Gail McMillan

The professional literature for library as well as computer science is rich with jargon and a term currently in vogue is "digital library." Unfortunately, too many authors write as if a digital library is a library with only digital resources and some programmed "services" and a distinct and separate entity from a traditional library. However, libraries have been and will continue to provide their users with materials in a variety of media, including digital. For nearly a decade, libraries have not only provided links between users and digital resources, they have been the originating source of online information. Virginia Tech’s Scholarly Communications Project (SCP), a department within the University Libraries, has nearly ten years experience in both publishing electronic works. It also has designed online systems that enhance traditional library services by migrating them to the Internet.

Established in 1989, the fundamental principles of the SCP were to work with the developing Internet technology so that faculty could experiment with new forms of scholarly communications--forms beyond the scope of the academic journal in print, for example. Since then, the SCP has developed the library’s electronic reserve system, built and maintained the infrastructure for electronic theses and dissertations, conceptualized and developed a digital image storage and access system, and worked with regional and international agencies to put timely news reports online, in addition to publishing electronic journals.

These and other digital initiatives combine the traditional library focus on user-friendly services, with a keen awareness and farsightedness that has resulted in building blocks to a content-rich future, even when predicting that future in clear detail is not possible. Library services such as those provided by the Scholarly Communications Project support innovative teaching and learning environments just as they continue to support traditional classrooms and individual research and scholarship. This is a digital library within a traditional academic lbirary, a combination that serves the university far better than either can independently.

To enable experiments in new forms of scholarship and to support and improve library services for the whole university, University Libraries allocates staff and other resources to the Scholarly Communications Project. Working within the means of 1.5 FTE and some student assistants, the SCP philosophy is not to turn away anyone who needs library support as it strives to fulfill its three-part mission: to experiment with new formats and technologies, to implement access to networked information, and to provide excellent library services.

As with a digital library, it is not enough that the SCP accumulate digital resources and design access and storage (including preservation and archival) systems; there is the responsibility to help users discover and use the resources. In addition to assisting editors, authors, and publishers in putting their works online, SCP increasingly supports user access to these works through one-on-one contact, workshop applications, and classroom presentations. A compelling need to provide user-friendly services and access is where the traditional library with digital resources differs dramatically from the digital library that may be content rich but service poor because only so much can be programmed and then the unpredictable human element requires human-to-human interaction. SCP combines rich digital content with automated access and storage systems, and user-friendly library services.

From the beginning, the staff associated with the SCP and the faculty at the university who requested their support have provided the unit's direction. They have collaborated to determine what experiments to conduct, resolved questions arising from the user community, addressed issues raised by the technology, and determined how to meet the needs of its users and clients. SCP has expanded considerably since developing EReserve and ETDs from behind-the-scenes and strong technical support unit, to a prominent role as liaison to students, staff, and faculty.

One of the initial activities of the SCP in 1989 was to determine how to publish an electronic journal. By the time the first issue its ejournal was ready in the fall of 1990, a desktop computer was rigged to also be a Gopher serverand the Journal of the International Academy of Hospitality Research was available as an ASCII (text-only) file.. In the ensuing years the SCP migrated its first four ejournals to the Web and added another 16 titles to its publication roster. Seventeen are the Web equivalent of printed academic journals that have been enhanced by hot links and 24-hour access. (An added benefit of the Web is the capability to gather statistics about accesses that is not possible to gather about academic journal circulations, at least at Virginia Tech.) 15 titles are currently available with two more under development. Nine are electronic-only journals. Eight of the 15 current titles are edited by Virginia Tech faculty and two by faculty at other universities. All are associated with scholarly societies or professional associations; and four are mirrored for MIT Press.

table 1

A goal of most new ejournal editors is to educate their subscribers. By publishing ejournals, their subscribers have an incentive to use the Internet and become familiar with navigating the Web. The editors’ goals are similar to the library’s, both want to provide not only access to their journals, but also to help their readers gain the knowledge to conduct further online research independently. The success rate for editors and SCP is very high.

As with online journals subscribers, there has been a continuing increase in the level of technological expertise among our faculty editors. At SCP we make it easy for faculty editors to put their journals online by providing them with very loose guidelines. They no longer want to know which word processor they should use, they want to know: will we accept PDF files in addition to HTML; how to imbed graphics in the text of the articles; and, what about making sure the links to URLs are working? Electronic journal editors have a much better understanding of the capabilities of the technology. While articles in HTML are each separate files, editors prefer that one PDF file contain an entire issue of the journal. They continue, however, to rely on the library to provide the access and archives (including security and appropriate back-up), as well as the user support functions, especially helping their students and new readers access online resources.

FACULTY AND STUDENTS BUILD THE DIGITAL LIBRARY

SCP receives almost all of its resources in digital formats and this is one reason the few staff at SCP can accomplish so much. When ejournal editors do not send SCP new issues already formatted in HTML or PDF, student employees convert them and link citations to references. Providing hard copy to digital format conversion on a production scale, however, is not in the library’s past and it is not a current or forthcmoing service. When SCP designed the library’s electronic reserve system, we gave faculty various options for making class materials available online, but we did not offer, and very few asked us, to convert print materials to digital formats. There are two reasons we can do this: (1) digitizing equipment and personal assistance is available in the library, and (2) Virginia Tech faculty receive with the equipment, software, and training to create their works for the online environment.

The New Media Center [URL] is a sophisticated and well staffed computer lab located near the Reference Desk in the main Virginia Tech library. Open to the public, NMC has 21 G3/233 MHz Power Macintoshes with PC compatibility and 18 Apple Color One Scanners, as well as select specialized equipment in an adjacent develop lab. Four full-time staff and ___ FTE student assistants are well-trained and have experience with software applications in graphics, desktop publishing, word processing, Web development, 3-D, animation, CD-ROM development, digital video, digital audio and other multimedia areas. Consulting services are available in the NMC, by phone, or at the faculty's office. The NMC faculty liaison also provides guidelines for long-term and major project support.

NMC is also heavily used by the university’s Educational Technology unit to conduct intensive faculty training sessions, especially in the summers. [URL] The Faculty Development Institute (FDI) is part of a university wide instructional strategy to invest in faculty and students by guiding faculty exploration of educational technology. Many FDI participants began the intensive four-day FDI workshops with little or no experience using computers. Others were "early adopters" who had been previously introduced to multimedia and were eager to discover new applications. Virginia Tech faculty who have participated in the FDI incorporate technological innovations into their instruction. The SCP director has covered EREserve, ETDs,a nd copyright issues for hundreds of faculty during FDIs.

Because Virginia Tech faculty receive training in combination with equipment and software they are fully prepared to create online works for electronic publication or online classes. The library is largely relieved of the responsibility for converting its resources to digital formats. The library can, therefore, devote its resources to delivering the digital content and to mapping and enhancing traditional services to online, user-friendly services. In addition, Virginia Tech’s graduate students have been required to submit their masters’ theses and doctoral dissertations in electronic formats, usually PDF, since January 1997.

For decades libraries have stored, occasionally circulated, and strived to preserve one copy of the final product of each graduate student's research, a thesis or dissertation. In 1994 the Graduate School invited the library to prepare for the inevitable arrival of theses and dissertations in electronic formats. SCP accepted the challenge and developed the procedures and mechanisims for storage and timely and access that resulted in a working prototype. The Networked Digital Library of Theses and Dissertations (NDLTD) has over 1200 Virginia Tech ETDs (as of August 1998).

The ETD project is an example of how libraries (independently and in collaboration) to improve services and increase the wealth of information available. SCP initiated the ETD project by bringing to gethe all the lbirary staff involved in processing theses and dissertations in paper so that they could design a parallel workflow for ETDs. Then SCP programmed and enhanced the processin to take fuller advantage of the ditigal resources and the Internet access. For example, the students and their committee chairs automatically receive an email message containing the URLs when their ETDs are approved by the Graduate School. Students may select access options from (1) unrestricted access; (2) VT only acccess; or (3) no access. VT ETDs can be found by browsing or searching from the Web and through the online library catalog.

Virginia Tech is not alone in prediciting that ETDs will eventually replace their paper counterparts. The Virginia Tech initiative, began expanding in the fall of 1996 to include other universities in part through a grant from the US Department of Education’s Fund for the Improvement of Post-Secondary Education. Currently nearly 40 universities in the United States and abroad are also committed to participating in the Networked Digital Library of Theses and Dissertations.

ETDs serve many purposes, including:

--- To allow graduate students to be more creative in documenting their final research

--- To improve timely access to the information within theses and dissertations

--- To save students money producing their final research projects

--- To reduce the need for additional shelf space in university libraries and archives.

By archiving and providing access to both student faculty generated works, SCP had to address issues such as online archiving [URL], unrestricted access vs. limiting users’ access, and intellectual property considerations [URL]. But, also of importance to the libraries is increasing the size of its digital collection without having to increase the staff for processing and continued maintenance of the collection (e.g., circulation and reshelving). The NDLTD will also contribute to maximizing access and services by experimenting with simultaneously distributed (i.e., multi-library) and centralized (i.e., single library) access.

ELECTRONIC RESERVE

When Virginia Tech’s University Libraries began discussin gexpanding its Reserve Desk functions in the early nineties, SCP asked to develop such a system that would enable instructors to put their class materials online. Initially, SCP’s electronic reserve system mapped Reserve Desk practices to Web processes, i.e., faculty sent (FTP) the library files of their course materials to manage. One draw back was time delays while library staffdetermined that files had been sent, verified to which course and instructor they should be linked, and moved the files to the appropriate directory. Advantages to the system included typical library services: once the faculty had sent the files they were relieved of having to deal with them or the students who needed the information. Improved library services included 24 hour access and programmatic backups so that files would never be inaccessible.

EReserve quickly evolved so that faculty could manage their own online class materials on a library server or link from an external server to Eresrve. Initially, the electronic reserve system required them faculty to register their classes for access through EReserve. Evolution did not ceased. In fall 1997, the library no longer needed staff to move files to appropriate directories when the SCP programmer created CGI and Perl scripts that enable faculty to programmatically link and upload files. As of fall 1998 semester, EReserve is linked to the university’s timetable and course descriptions, relieving faculty of the need to complete a Web form with this information. Faculty enter the same PID and password that they use for the university’s email system to become immediately linked to the Registrar's current list of classes. From their list of classes, faculty add and remove files in real time, or enter a URL to link their classes to existing Web sites. Access by students is also immediate

SCP collaborates with other units within Information Systems, the library’s umbrella administrative unit that includes Administrative and Distributed Information Systems. While the purview of EReserve remains in the library through SCP, the computer storing the programs and scripts as well as faculty and class information and their files resides in the Computing Center and is linked by the campus network.

Full copyright [expanded ver. near p. 20] compliance (in accordance with our interpretation of *US Code Title 17, sect. 108) is overseen by the library and access to the electronic reserve system continues to be limited to our university community. Whenever access must be restricted, as with Ereserve and some ETDs, all registered students and university employees are recognized as valid users whether on campus or off, through name proxy servers or IP addresses. [expanded ver. near p.17]

In the spring of 1997, SCP became part of a university-wide initiative to provide students with improved and efficient access to Virginia Tech's online and electronically enhanced courses. Called, VTOnline, the Web site [URL] presents all of the university’s electronic learning opportunities and information services. VTOnline presents information concerning online degree programs, short courses, extension activities, and public service initiatives. It also links existing and new network centered teaching initiatives, ranging from course and program innovation to intellectual property policies, assessment, and evaluation practices.

Another opportunity to serve our current and potential international community came in March 1997 when a Washington, DC news agency contacted SCP. NewsExpress selected SCP as one of 15 libraries to participate in its experiment to provide timely accdess to foreign language newspapers in PDF format. This was a welcome opportunity to expand the existing Virginia News Online service.

Working with regional news publishers, SCP had already developed procedures and written scripts that programmatically linked daily news reports to calendars. Through collaboration with Landmark Communications (owner of the Virginian Pilot and the Roanoke Times), SCP began providing access to local and regional news reports for the Virginia Tech university community and the Blacksburg Electronic Village (BEV) in 1995 (94?).

Initially, SCP had to overcome problems with uploading reports from computer tapes for historical issues and frequently changing access codes on the publishers older computer systems. In 1994 our experiment expanded to included pre-dawn automated dial-in and downloading and programmed HTML mark-up. New challenges included quality and quantity control because daily news reports, though completely text, grew quickly into very large files. One result was that programmed indexing enabling word searching was relegated to run on the weekends when less network traffic would be effected by slower system response times.

Initially our goal was to provide same-day news reports, but the newspapers’ out-dated technology as well as the publisher’s fear of losing paid subscribers, combined to frustrate our attempts. SCP ceased news file downloads early in 1997 but continues to provide networked historical news reports about the region and the state. The wealth of knowledge and experience SCP gained by working with these newspapers was next applied to television news reports.

SCP offered its experience managing daily newspaper reports on the Web, effective security precautions, stable and constantly available information resources, and experience with indexing and search engines to the regional CBS-affliiate, WDBJ7. The Roanoke station manager agreed to electronically transfer to SCP the daily news reports. SCP’s programmer adapted the scripts and programmed procedures for receiving file transfers of news reports already in digital format from the closed-captioning prepared for hearing impaired viewers. Within about 24 hours of broadcast, local and regional news reports have been available at http://scholar.lib.vt.edu/newonline/WDBJ7 since March 1995 (date?).

Having accumulated the experience and some hardware through the growth of the electronic journal and news collections, SCP staff were confident that they could handle anything available in an electronic format, though it could not provide scanning services to convert from hard copy to digital formats. However, when offered accompanying news videos we could not resist digitizing a few minutes of video with audio so that while reading a few stories, readers could also see and hear a televised interview. This experiment more fully replicated the content of the news reporting, while lacking faithful (archival) reproduction.

Prepared to expand its online news reports, SCP welcomed the March 1997 call from a NewsExpress about participating in an experiment to supply international newspapers online. The first newspaper, Le Monde, is available online in PDF format before it is available in paper on the streets of Paris or New York! About a year after this experiment began and after similar success with two Arabic newspapers, El Attat and Al Nahar, NewsExpress was ready to terminate the free access and offered continued access to Le Monde online. With just a little research, Virginia Tech determined that Le Monde Online would be less expensive and give more access that our current subscriptions to the paper and microfilm versions.

There is one exception to the SCP policy of accepting only digital formats for access and archiving. SCP designed a procedure for converting from hard copy to digital formats and a system for access and archiving early in 1996 when SCP led a small, interdisciplinary working group in developing a tactical plan as part of the University's Information Systems' strategic plan. The prototype called for digitizing 300 art/art history slides and entering identifying information (metadata) into online forms and linking the images and their descriptions into a database or "ImageBase." These digital images would then be incorporated into an Impressionist/Post-Impressionist art history course being taught fall semester 1996 by the Art & Architecture branch librarian, Dr. Annette Burr.

The working group envisioned many advantages to networked digital images over the typical analog slide collection. Instructors could display any images in any order during a lecture and would no longer be tied to the order the slides were placed in the carousel. Students could view the images outside of class at any time--whether the slide library was open or not. With the success of this prototype, slide collections around the university could be digitized for use in classroom lectures.

Our plan showed how, in the course of normal class preparation, faculty could select the slides to be digitized, photographic experts on campus would create high quality digital reproductions, and the library would store and provide access to the images. We envisioned our process working just as well for the entomology as for the history professor. The images selected for the prototype were slides for a semester’s-worth of lectures that had identifying information written on them. However, our system would allow the faculty to enhance or elaborate on that information prior to student access. An early implementation of the Dublin Core meant that every image is associated with appropriate information identified by these near-standard elements. An advantage of being in a university setting is access to relatively inexpensive student labor, one result being timely data entry and links between images and data.

During the spring of 1996, we created the work flow and designed the process to build the VT ImageBase content. The SCP programmer also designed a very basic searchable index of creator, title, and subject. During the summer the instructor pulled her slides, and PhotoGraphic Services prepared high quality TIFF images. Student assistants from the Art & Architecture Library and SCP entered the identifying data and derived small GIF images to display with the data in a user-friendly Web page. They also derived an intermediate size display image so that Web viewers could get a good look at the image within a reasonable timeframe. Neither the GIF nor the JPG images print at a quality worth stealing; for quality reproductions, users must contact Special Collections.

The TIFFs are the "archival" images; these are not stored on the SCP server but are archived on a machine at the Computing Center. PGS sends the "original" TIFF images to a file on the Computer Center’s machine that is accessed either by the student assistants when they create the user-friendly display images, or by PGS staff when a hard copy reproduction is contracted and paid for through Special Collections. The quality of the copies produced by PGS varies depending on the needs of the client. Costs vary accordingly--enough to cover PGS costs and to digitize one more image.

Several challenges had to be addressed for the original plan to be fully implemented. In addition to resources to digitize, the classrooms lacked good projection and display equipment for art slides to be viewed with the clarity--sharpness, distinct lines, and vibrant colors--now available with 35mm slides. These challenges remain, but the fast campus network connections and the continually growing disk storage space did not keep the prototype from being designed and tested. As a result of this strategic lack of appropriate classrooms, we have not persued full faculty participation in digitizing large collections of slides. We felt that improving instructors’ lecture content and classroom environments were essential to getting their participation in a large conversion scheme. With improved study environments for students who could access the images used in each day’s lecture, would not be sufficient incentive to get good faculty participation.

One year later financial support from VIVA contributed immensely to hiring student assistants and digitizing local, regional, and Civil War history images housed in University Libraries' Special Collections. From 250 images online for fall semester 1996, the VT ImageBase grew to nearly 8,000 images by January 1998, retrievable by subject, title, and creator. As of August 1998, about 14,000 digital images are retrieval by 19 fields.

The library has a philosophy of open access., Virginia Tech is a state institution and, therefore, its library is open to any citizen of the Commonwealth. Accordingly, the Scholarly Communications Project has encouraged all of its collaborators and contributors to provide unrestricted access to their works. We have met with varying degrees of success.

Our ejournal editors have been willing to give unrestricted access a try with the caveat that they can change their minds. We, of course, agree, and are even looking forward to the day when we will need to design an efficient system that will allow selected readers access to specific works.

Otherwise, we have negotiated university-only access when access had to be restricted. This has been the case with online newspapers and commercial ejournals. Our early work with regional newspaper publishers was very enlightening when they required that we limit access. They could not be convinced that a typical customer continue subscribing to the local newspaper even if it was available online without charge. My family’s morning ritual is to have breakfast together at the same table while everyone looks at a different section of the newspaper. The newspaper publisher seemed to think that the four of us would huddle around our 15" monitor with our cups of coffee, glasses of milk, bowls of cereal, and plates of bagels with cream cheese and jelly--all content to read the same screen’s worth of news at the same rate! Not a likely scenario in my family or most anyone else’s.

There is one occasion when SCP willing and consciously limits access to our university community and this is when traditional library services involve access without permission to someone else’s intellectual property (i.e., frequently referred to as copyrighted materials). EReserve is an example as is the VT ImageBase. In both systems, faculty scan someone else’s work without permission so that their students can learn from those resources. Just as the library would have served copies from its Reserve Desk, we now provide restricted access through the Web to our university community.

However, at the Reserve Desk you do not have to be a student to check out materials. As with the Circulation Desk, the Media Center, or any of our branch libraries, you just had to show that you were a citizen of the Commonwealth (e.g., with driver’s license). Now, the digital materials available through electronic reserve are available only to members of the university community or to people on campus with access to the network (e.g., Internet workstations in the library or computer lab). Access to EReserve, for example, a is more restricted than the traditional library Reserve Desk service, so that the library will not violate the rights of copyright holders and we will comply with copyright law.* The notion that faculty and students should be validated to ensure that they should have equal access to library resources is a new concept to state supported institutions like Virginia Tech with a library that did not restrict access to just VT students.

Access is allowed through two validation systems. The most commonly known system is casually referred to as "IP address." That is, the server with the information you want validates that the computer you are using is registered with the university’s computer name server. Similarly, but through a somewhat more cumbersome process, university faculty, staff, and students who are away from campus computers, can use the "proxy" functions that are components of Web browsers such as Netscape and Explorer. This function has the user enter their personal identification (mine is gailmac, many are alphanumeric) and their password as it is registered with the university. One must also enter the Internet address of the host (often the mail) server. This system validates the user rather than the computer, but both require that the university maintain an up-to-date register of computer addresses and current university community members’ user Ids and passwords.

Many of the library’s online systems rely on these name servers to control access. In addition to electronic reserve, systems using online computer/user validation include requests for library materials entered online for interlibrary loan and document delivery. The library has also contracted with many vendors for Web access to a variety of indexes and full-text databases. These contracts almost always require that the library guarantee that only its students, faculty and staff have access and that others who may have access are legitimately coming through the campus computer network, just as they would have at one time walked through the doors of the library and taken the resource off the shelf.

Unlike commercial publishers, the library as publisher does not distribute its products and, therefore, never runs out of stock. The library also tends to archive indefinitely what it has accessioned. Therefore, when the SCP was initiating new works and aligning new procedures for online resources with past practices with hard copies, we seriously considered the thorny issues of archiving and security. Initially many of our authors, editors, and collaborators felt uncomfortable knowing that there is not always a paper back-up.

Having a work in paper provides many with a false sense of security. Historical experience in libraries has demonstrated that a work that is checked out, is not always returned. When this happens, there usually isn’t a back-up copy. Back-ups are precluded for several reasons, most of which are associate with costs. It is too expensive to acquire two of everything, one for patrons and one for the archives. When this is done, the added costs of processing, shelving, and storage cannot be ignored.

One of the most outstanding and beneficial attributes of digital resources, is that the library and the SCP are committed to security and archiving. We make frequent back-ups, storing them online and off-line, in the same room as the server and in other rooms and in other buildings. When a drive fails, there are others at hand and there are copies of the information resources that can restore any work that was lost. This is not the case with most library resources in non-digital formats. These copies are programmatically generated and do not require labor intensive processes to transfer them from one location to another. They are not bulky and do not require new buildings be constructed to house an ever increasing collection, as is the case with books and journals.

The Scholarly Communications Project's staff has always been small. Initially there were two half-time positions, with Library Automation sharing a programmer and Technical Services sharing a serials librarian. In 1994/95 a half-time technical assistant from Technical Services joined the SCP to focus on HTML tagging for electronic journals. In 1994 the SCP director officially separated from Technical Services to devote full-time to the Scholarly Communications Project. The technical assistant’s position was subsequently upgraded to full-time, in part because there was less need in Technical Services for data entry, but also because technical processing was de-emphasized to help fulfill an administrative goal to put more people at public service points for increased personal interaction with library users.

The gradual weaning of two positions from traditional technical services provides some evidence of the changes technology has instigated within the library organization. Further evidence is the 1996/97 upgrading of the then-vacant technical assistant’s position to that of a programmer so that manual tasks such as file downloads, some HTML mark-up, and forms could be completed and processed programmatically through in-house scripting. In 1998 the programmer was upgraded to a programmer/analyst.

While gradually increasing staffing for SCP, personnel resources are minimal considering the number of titles it publishes and the systems it designs and maintains. Therefore, SCP rarely solicites new electronic resources to publish. We know many more faculty participate in scholarly journal editorial boards, for example, but we are afraid of success--of attracting more work than we can handle well.

Originally a Gopher server, the NeXt 3.3 called Borg or scholar, began operating as a Web server in February 1993. During the first half of 1995, SCP expanded from one principal server to three servers, splitting the large and rapidly expanding image and newspaper files from the files of scholarly publications and electronic journals. Our October 1994 archive required 479Mb; up from 78Mb of storage we were using in January 1994. In June 1995 4Gb of scholarly electronic publications were available through the Project's servers.

In October 1997, scholar migrated from the NeXt to a Sun Netra Server. It has a 200 Mhz Ultrasparc processor and 128 Mb of RAM, with several gigabytes of disk space allocated for the operating system and related tools. It now runs the Netscape Enterprise Server on Solaris 2.6 and Perl 5.004_01. To backup our system, we have a Sun 8mm Ultra Wide SCSI tape drive that takes 170m tapes with a compressed storage capacity of 40Gb.

Libraries are evolving more than digital content and services; they are frequently called upon to pick up services that could be provided by other areas of the university. Likewise, an evolving role of the Scholalry Communications Project has been to develop expertise in areas tangential to online scholarship and then to coordinate and share the knowledge gained, as we did with copyright. Largely through our work with ETDs and EReserve, but also ejournals and digitizing images, we developed knowledge that became known throughout the university (especially since our university lacks a copyright expert).

The library has long been a guardian of copyright through thoughtful and law-abiding policies for interlibrary loans, reserve materials, photocopy services, and others. However, when these Web sites developed, we discovered inconsistencies in policies and lack historical understanding of the details of our policies. Therefore, SCP and selected User Services staff (including branch librarians, reference librarians, and Reserve Desk staff) evaluated policies, agreed upon consistent practices among services, and SCP developed a Web site (http://scholar.lib.vt.edu/copyright/) to share what had been learned with the rest of the library and the university community.

Faculty, especially those expanding into the realm of distance education, had also been asking about copyright laws so the SCP began sharing its knowledge at the summer 1998 faculty development institutes. Hour-long sessions began at the Web site, focussing on the rights of creators/authors, a thorough review of fair use, and public domain issues.

The Scholarly Communications Project is an example of how libraries, both independently and through collaboration, improve their services and increase the wealth and the quality of information by creatively using existing staff and evolving technologies. Experiments in electornic publishing of faculty and graduate students works, online course materials, and networked news reporting have becom standard library resources and services. On-campus and remote members of the community have access to both open and restricted resources. It has improved patron services as well as broadened instructional activities. With faculty and student support and collaboration with units outside the library, a new variety of resources, access methods, and services are now available for extended campus users as well as traditional library users. Especially significant is the library’s ability to evolve into new roles as needed.

The Scholarly Communications Project at Virginia Tech’s University Libraries has created a variety of partnerships with units and individuals within its university community to produce unique online resources particularly suited to incorporating digital library resources and services. Throught activities like those at http://scholar.lib.vt.edu, libraries demonstrate that they are vital and fulfilling their functions to provide current as well as historical works and expanding traditional services to meet the needs of extended campus learners and faculty.

From Laser Disk to Internet Access: Catalogue of the History of Architecture

Begun as an experiment in 1995 (?), SCP personnel extracted architectural images from a digital video laser disk to give the Virginia Tech community timeless access to the historical architectural images. Staff rekeyed and scanned the index (sometimes OCR worked!) and linked the images to the index.

As more and more people use the Web, SCP fields a growing number of requests for access to the images, especially by students and teachers. These requests are forwarded to the author of the laser disk, Professor Humberto L. Rodriguez-Camilloni (check name).

From Photos and Negatives to Internet Access

In the spring of 1994, the university's PhotoGraphic Services unit began digitizing images from the Libraries' Special Collections Department. Two sample collections became available: original watercolors of cadet uniforms worn by Virginia Agricultural and Mechanical College and Virginia Polytechnic Institute students, and fragile glass negatives from the Norfolk & Western Railroad Archive including railroad engines, bridges, and advertisements.

Experiments in digitizing, storing, identifying, and presenting non-textual files were a new challenge. SCP staff added brief text and appropriate hypertext links that connected, for example, N & W railroad pictures to the Virtual Railroad, or an older style cadet jacket to its later version.

Interest has grown steadily in providing networked digital images and encouragement from VIVA, the Virtual Library of Virginia, lead to preliminary scanning projects focusing on historical resources unique to southwest Virginia. SCP provided staff, equipment, and software resources so that several small collections from the Special Collections Department, including its University Archive, could benefit from this early experimentation.In addition to ImageBase, access, there are hot Web links from the descriptive guides and finding aids to selected digital image collections.

The future of SCP should include stepped-up outreach to faculty to better inform them about what is available from the library and to advertise SCP initiatives. SCP should also build on current strengths, such as promoting greater innovation in electronic journals and making ImageBase more broadly applicable to campus. It should look for innovations CSALT/SOLE-like opportunities. It could also develop a family of standards for scholarly e-publishing. Further work needs to be done in a variety of areas, including intellectual property mangagment, archiving and preservation, and multilingual appeal (not to mention metadata; etext to MARC; and keywords to thesaurus vocabulary).

A REAL CONCLUSION MAY BE HERE!!! [from Academic Support Units: Process for Program Review--Based on review of: 1996/2001

Scholarly Communications/Special Collections actively promotes the research and outreach missions of the university by (1) providing a mechanism for electronic publishing of research and scholarship conducted by members of the university community; (2) providing an accessible archive of current and historical research; (3) by turning experimental scholarly communications into standard operating procedures. These activities also support students' learning and faculty teaching The Libraries has accomplished these and more with in-house expertise, equipment and staff.

Without support from Scholarly Communications/Special Collections university research and scholarship would be delayed in getting to students and faculty by traditional publishing and library processing routines. Students, researchers, and scholars would not have constant access to the wealth of our university community's research and scholarship available in electronic journals, or to course materials, and digital images and unique materials currently housed in Special Collections. Their use would be limited to the operating hours of the library and to availability when other patrons have not checked out the materials, or to purchasing individual copies. The Libraries will have to continue to pay for staff costs associated with technical processing (e.g., binding, labeling, security stripping) and circulating, shelving, and interlibrary lending without the electronic access provided by SCSC.

The effectiveness of some of these services have been measured through student and faculty surveys. All students and faculty having class materials accessible through the Libraries' EReserve system were surveyed spring semester 1995. 75.4% rated it 1-3 on a scale of 1-5, with 1 being very easy to use. On the same scale, 83.3% of the faculty rated it 1-3. The same percentage of faculty had no trouble converting their files to the required form and 100% preferred EReserve to traditional Reserve. The second draft of a Web survey of EReserve users for spring semester 1996 is being reviewed. Access statistics for course materials are also available for each semester of availability at http://reserve. lib.vt.edu/ Increased use is another measure of satisfaction. Spring semester 1995 eight faculty had course materials available for nine classes. Spring 1996, 31 faculty were teaching 39 classes using EReserve materials.

EJournal editors/readers have not yet been surveyed, but access logs show dramatic increases in users during the past three years. Gopher access increased nearly 300% during 1993-1994; increased accessed ranged from +93% to +522%. World Wide Web access to ejournals increased more than 1100% during 1994-1995. Another indicator of the success of ejournals is in the growth of the number of ejournals published. The Scholarly Communications Project published five ejournals in 1993-1994 and it had 14 ejournals to its publications roster in 1995, and two more became available in March 1996. The demand for this library service has increased dramatically without the library seeking additional ejournals to publish and without increased staff.

In both of these sample areas, EReserve and EJournals, Libraries' services have changed over time to respond to user needs. For example, EReserve originally accepted only PDF files but now all file formats are accepted, including links to external home pages, and EReserve provide the faculty space on the EReserve server for their course home pages. The limited staff resources can do so much because faculty have the responsibility for preparing e-course materials (including scanning articles) and lectures, and for determining how their materials will be presented, accessed, and archived.

Future activities of the SCSC include a Digital Imaging Project where the Libraries partners with PhotoGraphic Services and the Art/Art History Department to convert 300 of +100,000 art/architecture/art history slides to develop a prototype for incorporating art into the design of digital course materials and digital student study aids. The future success of this project depends somewhat on upgrading classroom projection equipment, but the current status of computer labs makes the usefulness to students well worth the effort.

Another collaborative effort is the proposed Virginia Tech Digital Library project where the Libraries partners with Computer Science and Computer Network Systems (SO--is this who Lucy and Robert are affiliated with?) and private industry and other research universities (what to call the IBM consortium we're now part of?) to develop a prototype digital library. The VTDL would integrate the current bibliographic database (VTLS) with access to reference databases and Internet resources, including SCSC materials such as electronic theses and dissertations, digital images (including those converted from art slide collections and Special Collections), EReserve course materials, and more.

SCSC is set up to experiment in scholarly communications but not to put its prototypes into production and heavily used services. It struggles to maintain EReserve at a high level of service, for example. The advent of ETDs in 1996/97 will certainly test the limits of its resources. The last few years have not seen an increase in the budget for this unit . Dramatic increases in the number of faculty and students served with additional staff to assist them in creating innovative electronic information resources including innovative electronic journals and course materials. Additional electronic publications are certainly in the offing with a university community as focused on research and scholarship as is Virginia Tech.

*****

ETD processing was scaled up, resulting in approximately 200 new titles being network-accessible in 1996/97.

SCP installed the OpenText LiveLink search engine in January 1997, providing faster and more accurate results than did the previous freeWAIS (Internet shareware) search programs. Throughout this fiscal year, the continually growing collection of ETDs was reindexed to enhance user access and to serve as a model for other libraries and universities considering participation in the Networked Digital Library of Theses and Dissertations. Subsequently, SCP provided search access to its electronic journals.

Scholarly Communications and Special Collections staff cooperated with PhotoGraphic Services to advance the prototyped "digital image database" and scale it up to a production-oriented ImageBase with nearly 3000 fully identified images (fy 1996/97). The time lag between digitizing and identification is minimal with a cadre of student workers. All digital images are described using the near-standard Dublin Core metadata tags. Without VIVA support, digitizing would have been severely limited, but now the library can provide access to images including those from the Norfolk & Western (2250), local history (475), Earl Palmer’s Appalachia (150), and VPI Women (75).

An Nahar, began almost daily online publication in June 1997

The European, began weekly online publication in April 1997

Le Monde, began daily online publication in March 1997 (plus supplements)

improved definitions of fields

shared online archiving of TIFFS (archival quality images) with Computing Center

The loss of Peter Haggerty fall semester 1996 and then James Powell in December, left Scholarly Communications heavily dependent on the good will of Library Automation for computer support. The upgrade of the technical assistant’s position to that of a programmer, meant the welcome addition of Tony Atkins (by way of North Carolina with a Bachelor of Fine Arts from the University of Southern California) in April 1996. He has already noticeably improved displays on the Web and scripting for ejournals, VA News, and ETDs. Assuming James Powell’s position as Technical Director, Valeri Poltavstev, joined the SCP staff in May 1997. His first accomplishment improved computer access security both in SCP and Special Collections.

Scholarly Communications and Special Collections will use technology to promote access to information resources at Virginia Tech and the University Libraries. Emphasizing collaborative strategies that respond to and anticipate the needs of the university community and the public at large, we will extend access to scholarly electronic publications and to one-of-a-kind resources.