The Library as Laboratory


John Unsworth




Presented at the Annual Meeting of the American Library Association

Session of the ACRL Law and Political Science Section

and the ARL Office of Scholarly Communication:

"The Politics of Scholarly Communication in the New Millennium"

Sunday, June 27, 1999

9:30 a.m.-12 p.m.

New Orleans, Louisiana



The description of this session in the conference program notes that “Technological changes . . . are increasingly requiring collaboration among university faculty members, administrators, librarians, and scholarly publishers,” and it goes on to claim that we will “encourage librarians to help promote an environment conducive to effective research.”


I'd like to follow through on that, or at least on part of it. I'm not going to make the argument for doing electronic scholarly research--I've made that argument elsewhere, and often. I'm also not going to spend much time on the new relationships that scholarly publishers need to establish with scholars, though if you're interested in that topic I invite you to have a look (on the Web) at the paper I just delivered in Austin at the AAUP's annual meeting. Instead, in the time I have today, I will argue that the library is the most suitable place in the university to house electronic scholarly research projects in the humanities. Metaphorically speaking, the library has always functioned as the laboratory for the humanities, but it needs now to literalize that metaphor, and doing so will require new commitments of two precious commodities--space and money--which will inevitably displace other projects and priorities, and will therefore meet with some resistance. Nonetheless, I think the future of the library lies in fulfilling this new role.


At the outset of this session, Ken quoted Tip O'Neil to the effect that all politics are local--I'd agree, and I'd add that the devil is in the details. With those maxims in mind, I hope you'll forgive me for delving into some local details in what follows--I think the details are extremely important to understanding how this new collaborative relationship actually works, in applied political and economic terms, and how (or whether) it might be replicated at other institutions.


The Institute for Advanced Technology in the Humanities, where I work, is housed in Alderman Library, the University of Virginia's graduate library. IATH has been in Alderman since it began, seven years ago. The Institute takes up about 2000 square feet--and that's a conservative estimate focusing on our main third-floor office space, one that doesn't count our machine room on the first floor or the faculty studies occupied by ongoing IATH projects on the fifth floor. The library donates this space to the Institute--it was formerly occupied by the Microfilm department, which has been moved to a considerably smaller space on the other side of the third floor. We pay for our own networking costs and we pay our own phone bills, but we pay no rent to the library (or anyone else). When it's time to recarpet or repaint the Institute's space, we split these costs with the library. We have about a dozen offices (mostly cubicles), some public meeting space and some public workstations (public in the sense that they can be used by any IATH fellow or project assistant), and our own keys.


Now, allow me to explain a little about how the Institute works:


* Budget: IATH is a permanently funded research unit of the University, with budget coming from the office of the Provost. We cost the University about $350,000 a year, with an equal or greater amount coming in, in any given year, in the form of grant funding. From 1993-1998, the University has invested just under two million dollars in the Institute, and IATH has raised about 3.25 million from outside sources.


* Reporting: I report to the Vice-Provost for Research, which means that IATH is outside the College structures at the University--not part of the College of Arts and Sciences, not part of any other school or department. In fact, we are also not part of the library, from an administrative point of view.


* Staffing: We have nine full-time staff, and a number of students working part-time on IATH projects. Two of our full-time positions are systems administrators with expertise in configuration and deployment of document management, SGML, and database software; one concentrates on imaging and modeling applications; one is a Java programmer; one is one is a development officer; one provides administrative support and fiscal administration; three are faculty-level appointments (an associate professor in computer science--Worthy Martin--who serves as our technical director; an SGML and information systems expert from a library background--Daniel Pitti--who serves as our project director; and an associate professor in the English department--me--with research interests in publishing history and scholarly communication.


* Fellows: Two fellows in residence are selected each year through a competitive application process. Faculty in any humanities discipline are eligible to submit proposals for long-term, large-scale research projects that use information technology as a research tool and publishing medium. Fellows are granted half-time teaching release (donated by their academic departments), and their departments also provide ten hours a week of student assistance during the academic year. IATH matches those students hours, and picks up the slack in the summer, and also provides a $10K project budget, office space and equipment, programming, information systems design, and consulting for the life of the project. We also work in less intensive ways with associate fellows (UVa faculty with no teaching release, no office space, more modest staff support from IATH), and networked associate fellows (researchers from outside UVa; intensive support of these projects is usually contingent on grant funding).


The Institute is one of several electronic centers housed in Alderman Library and next door in Clemons Library--these include:


* The Electronic Text Center: Established in the same year as the Institute and located next door to us, the Etext Center is run by David Seaman: it provides walk-in service and training to faculty and students interested in finding, creating, or using electronic text resources. The Etext Center also builds electronic collections, and its staff spend a fair amount of time creating or enriching elements of these collections, either on their own or in conjunction with faculty.


* The Geostatistical Information Center: Formerly maps and social science data, the GeoStat center is run by Patrick Yott, and it provides a wide range of consulting, programming, and collections development services with respect to mapping software, statistical packages, and the visualization of statistical data. It's also on the third floor of Alderman, right down the hall from IATH and Etext.


* The Virginia Center for Digital History: This is the newest of the electronic centers housed in the library, having been established only last year. It is located next door to IATH and is an outgrowth of an Institute project--the Valley of the Shadow, Ed Ayers' civil war archive. VCDH works with faculty at UVa and elsewhere to support teaching and research of history in electronic form. Like IATH, it is administratively distinct from the library: it is funded by and reports to the Dean of the College of Arts and Sciences.


* Digital Special Collections: Edward Gaynor is in charge of digital special collections, a unit within Special Collections that focuses on digital imaging of rare materials, the provision of networked finding aids, and mounting the online components of Special Collection's exhibitions.


* The Robertson Digital Media Center: Rick Provine runs this recently endowed center in a newly renovated section of Clemons Library, next door to Alderman. The Digital Media Center specializes in digital audio and video: like the other library centers, its mission combines collections development, training, and collaboration with faculty and students in teaching and research. Starting this fall, the Digital Media Center will also house our new Director of Media Studies, Johanna Drucker, and it may also provide support to a new Media Studies Postdoctoral researcher supported by the Woodrow Wilson Fellowship Foundation.


In addition to its centers, the library has an overarching digital libraries division, charged with designing and planning the systems that will manage and deliver the objects in a unified digital library in the future.


Outside the Library, there are also a couple of relevant programs run by the University's campus computing organization, Information Technologies and Communication (ITC). These include:


* The New Media Center: This is a walk-in service center where staff assist faculty and students in digitizing, modeling, imaging, and working with audio and video. The New Media Center has existed for about four years, and its services overlap in some ways with the library's Digital Media Center.


* The Teaching + Technology Initiative: This program, now about three years old, provides release time and technical support to faculty members who propose to integrate information technology in undergraduate teaching. It is managed by John Alexander, in conjunction with the University's (originally Lilly-funded) Teaching Resource Center. I'm a TTI fellow this year, with a project on 20th-century American Bestsellers.


Clearly, then, the Institute is not an isolated effort, but is part of a much larger picture. Most of that larger picture is made up of initiatives that are library funded; some, like IATH and VCDH, are funded from other sources but housed in the library. Only a couple of these--the ITC initiatives--exist entirely outside the library's domain. It is worth noting, by the way, that it's much easier to stick to a limited mission and do it well in an environment where other bases are covered by other organizations. Beyond the general benefit of being one of many centers, there are several benefits, from my point of view, to a library-centric arrangement:


* For IATH's fellows, being in the library means they have ready access to library collections, especially those primary materials not in digital form. Just like any other faculty, fellows also can benefit from the assistance of staff at Library electronic centers in relevant portions of their projects. Also, the large number of electronic centers, all of which employ students on a part-time basis, means that there is a sizeable pool of students who are trained in some fairly arcane software tools, who understand something about building digital collections, and who can be recruited to work on IATH projects. It might seem that networked humanities research projects such as those sponsored by IATH could easily be distributed across campus, and that no physical facilities should be required--but in fact, I believe the most important part of IATH's success has been the existence of a central place where faculty from across the humanities and technical staff versed in humanities computing can work together. The Institute has become the University's most genuinely interdisciplinary space, and its staff and fellows benefit in a number of ways from cohabitation.


* For IATH technical staff, being in the library makes it easy to learn from and collaborate with staff in other library digital centers (most of which are very close at hand). The Institute's location and its proximity to library operations also subtly change the way the staff thinks of their jobs, in my opinion: though they work with information technology, I believe they come to think of themselves not only as technologists, but also as information professionals.


* For the Institute in general, the same is true, I believe: being in the library has had a definite impact on the way we perceive our mission and on the methods we use. Granted, the goal of supporting long-term computer-based research projects would probably have driven us in the direction of non-proprietary hardware- and software-independent encoding schemes, but I think our setting has helped to promote and reinforce an awareness of long-term preservation and access issues, an interest in best practices, and a prejudice in favor of community-based standards. In fact, in retrospect, it seems quite obvious that the Library was the best, and perhaps the only, place to establish a place like the Institute for Advanced Technology in the Humanities.


What's maybe not so obvious is why the Library puts up with us. I think there are several answers to this question, though:


* Kendon Stubbs: The first and, in many ways, the most important reason that the Institute is in Alderman Library is Kendon Stubbs. Kendon is Associate University Librarian and for the past seven years, he has consistently championed and promoted library digital centers and the use of library space for humanities computing initiatives. Many others have been involved in the development of these centers, and others, including the director of the University's Libraries, Karin Wittenborg, have seconded Kendon's commitment to them, but I think few people would dispute that Kendon has been a driving force in turning the University of Virginia into an international leader in the fields of digital libraries and humanities computing.


* Public Relations: Because these library digital centers exist, and because faculty and students come to them in order to work on their research and teaching projects with library staff, the library has come to be seen as a place that houses human resources as well as information resources--a place you go to find expertise, advice, training, and consulting, not just somewhere you go to get books and journals. In fact, as the library steadily provides more and more of its books and journals in electronic form, and even permits faculty to use the web to order the delivery of physical items to faculty offices (using LEO, Library Express On-Grounds), I would say the general perception of the library is shifting increasingly away from the repository and toward the laboratory, or (to use a favorite neologism of the science community a few years back) the collaboratory. In short, places like the Institute and the Etext Center promote a faculty culture that regards the library as an essential space for experimentation, discovery, and creation, and regards librarians as expert colleagues.


* Shared Computing Infrastructure: The hardware, software, and human infrastructure required to run something like the Institute is considerable, and it overlaps quite a bit with the requirements of library centers building and delivering digital collections. The Institute uses no library funds for this infrastructure: it attracts funding from other parts of the University and from sources outside the University, and it shares infrastructure with the library. For example, the Digital Media Center uses a substantial amount of disk on our main server, a Sparcserver 1000 with 250 GB of disk and a gigabyte of RAM, to deliver streaming audio and video. They, in turn, share the media server software with us, and we back the whole thing up using our staff and equipment. We have shared a full-time position with the Etext Center until recently (a programmer analyst specializing in text delivery and text database systems), and in the past we have provided considerable systems administration to the Etext Center in particular. We often will consult with other library centers when an expensive software or hardware purchase is in the offing, to see if the use and the expense of that resource can be shared, and in recent years we have pooled our resources in this manner to purchase the IBM Digital Library package and a high-end server to run it on.


Out of the collaboration between the Library, the Provost's office, humanities departments, and humanities faculties, more than forty projects have been started--the Rossetti Archive, the Blake Archive, the Valley of the Shadow, the Pompeii Forum Project, the Piers Plowman Project, as well as projects on Walt Whitman, Emily Dickinson, the Salem Witch Trials, Tibetan mystical literature, and many others. Some of these have gone on to be published by academic and commercial publishers; others, like the Blake Archive, have so far been able to recover their costs through grant funding. Its clear that we don't yet know what the best funding model will be for projects of this sort--or whether there will be many, or not even one. But for exactly that reason, I would say that we need publishers to be involved and to take a few chances, along with the rest of us.


So, what lessons can be drawn from the past six years at the University of Virginia?


* humanities research computing thrives in a library setting


* there are tangible costs associated with establishing new centers of expertise (and I would add that the most significant expense, both in dollar amount and in impact, is the money spent on staff), but those costs can be spread across different units of the University and can be offset (though not replaced) by outside sources.


* nonetheless, someone in a position of leadership has to be the first to put resources on the table


* the most important initial resource is a commitment of space, since space is required in building a laboratory of the non-metaphorical sort


* an isolated effort in the area of humanities computing is likely to be overwhelmed with incompatible requirements, whereas a more narrowly focused effort set in the context of other, complementary efforts has a much better chance of success


* initiatives that grow out of and correspond to traditional university functions have a clearer sense of their own mission than centers created ab novo


* nonetheless, humanities research computing is inherently interdisciplinary, and there are both intellectual and fiscal economies of scale to be had in creating this kind of center above the department, and perhaps even above the college, level


* The place where research across the disciplines of the humanities has traditionally been conducted is the library, which brings us back to the first point, that humanities research computing needs its laboratory to be in the library.