What kinds of material will scholars create in the future?
By David RobinsonI’m writing in from the Chicago Colloquium on Digital Humanities and Computer Science, taking place this weekend near the campus of Northwestern University. Most of the papers are from interdisciplinary teams in which humanists and technologists are working together to study traditional humanistic areas. This morning’s schedule had three presentations, and comparisons among the three are an interesting lens on the state of scholarly publishing.
The teams are building and using new computer tools specific to their scholarly interests. The first presentation was about the Orlando Project, which uses a computerized system to track the interactions among women writers. It is, basically, a very thorough timeline. Thanks to a large-scale human cataloging effort, the information is equally readable to computers and to people. Computer searches using this tool can point the way to previously unnoticed information about who wrote, read, and communicated with whom, in what order and at what time—the bread and butter of intellectual history.
A second paper concerns an effort to encode the alchemy-related manuscripts and notes left behind by Isaac Newton. The project, The Chymistry of Isaac Newton, has (among other things) created unicode characters to encode the alchemical symbols that appear throughout Newton’s notes, enabling scholars to conduct computer searches for the appearance of these terms.
The Orlando Project, which was initially supported by research grants from the government of Canada, is now being published (and sold) by Cambridge University Press. The Chymistry project is free and open, supported by a grant from the National Endowment for the Humanities (in both cases, the projects have also received support from other sources). I’m afraid the comparison gives further ammunition to the critics of univerwity presses: The Cambridge project is very slow to load, and has an institutional sales model. Prices, it says, depend on library size.
The third presentation concerned several pending efforts to get lots of users to help encode corpuses of data that are of particular interest to humanists. These systems (one for novels and the other for the letters of Bertrand Russell) have users assisted by computer software look at images, and encode both the text and the metadata about, for example, the people and events to which the texts make reference. The model is that students, retirees and others volunteer to do the work a little bit at a time. “Wikinomics” for the humanities, the team says.
So far, in our own online symposium, we have focused a great deal on access, pricing, and quality control. As far as what it is that scholars are creating and distributing, however, most of our conversation applies comfortably to written essays, journal articles, preprints or books—the kind of thing that scholars all across the academy have been producing for centuries.
The advent of these new databases, many of which are known principally to the small groups of specialists to whom they cater, opens up a new group of questions about the future of scholarly publishing. One striking thing about these projects is the extent to which they are independent of one another: software is often written from the ground up, or heavily modified, to create a database that will be about a particular narrowly focused corpus of text. There are some standard tools, but there is also ample room for further consolidation. There may be an unmet need for specialized institutions that help data “go digital.”
Publishers, in the pre-digital model, were providers of practical expertise about the machanics of printing and distribution. For databases (and, indeed, any published data with which a user can interact) it would seem natural that an analogous institution could usefully exist. But university presses, as they now are, don’t have that expertise. Can they get it?
You might think that these questions apply only to a small subset of scholarly publishing projects: the largest ones, involving authoritative publication of large bodies of primary source material. Such large efforts have thus far been the main place where these questions emerge, but that in part reflects the lack of standard tools that might let scholars build interactivity into smaller projects.
For example, what about a social science paper that contains a quantitative model of a social phenomenon? There’s no reason, in principle, why such models have to be presented statically. The authors, in addition to explaining the assumptions on which they settled, could allow readers (viewers?) to move sliders that modify the assumptions and play out the results on whatever graphs or charts are used in the paper.
Even the basic academic act of citation, which is present in nearly every academic paper, might be dramatically improved (and, in some cases, already has been) with basic hypertext. This format can avoid crowding the screen with references, while still making it effortless for anyone to learn what has been cited, or to proceed directly to the cited material. But a great deal of the electronically published material today, presented as printer-ready PDF files, needlessly omits even this most basic way in which technology can improve the practice of scholarship.
I still do wonder how best to distribute static text. But that’s only one part of the story.
October 22nd, 2007 at 9:25 am
Thanks, David, for the helpful report. The point you raise about the technological role of presses in the digital age is crucial (and it is, after all, part of the impetus for Laura Brown’s report). Our experience with producing digital humanities books in the Gutenberg-e project was both that few scholars had imagined New Style digital books (they were producing digital text), but that those who were New Style thinkers seldom had access to the human resources necessary for the implementation of their ideas — they did not have digital humanities support at their universities (unless they were at U. Va. or a few other places) and the presses did not have staff competent to produce the new material. And of course everybody was short of the software needed to express New Style ideas. That situation has not changed very much. There are still very few campus-based digital humanities technical centers, and the presses are by and large not up to more than the production of digital text. The theory is there, however — take a look at Willard McCarty’s 2005 book, HUMANITIES COMPUTING, the most comprehensive and ambitious attempt to theorize the field.
October 22nd, 2007 at 10:33 am
The people behind SSRN, a network for sharing preprints in the social sciences, have just launched a parallel effort in the humanities:
http://chronicle.com/temp/email2.php?id=wNnczTfwGHbbJpgq9Ydj2nKqZZjmnjjh
October 25th, 2007 at 10:40 am
What Stan says is, alas, true. It is not as though presses aren’t trying. At Penn State we have an Office of Digital Scholarly Publishing operated jointly by the press and library, and one of our projects (jointly with Cornell under a Mellon grant) is to develop the DPubs open-source publishing platform for use in publishing a wide range of materials, which could include databases as well as conference proceedings, and regular books and journals. It is constructed in modular fashion so as to be adaptable by other universities to their particular needs. As I have said elsewhere, we’d all like to be doing more cutting-edge publishing like the Gutenberg-e and ACLS Humanities E-Book projects, but these are very expensive, and no one so far has figured out how to make them sustainable, even within a licensing scheme let alone as open access. My question about the Chymistry Project would be precisely: what happens when the NEH money runs out? It takes a major effort, like the very worthy Stanford Encyclopedia of Philosophy, to raise funds sufficient to create an endowment that can assure its long-term sustainability. It is highly unlikely that all publishing in the humanities can be sustained on the basis of endowments.