"The University of Virginia's Electronic Text Center: An Interview with David Seaman."

Virginia Librarian, April-June 1993 (pp. 6-10).

[Page 6]

Alerted to the development of an Electronic Text Center by retired University of Virginia Library director Ray Frantz in his recent interview for Virginia Librarian (Fall, 1992), we asked Electronic Text Center director David Seaman to describe his operation. That interview, conducted via electronic mail, follows.

Dan Ream: How can you best describe your Electronic Text Center to a librarian who may be unfamiliar with the technology involved?

David Seaman: The Electronic Text Center collects and prepares texts for inclusion into our online text service, it makes available hardware and software that permits the computerized analysis of text, and it provides guidance and training for these new scholarly tools. By so doing, we hope to make networked electronic books a mainstream part of the teaching and research resources on which University of Virginia faculty and students draw.

The on-line collection of texts is available 24 hours a day to any patron with access to a computer and a modem. The initial set of on-line texts includes the new Oxford English Dictionary; the entire corpus of Old English writings; selected Library of America titles; several versions of Shakespeare's complete works; hundreds of other literary, social, historical, philosophical, and political materials (chiefly from the Oxford and the Cambridge Text Archives); and the currently released parts of two massive databases from Chadwyck-Healey: J-P. Migne's Patrologia Latina, and the English Poetry Full-Text Database, comprised of the complete works of 1,350 British poets from AD 600 to 1900.

The Center is located in Alderman Library, and is open most of the hours the Library operates. It contains a variety of different computers with large color monitors, and software that allows users to search and analyze texts, to view digital images, and to record digital sound. The large monitors are essential because they provide sufficient space to work on several things at once; a user may well want to be able to search multiple databases, or use a database alongside a word processor, the library's on-line card catalogue, and an electronic mail account. We also have laser printers, CD-ROM drives, and scanners which can turn printed text into computer-readable forms and can generate high-resolution color images. Software packages in the Center allow scholars to build indices, concordances, word-lists, collations, statistical analyses, and hypertexts.

DR: What does an E-Text Center offer that traditional libraries cannot; or, why would I want this capability in my library?

DS: Libraries, traditionally, have been quick to embrace technology that allows better control of and access to the information they contain. In recent years we have seen the arrival of computerized catalogues, bibliographic databases, and now on-line full-text collections, all of which are rapidly becoming a mainstream part of a traditional university library.

[Page 7]

Having the majority of our electronic texts available on-line affords significant advantages:

On-line texts are freed from the temporal and spacial confines of the library. A user with a computer and a modem does not have to go to the Electronic Text Center to work with most of the texts that it holds, because he or she can access those materials from home, office, or dorm room. Moreover, the electronic text can support many simultaneous users.
In addition to more convenient access to our users, an electronic text collection permits more flexible use. Hypotheses can be tested over massive amounts of data with great speed: a UVa student can use the on-line texts to count and read in context the occurrences of blood images in Macbeth, for example, or search for the earliest recorded use of a word in the English language, or trace an idea or image through the works of Tertullian or Anselm.
On-line access to the texts also allows us to provide the same search and display "front-end" for all our collections. Having been taught to use one database, a user has the knowledge necessary to search all current and future databases, thereby overcoming the frustrations often involved with using CD-ROM products, each of which may have a different interface.

DR: Whose "vision" was your E- Text Center and how did your center end up in the library, rather than in the campus computer center or elsewhere? Was this a difficult negotiation process?

DS: We are funded from the Library's budget, with expertise and some equipment provided by our Academic Computing Department, and some start-up funds provided by the Dean of Arts and Sciences. We are lucky at UVa to have close and productive ties between the Library and Academic Computing; the University's move into large scale humanities computing is fulfilling a desire shared by both Library and information technology faculty.

The founding vision for the E-text Center was that of our Associate Librarian, Kendon Stubbs, who realized eighteen months or more ago that the time was ripe for an enterprising university library to move into on-line electronic texts and related services in a serious way. By that point the equipment needed for such an undertaking was getting cheaper, large commercial databases were being announced, and a campus-wide computer network was already being installed at UVa. Working with James Campbell, North Europe Bibliographer and head of our Electronic Information Committee, Kendon Stubbs set about the process of arranging space, staff, and equipment for a new type of Library service. The addition of John Price-Wilkin to the library faculty as Information Manage ment Coordinator has strengthened the program greatly.

The Electronic Text Center is in the Library, then, because the initiative and the continuing support came from the Library. The heavy and increasing use we have seen for the Center and online texts bears out the Library Administration's judgment that there was a pent-up need for such services if they could be offered in a supportive and easily used manner. And in a fundamental way, I think, an electronic text initiative belongs in a library because it is a textual as much as a technical endeavour although the two are sometimes difficult to separate. Certainly, the textual, bibliographic, and educational skills needed to evaluate, prepare, and present electronic texts to users unfamiliar with such services are all found in libraries.

DR: What are the goals or missions of your center?

DS: A principal aim of the Electronic Text Center is to help create a new user community and to facilitate a new level of computing within the humanities at Virginia. To this end we work daily with users to introduce them to new working methods, new teaching possibilities, and new types of equipment. The Library was adamant from the beginning that these new services had to be introduced and taught through ongoing workshops and demonstrations in order to create a user base which clearly understands that the Center's equipment empowers them not only to manipulate e-texts in a variety of ways, but also to create them if they are not currently available.

We aim to maintain a high degree of bibliographical awareness in the handling of electronic

[Page 8]

texts, ensuring that the texts in the on-line collection-and any changes made to them-are properly described. In some cases we will not know what edition has been used to create the electronic text, and we make this fact clear in the description of the text.

We are also concerned to maintain our on-line data in a standard tagged format-known as SGML, or Standard Generalized Mark-up Language-that will ensure that the electronic texts, with all their typographic, spacial, and structural instructions, will outlive the software we currently use to search and display them.

We hope too that the Center and on-line archive will provide a potential model for other institutions as they create an "information technology community" at the college level by unifying the creative energy and expertise of technical and non-technical departments .

DR: Please discuss the benefits to your faculty and students from this new Center and its status as a "bonus" for them in times of poor salary increases and tuition increases.

DS: As academia goes through a period of shrinking resources and growing workloads, the research and pedagogical services that electronic texts offer are likely to be doubly welcome. We are fortunate that a well-implemented on-line library catalog--Virgo-- has been in place for a number of years at Virginia, and has prepared users in all disciplines to use and expect on-line services from the Library. I have no doubt that the training efforts of the Virgo staff have made the implementation of our new e-text services much easier.

Our first semester of operation was necessarily a time of experiment and fine-tuning; nonetheless, there were significant research and teaching projects using the services of the Center and the on-line texts throughout the Fall. These included:

A Shakespeare survey course created a teaching tool using text, images, and digitized sound from different productions of The Merchant of Venice, to run alongside the on-line collections of Shakespeare's works.
An undergraduate survey course used our holdings in the 19th century novel and added Frances Brooke's Lady Julia Mandeville, an 18th century Canadian novel, to the collection.
Specialized software in the Center has been used to provide frequency lists and selective KWIC (Key Word In Context) concordances for a variety of texts, including Pride and Prejudice, Julian of Norwich, and the three texts of Piers Plowman. Preliminary results from the latter project have already been presented at a medieval conference.
Scholars from Religious Studies have searched the Hebrew bible, the Talmud, and hundreds of books of rabbinical responsa on the Global Jewish Database CD-ROM.
An English composition class used our services to gather Bush/ Clinton position papers from the Presidential campaigns.
A French professor has set up a language tutorial program for his medieval French course; the resulting activity has brought numerous undergraduates into the Center who may not have discovered us so soon otherwise.
Graduate bibliography students have used collating software,

[Page 9]

image scanning, and digitized sound while preparing and presenting editing projects.
A medievalist has digitized a manuscript facsimile, using the ability to enlarge portions of the scanned images as he transcribes them.
A UVa-based journal, Callaloo, has used our scanners to enter submissions for copy-editing
The Old English Corpus has been searched for examples of personified ship names, and the results used as the basis for a graduate paper.

In addition, we have worked with the first two Fellows of the lnstitute for Advanced Technology in the Humanities, an IBM- supported research center housed in Alderman library.

Precise usage of the on-line texts is difficult to gauge, because we do not necessarily see or hear from users. I imagine that some of those patrons who took training sessions have gone on to use texts from their offices and homes, although we do not have any record of their projects.

DR: How are you promoting your center to faculty and integrating it into the curriculum at UVa? Please describe the training you offer.

DS: We knew from the beginning that it was not enough to announce our services and then sit back and wait, and that we need to bring users to the Center and to demonstrate its services firsthand. New users need to see for themselves that they can sit at a large color monitor and simultaneously search multiple on-line databases (say, the Oxford English Dictionary and the English Poetry Full-Text Database) while manipulating color images of manuscript pages (which they may have just created in the Center), and then can open another window to e-mail a colleague about the results, or to log into another library's catalog before using our on-line document delivery service to order a book through Interlibrary Loan. Such a hands-on demonstration typically overcomes any initial trepidation a new user may feel.

At the beginning of the fall semester we sent an announcement to all faculty, and followed up with articles in local publications- the University Journal, Inside UVa, and the Virginia/ William & Mary Football Program. We have had some limited success so far in making our situation better known outside Charlottesville, principally through Internet postings and a description of our services in the Chronicle of Higher Education. In addition, presentations have been given at an Association of Research Libraries Conference in December, and at Virginia Commonwealth University.

We give many impromptu and formal presentations to faculty, students, and staff, which supplement the regular training sessions in the on-line search tools that the Library's Information Management Coordinator runs from an electronic classroom located in the Library's Reference Room. These classes include basic hands-on introductory sessions, and then more advanced tutorials usually based on a single database (the Patrologia Latina, the Old English Corpus, etc.).

DR: What can non-affiliates of UVa access from your center and how can they do it?

DS: Because of contractual obligations, access to the Electronic Text Service is restricted to UVa students, faculty and staff. However, the on-line University of Virginia Information Service can be accessed remotely.

DR: Please describe plans or hopes for future growth or expansion of your Center's holdings or services.

[Page 10]

DS: The e-text collection will continue to grow rapidly, I have no doubt, especially as more publishers offer electronic data for inclusion in campus-wide information systems. I hope this will lead to a diversification of our on-line text collection-the marketplace has dictated that our holdings are currently strongest in English and latin texts, with literary material predominating However, I am keen to stress to the University community that this endeavor is not by design Anglo-centric, and is not intended only to serve one or two key academic departments.

We will also see other electronic services developing at UVa, to work in consort with the e-text holdings: on-line art collections, a Multimedia Center, and increasing numbers of items on the Grounds Wide Information Service are all likely.

DR: Who staffs your center and what are their backgrounds? Are they librarians, computer scientists, other?

DS: The Center is staffed by one full-time faculty member (me), and five part-time graduate assistants, working 10-15 hours a week. We all come from the humanities departments here at UVa. I worked for several years as a graduate assistant in Special Collections, while I was a teaching assistant in the University's English Department. Of the rest of the staff, three are from the Ph.D. English program-Peter Byrnes, David Gants, and Kelly Tetterton; the other two, Peter Kastor and Jamie Spriggs, are graduates in the History and French departments respectively.

Most of the Center staff has had some bibliographical train ing prior to this job, which has been invaluable as we try to make sense of the texts that come in from the Oxford and Cambridge Text Archives and elsewhere. Rarely are these texts properly described, and we have had some success in locating source texts. We have all had to learn new computing methods and programs, but our previous use of computers in our own teaching and research provided a strong foundation for learning new systems. The staff members' strong research and pedagogical backgrounds gives us a good understanding of what a humanities researcher and teacher may want from such a system Our close contacts with our home departments also helps us to lure in new graduate student and faculty users, and can lessen the nervousness some users feel at being faced with an array of unfamiliar machines and software packages.

The staff's work in processing texts-checking, describing, and adding SGML tagging-is helped enormously by an innovative Library Staff Sharing initiative. Early last summer we asked for volunteers to check and tag texts, and the response was enthusiastic. Our volunteers have varying amounts of time to dedicate to this activity, of course, but their work in proofreading and preparing texts continues to be invaluable.

DR: How innovative is your center? How many others like yours are out there?

DS: There are many important e-text projects underway nationally, and a small number of institutions with some on-line materials or with library-based centers providing text analysis software and CD-ROM based texts-operations at Columbia, Georgetown, Indiana, Iowa, Princeton, Rutgers, Stanford, and Toronto all come to mind. However, to the best of my knowledge there is no other university or college currently that has the combination of a staffed E-text Center that is open extended hours, and a large on-line archive of electronic texts that are available to all of its faculty and students. This situation will change rapidly if the widespread interest we have evoked is any indication.

DR: Where are you located and how can you be contacted?

DS: The Electronic Text Center is located on the third floor of Alderman Library. We are open Monday-Thursday 9 am-10 pm, Friday 9 am-6 pm, and Sunday 1 pm-10 pm. For more information, or to arrange a visit, e-mail us at etextcenter@virginia.edu or phone 804-924-3230, or write to The Electronic Text Center, Alderman Library, University of Virginia, Charlottesville, Virginia 22903.

Sidebar: Standard Generalized Markup Language

The texts in our on-line collection are marked up with SGML tags that use letters and phrases within angled brackets to convey such information as structural divisions-title page, main body of text, scene, stanza, page, paragraph, etc. and typographical elements- changes in typeface, special characters, etc.

These tags preserve elements of the structure of a text, aid one's ability to constrain searches to particular features, and help one navigate and use a text. For example, one cannot limit a search to a particular chapter in a novel if there are no markers in the text identifying where chapters begin and end; one cannot view a word from a play in the context of a scene if the scenes are not tagged. SGML tags are simply other ASCII characters typed in as part of the text, and they travel with the text if it moves from computer system to computer system.