well the available systems of collection and dissemination on which our careers may increasingly depend.