DATE 11/30/1998



SUPPORTING DEAN/CHAIR Gordon Braden, Chair, English Dept.


I. Project Title and Summary Description

Title: Bestsellers Database Project


ENTC 312 uses best-selling 20th-century American literature as a means of understanding 20th-century America. One bestseller from each decade is assigned as required reading, and beyond that, each student chooses a single best-seller (from a list of the top ten fiction bestsellers for each year from 1900 to 1994) as the focus of his or her own assignments. Once chosen, the book is no longer available to be chosen by others. The course assignments include a bibliographical description of a first edition, a publication history, a biographical sketch of the author, a reception history, and a critical analysis of the work in its cultural and literary contexts (or, if copyright permits and if the student wishes, a full-text electronic edition prepared to the standards of the Electronic Text Center). The course syllabus (for Spring ’98 and ’99) is available at and the assignments, submitted into a web-accessible database, are available at:

II. Pedagogical Aims of Project

"ENTC 312: 20th-Century American Bestsellers" is being taught for the second time during the spring semester of 1999. Enrollment is 75 students, and the course fills Department distribution requirements for the English major. With 940 bestsellers to choose from, and 75 students taking the course each semester, this course could be taught for many years.

The pedagogical benefit of the Web-based assignments is significant: because students are being asked to gather information that isn’t already compiled in one place, and in fact in some cases may or may not be available at all, they are driven to use a very wide array of research resources, ranging from traditional printed library reference works, to electronic databases, to (in some cases) personal interviews with publishers, special collections librarians, and even authors. In addition, because students have the sense that they are contributing to a permanent and published resource—a unique database of information on 20th-century American popular fiction—and because they can see one another’s assignments as they accrue, student motivation is very high.

This course has already proven its ability to engage undergraduate students in library research unprecedented in its depth and sophistication. It also has proven its ability to teach students the basic nature of real research—namely, that some questions have no answers, and that the more you know about a subject, the less certain you may be. What remains to be seen is whether the information compiled by students in this class actually constitutes a research resource for others, whether the current techniques for grading are effective in a course where two or three people evaluate original work by seventy-five researchers, and whether elements of this course can be extended to other contexts.

On that last question, I am fairly confident—the kind of information that is being collected in this course is quite generic, and could be collected in a number of different courses that deal with books as objects, with publishing, or even with library science. For example, I can easily imagine using assignments like this, compressed into a shorter time-span, to train incoming graduate students in library research skills; I can also imagine using a database like this (with some modifications) in bibliography courses of the sort taught at the graduate and undergraduate level in the English department. In fact, because this course has proven its ability to teach library research skills so well, I am planning to teach a version of it as an ENLT 200-M course, the course rubric for introductory seminars for prospective English Majors. I also think that the way ENTC 312 is organized—with a large number of students, individualized and web-based research projects for each student, involving the collection of text and image data, and the submission of that data through a very simple and straightforward web-based set of forms—is eminently generalizable. The combination of individual research, immediate publication, and intramural competition, is what makes this course unique, but these are pedagogical strategies that are in no way tied to the course content, the particular discipline in which this course is taught, or even the technical methods being used to mediate the electronic processes used in the course.

Less clear to me, and central to this proposal, is the question of the quality of the information gathered and submitted by the students in this course. This is an important issue to understand—not only in order to evaluate more systematically the pedagogical effectiveness of the generalizable techniques being used in ENTC 312, but also in order to establish the value, to other researchers, of the database being created by successive semesters of ENTC 312.

III. Preparation and Feasibility

Though the technical mechanisms which make the course possible are already in place and working, they were programmed by a novice (me), and therefore there is a good deal of room for improvement, not only in elegance, portability, and robustness, but also in the areas of data security, privacy, and general process automation. It may well be that the current mechanisms, which involve Perl and an MSQL database, need to be thrown out and a new set of tools used instead: certainly, MSQL imposes some significant limitations on searching and manipulation. Therefore, the first step on the technical side of this project would be to assess those mechanisms, and compare them to alternatives. Once the best method is clear, I would envision working with a student (or TTI) programmer, rather than doing the programming myself—not because I’m unwilling to learn, but because I’d like to avoid recreating the shortcomings of the current setup, shortcomings which reflect my own limitations as a programmer. Updates and enhancements to existing database structures are always onerous, and for that reason, as part of this proposal, I would like to involve some of my colleagues in an informal evaluation of the nature and extent of the fields I’ve established for collecting this data, probably in parallel with the reconsideration of the technical mechanisms, so that if a migration and a restructuring are required, they can be timed to coincide. Since the course is already being taught, and is already using the kind of mechanisms that are required, we have a ready test-bed of existing data that could be used in development.

IV. Evaluation and Assessment

One important measure success during the fellowship project will be a completed assessment of the quality of data produced by students—regardless of the outcome of that assessment. If the outcome of the fellowship year, on the technical side, is a more stable, more portable, more secure set of data-entry forms and database components, I will consider the project a success in technical terms. The pedagogical effectiveness of the project is, to my mind, already established, but further measures of that would be the end-of-semester feedback from reference librarians and others who support the students’ work, and the students’ own course evaluations.

V. Collegiality in Project Development

I am well acquainted with the TTI program as an observer, have attended several presentations by fellows in the past, and have worked at IATH with a number of faculty who have been through the TTI program. I look forward to the collegial interactions that TTI encourages, and I’m aware of the benefits of such interactions through my own experience at IATH.

VI. Dissemination

The English department can be made aware of work I might do under the TTI program in a number of ways. First, through its own web site, which features "web-related courses." Second, as chair of the department’s Web and Technologies committees, I have regular contact with a number of colleagues with a demonstrated interest in matters electronic. There are also occasional workshops in the department on teaching, as well as opportunities to give talks to colleagues and graduate students on a topic of one’s choosing—either of which would be a logical venue for reporting to the department on the TTI work. Beyond the department, there are a number of ways to disseminate this project: I have already used the humanities email discussion list, Humanist, to poll colleagues at other institutions for interest in, and comments on, this project; I would continue to use that (and other like) channels. I also travel frequently give faculty seminars on teaching and technology at other institutions (Lawrence University, Princeton University, William and Mary, University of Vermont are on this year’s list) and this work could be part of those presentations. Finally, I regularly participate in conferences, and will be one of the local hosts (along with the TTI and the Etext Center) for the 1999 conference of two international humanities computing organizations, at which conference I expect an opportunity to present this work, probably in a poster session.

VII. Brief Summary of Equipment and Support Needed

A desktop computer with an ISDN modem would be very useful, as much of the work I do on this project is done from home, and the laptop that is currently my only home computer is somewhat limited in both speed and display space. Depending on the assessment of database tools to be used for this project, I may require funding for some server-based, web-accessible database software. But most of the budget should go to graduate assistants (for data assessment) and to student (or TTI) programmers (for programming work on the course mechanisms). Graduate students in English or other humanities disciplines would have the skills necessary for the data assessment—the issue there is not training, but rather the number of hours it will take to replicate the research done by 75 to 150 individual students over the course of one or two semesters. Assuming an experienced programmer is available, no special training should be required in that position either.

I don’t anticipate a need for travel funding, for permissions fees, or for other services and supplies in connection with this project, nor do I anticipate requesting funds for release time or summer stipend, as I already teach only half-time, and the other half of my time I am on a twelve-month administrative appointment as director of the Institute for Advanced Technology in the Humanities.

VIII. Budget

Research Assistants (data): 2 @ 10 hrs./week @ $12/hour x 15 weeks = $3600

Programmers (database): 1 @ 20 hrs./week @ $14/hour x 20 weeks = $5600

Software (est.: may not be required): $3000

Hardware (desktop PC, 450 MHz, ISDN, etc.) $3500

ISDN service (one year, ESI UVa rate) $450

Total: $13,150 - $16,150