This semester (Fall 1995) I am teaching a graduate introduction to medieval Catalan language and literature. Catalan is the language spoken in NE Spain (including Barcelona and Valencia) and the Balearic Islands. The medieval literature dates from the 13th to the 15th c. and includes some of the greatest writers in medieval Europe. The poet Ausias March is the best poet of the 15th century, anywhere, and the chivalric romance, Tirant lo Blanch (Tirant the White), was translated into English some years ago and became a supermarket bestseller.
The course has four students at Berkeley, two at UC Santa Barbara, and one at UC Irvine (the first time a multi-point distant learning course has originated from the Berkeley campus). While medieval Catalan might appear rather esoteric and the class relatively small, in fact the course was set up as a pilot to solve a serious problem in this era of budget constraints: How to offer lesser-taught foreign languages, which generally have very low enrollments on a given campus and require fairly specialized and therefore expensive instructors? Such courses are needed in order to prepare students, both undergraduate and graduate, in many fields, particularly humanities and social sciences (e.g., for field work in foreign countries). One solution is to use distance learning in order to pool the students from the various campuses, thus decreasing the per-pupil cost and allowing specialized resources to be shared.
The class meets for three hours once a week, using the distance learning classrooms on the various campuses. In addition to class meetings, there are: (1) an e-mail reflector group for the class. All messages sent to it are automatically re-broadcast to all students in the class; (2) a WWW home page (URL: http://www.lib.berkeley.edu/~catalan) which contains the sorts of materials usually provided in a paper course reader: Purpose of the course, methodology, grading standards, syllabus, reserve list, bibliography, weekly reading assignments, and background materials. The major innovation is that the background materials contain hypertext links to other relevant home pages in the U.S. as well as in Spain, thus offering access to a wealth of material that students would never see otherwise.
I shall sketch out briefly some of the problems we ran into, organizational, technical, and financial. With regard to organization, the first problem encountered remains the most difficult to resolve: the discrepancy between Berkeley's semester system and the other campuses' quarter system. Since Berkeley starts fall semester a full month before the other campuses, students were forced either to forego part of their summer plans or watch the first three classes on videotape. In addition, administrative arrangements for the granting of course credit are still clumsy, and reservation of the distance classrooms had to occur outside of the normal classroom reservation process. The latter problems will presumably be corrected as distance education becomes more closely integrated into the normal running of the institution; but the quarter/semester problem must be resolved. Otherwise distance learning cannot offer a viable alternative to campus-based instruction in the UC system.
Technical difficulties ranged from the fundamental to the trivial; but even the trivial caused a great deal of user aggravation. Early on we dropped the idea of true electronic reserves (even page images) primarily for technical reasons. My relatively modest reserve list of 40 items was far too large to consider asking the library to make available digitally, quite apart from solving the copyright problems.
There were a number of infrastructure problems: At one point, for example, I was using four different PC's, none of which was completely configured with all of the necessary hardware, software, and networking capabilities. We have also not been able to find suitable laboratory sites running UNIX X-Windows at Irvine and Santa Barbara. This capability is needed in order for the students to access the Dynatext data base of digitized leaves of medieval Catalan manuscripts held at Berkeley's Bancroft Library, a total of almost 1000 images. While enormously promising, the Dynatext version currently available is not yet adequate for instruction. For example, it allows zooming, which is necessary to see the fine detail on a manuscript, but only at the cost of allowing the user to modify the original image file. Read-only zooming as well as image processing capabilities are essential.
The network connections still require a great deal of nursing. At one point we lost the connection with UC Santa Barbara for two weeks and again had to resort to videotape sent by regular mail.
Current WWW browsers like Netscape are inadequate for serious work in many disciplines, since HTML does not yet support the character sets needed for phonetic or scientific notation except as digitized and therefore unsearchable page images.
The distance learning classroom at Berkeley is a state-of-the-art facility, while those at Santa Barbara and Irvine appear not to have reached the same level of sophistication. What one can do in the class depends on the facilities at all sites: the least common denominator phenomenon. Thus students at the distance classroom must be able to see both the instructor and the materials presented--computer screen or electronic overhead projector (Elmo), simultaneously--, just as in a normal classroom, instead of having to switch back and forth between the two. The bandwidth for the video representation is adequate, but the "back channel" line can only handle static graphic images, so that it is impossible to scroll down a computer screen, for example; although once scrolling stops a screen dump can be displayed. Use of the overhead projector for previously prepared materials works very well; but any kind of ad hoc elaboration still requires a blackboard, preferably electronic.
Leaving aside the preparation of the Dynatext data base of medieval manuscripts (funded by a $500 grant from Berkeley's Gaspar de Portola Catalonian Studies Center), the cost of library staff time in developing the digitized instructional materials was substantial; yet without such staff help it would have been impossible to do the class. For starters, design and implementation of the WWW home page took about 40 hours of staff time. Over and above that one library staff member has been devoting between 10 and 15 hours a week to preparing class materials, while two others have put in a total of about 25 hours proof-reading HTML-marked-up documents.
Thus, not counting my own time (between five and six hours a week to convert handwritten notes into a form suitable for digital presentation) and that of the students (so far some 20-25 hours for transcription and proof-reading), library staff have put in, very roughly, some 150 hours of staff time into this one course.
Conclusions This class, taught to students at two other campuses, would have been impossible without electronic surrogates for traditional, paper-based class handouts or course readers. Distance education without a digital library is not possible. However, no library, computer center, or instructional technology program can afford to provide this level of hand-holding for every class. Instead, libraries must begin, in collaboration with other partners, to make available the tools necessary so that faculty members and students can create these resources themselves. THIS TECHNOLOGY WILL ONLY AFFECT THE MARGINS--THE EARLY INNOVATORS--UNTIL AN INSTRUCTOR CAN SIT DOWN THE EVENING BEFORE A CLASS AND PUT TOGETHER ALL THE MATERIALS NEEDED FOR THAT CLASS. Right now this is only a pipe dream.
Commercial software is hopelessly inadequate. If airplanes crashed with the frequency of Windows, none of us would ever fly. Unfortunately, we must depend on the commercial marketplace to remove some of the technical barriers by building on expertise already possessed by many faculty members (e.g., automated conversion of word processing files to HTML). It is devoutly to be hoped that HTML will converge rapidly with SGML, which has already solved virtually all of the character representation problems mentioned above, as well as many others; or at the very least make use of UNICODE.
While machine-readable text, because of its search capabilities, is the ideal for instructional materials, at least in the near term digital page images provide a reasonable alternative, especially for any discipline which needs special character sets, images, graphs, tables, etc. However, to make this feasible, faculty members must have ready access to high-quality scanners and software as well as the capability of loading the resulting images into WWW home pages. Right now most network administrators are reluctant to grant the required access to servers.
Despite these problems I have been enormously impressed with the potential of distance learning and I regard the current project as a success. However, it seems clear to this survivor of the digital wars that libraries that want to be involved in distance learning have but two choices, one of which is no choice at all: They can attempt to continue with their traditional role as intermediaries between the end user and the information, although now reversing the direction of the exchange; they will now take information from one end user (the faculty member) and put it on the WWW in order to make it available to another (the student). To do this as we have tried to do in this project will require prohibitive amounts of staff time.
The alternative is to provide the software, hardware, and training necessary so that the instructor can develop his or her own course materials without investing "an arm and a leg" in the process. A digital "priesthood of all believers" must replace the librarian's hieratic functions in a radical recasting of traditional librarianship.
The Bancroft Library
UC Berkeley, CA 94720-6000
FAX (510) 642-7589
[ Return to Top of Page] [ Return to Conference Program page ]