A Brief History of the Institute for Advanced Technology in the Humanities

John Unsworth, Director (1993-2003)

The Institute was founded by the University of Virginia in 1992, in response to a call for proposals from IBM. This part of the history was before my time, but as I understand it, Bill Wulf, UVA's AT&T Computer Science Professor, who received the invitation to submit a proposal, suggested that the focus of the proposal should be on humanities computing, for three reasons: first, humanities departments were (and are) among the strongest at the University of Virginia; second, computers were becoming sufficiently capable of multimedia to be interesting to humanists; and third, Bill had a hunch that some of the most interesting computing problems in the last decade of the 20th century would be in the humanities.

A second character enters the history at this point: Alan Batson. Alan was also a faculty member in computer science, and at that time was head of academic computing at the University. Alan was also a member of the steering committee that designed the Institute. The other members of this steering committee were Kendon Stubbs and John Price-Wilkin, representing the Library that became IATH's home; Randy Pausch, Bill Wulf, and Ira Jacobson, from the College of Engineering; and Jerome McGann, Ed Ayers, and Dick Sundberg, representing the College of Arts and Sciences. McGann and Ayers later became the Institute's first two fellows, and both their projects are still actively developing.

Alan's contributions were many, but two stand out: first, he contributed the Institute's two full-time and one-half time staff members to the Institute's staff, for the first three years of its operation--Thornton Staples, Ross Wayland, and Rob Cordaro, all employees of Academic Computing, and later ITC (Information Technology and Communication). In the planning year, he also contributed Deborah Mills as a manager, without whom my job as the first director would have been a good deal more difficult than it was. Alan's second criticial contribution was to insist that the focus of the Institute should be on research, rather than on teaching. I believe there was considerable pressure to turn this opportunity to the direct support of teaching with technology, but Alan insisted, against all comers, that if research was what was rewarded in a research university, then the focus of this project should be on promoting computational research in the humanities. Teaching innovation would flow from changes in research behavior, and we already had a decade of proof that trying to change faculty culture in the humanities by putting technology in the classroom didn't work.

I have always understood the Institute's recipe for success as the recipe for stone soup: everyone puts in something, and the result is more than any of them could have produced alone. In the past ten years, the Library has been very generous in contributing space for IATH, ITC has contributed staff, in the early years, and systems support in more recent times, the Provost has provided IATH's operating budget (increasing from $250K to about $450K per year over the decade), key faculty members have contributed their effort and put their reputations on the line, and academic departments have provided release time and modest support for student research assistants. On this foundation, the Institute has developed dozens of long-term, large-scale projects, attracted millions of dollars in outside funding, and produced some of the finest examples of humanities research computing, on both technical and intellectual terms.

Mission and Method

This is the mission statement of the Institute for Advanced Technology in the Humanities, quoted from IATH's home page at http://www.iath.virginia.edu/ :

As a research unit of the University of Virginia, IATH's goal is to explore and expand the potential of information technology as a tool for humanities research. To that end, we provide our Fellows with consulting, technical support, applications programming, and networked publishing facilities. We also cultivate partnerships and participate in humanities computing initiatives with libraries, publishers, information technology companies, scholarly organizations, and others interested in the intersection of computers and cultural heritage.

With respect to IATH's faculty fellows and their projects, I would describe IATH's mission as making sure (sometimes over the objections of the fellows themselves) that we build things to last, rather than going for immediate gratification. If IATH projects last ten years or more, we want to make sure that the fellows (and IATH's staff) don't spend the last five years redoing all the work they did in the first five, in order to keep it current with technology. This has meant a general commitment to open, non-proprietary standards (like SGML, XML, XSL, Java, Javascript, etc.) over proprietary methods, and it has also meant a commitment to separating data capture methods from data delivery methods. In practical terms this means, for example, encoding text in XML using TEI or some other DTD/schema specifically designed to carry, say, literary texts, and then working out a way to render that TEI into HTML for browsers, rather than just encoding the texts in HTML directly. The reason for this is simple: HTML tells you very little about the content it carries or about the structure it expresses, and that means that future transformations from today's HTML into tomorrow's HTML or some other encoding will have very few hooks on which to hang automated transformation--instead, everything might need to be re-encoded by hand. IATH's more indirect method insulates content from the rapid change in delivery mechanisms, and allows greater and more specific expressiveness in the encoding. Taken together, those two things add up to longevity in a rapidly changing medium. Inevitably, though, this indirection sometimes produces frustration, since more mediation means more intermediates, and less direct control over all aspects of the project. This is also an inevitable side-effect of true collaboration, in which no one person controls all the knowledge or all the expertise required to get things done. Even so, longevity remains IATH's goal with respect to the projects it produces, and collaboration remains its method.

Programs

Fellowships in Residence are available to any full-time academic faculty member in any humanities department at the University of Virginia, through an annual competitive application process. Applications must describe a humanities research project that uses information technology as an analytical tool (and not simply as a means of composition or a publishing medium), and they must be accompanied by a letter from the applicant's department chair promising to provide half-time teaching release for one year, and 10 hours a week (during the academic year) of research assistance from a student, if the project is selected. Applications are screened by a selection committee, consisting of past fellows, staff members, library faculty, and others.

From 1993 to 1999, the Institute took on two new fellows in residence each year, for a one-year residency, but with an open-ended commitment to support the fellow's project. This open-ended commitment has some obvious costs and it has been controversial among staff at times, but I believe it has also been one of the keys to our success: projects like the Valley of the Shadow or the Rossetti Archive--IATH's first two projects, both very well known and still actively developing--take time to come to fruition. In 2000, though, faced with a flat budget, static or declining staffing, and steadily increasing commitments to fellows, we took a year's hiatus from adding new fellows. Thereafter, starting in 2001, we decided to add one new fellow a year, instead of two, and to provide a two-year residency, instead of one year--though we haven't asked the fellows' departments to increase the teaching release, which continues to be half-time release for one year. We have continued the Fellows in Residence program in that form until the present. The two-year residency seems to work well in several ways: first, it gives us a full year to work on designing the project--and design is a critical piece of these projects, given that they take shape over years in a medium that evolves in cycles of a few months. Second, it allows us to put off grant writing until the beginning of the second year, but then allows us a full year to draft proposals, share them with program officers, revise, and submit; also, by the time these proposals go in, there is almost always a substantial (and genuine) draft of the project to show, on the Web. When Fellowships in Residence lasted only one year, there was considerable pressure to produce proposals during the first year: this inevitably led to spending time mocking up demo sites, which took time away from designing and building the real thing; in addition, since the fellows were not very far along in their projects, they often didn't produce very compelling proposals.

IATH also sponsors two other kinds of fellowships: Associate Fellowships and networked associate fellowships. Associate fellows are identified through the same selection process that is used for fellows in residence, and they are University of Virginia faculty. These fellows do not get teaching release or office space at the Institute, but we do provide them with server space for their project, with consulting and support, and with grant-writing services if needed. The purpose of the Associate Fellowship is to provide appropriate support to a project that is not ready for a fellowship in residence, and needs to ramp up, or to one that simply needs help in finishing something. David Germano (of the Tibet project), for example, started as an Associate Fellow, and in the next year became a Fellow in Residence; at the other end of the scale, Mike Gorman used an associate fellowship to finish a migration to the web for his project on Bell's invention of the telephone. Networked Associate Fellows are faculty researchers from outside the University of Virginia. This category of fellowship is intended for groups who need to use the network to collaborate; the Institute provides them with access to our computer resources, occasional consulting, and grant-writing services. The Blake Archive is an example of a project in this category.

Budget

During the first two years of its existence, IATH's budget was a temporary one. The expectation, made clear when I was hired as Director, was that the Institute would become self-supporting within three years. During my second year as Director, I had an outside consultant work with me to draw up a business plan for accomplishing that goal. The plan calculated that, if we doubled the size of our staff (from 3 to 6) and spent a substantial amount of time consulting on electronic publishing and web issues with cultural institutions, we could expect to carry the cost of the additional staff and about half of the original staff. When presented with these projections, Provost Peter Low decided to remove the expectation of self-support, and he made the Institute's budget permanent, charging us simply to do research, rather than running a business to support our research activities.

Over the next several years, the Institute's budget grew, on paper, but the actual level of annual state funding remained at about the $250,000 level; meanwhile, in real terms, the staff and the list of projects grew steadily. In 1996, ITC transferred about $130,000 from their permanent budget to IATH's, allowing us to own the 2.5 FTEs that had up to that point been on loan from ITC, and hire our own people: on paper, this increased IATH's buget from about $130,000 to about $260,000, but in fact the level of University support for IATH remained the same. Also in 1996, the University increased IATH's budget by about $45,000, when Ed Ayers negotiated salary for a technical director for IATH, as part of an agreement not to leave the University of Virginia. Paul Jones was hired on that funding, at 2/3 time, with 1/3 in ITC, but Paul returned to UNC the next year; shortly thereafter, Worthy Martin, associate professor of Computer Science, joined IATH as technical director, and he remains with us today. Also during this period, the Vice Provost, Gene Block, agreed to provide up to $65,000 a year to underwrite deficits at IATH, as a guarantee that the central administration would help us to raise enough endowment to generate that amount of income. That never happened, but in 2000, $85,000 a year was permanently added to our budget by the Vice Provost's office, and the deficit agreement came to an end. Also in 2000, we had a one-time cash infusion from the Provost to renew our technical infrastructure, and we were granted three years of help with unix systems administration from ITC, valued at $30,000 a year. At this point, through these and other increases, our annual budget had grown to about $450,000 a year in state funding (and about that much again from outside sources); it increased slightly in 2001, to about $485,000, but then it began to decline, as the state cut the University's budget in 2002 and 2003. Even so, our budgets have balanced in recent years--though we have achieved that this year by not hiring to replace departing staff. At this point, when all our positions are filled, we normally employ ten full-time people, and we receive about $400,000 a year from state funds (and we are now paying for systems administration assistance from ITC, out of our regular budget). As one would deduce from those figures, we are heavily and increasingly dependent on outside grant funding.

A point worth mentioning here, in that connection: we have never been given permanent funding for our development position (indeed, that is the reason for the deficits we once asked the Vice Provost to cover). We used a part-time student for this in the early 1990s, and then in the late 1990s and early in this decade we have had an unfunded full-time development officer. I have been willing to scrape together the salary for this position, with a combination of begging and borrowing (no stealing, yet) because of the importance of grants to the Institute's overall financial picture. Over the last ten years, we have brought in well over seven million dollars in grant funding, largely through the work of our development officer. During that same time, we have reaped very little benefit from University Development--though in the past year, they have helped to support the development officer, as a one-time gift, and we are grateful for that. It's clear, though, that we cannot depend on the kindness of others--we need to employ our own development officer. It would greatly ease the budget situation at the Institute if the University would recognize that necessity, and fund the position.

Staffing

In the beginning, IATH's two full-time staff members were each assigned to work with one fellow, in the first year (Thorny Staples worked with Jerry McGann, and Ross Wayland worked with Ed Ayers). When I arrived, it was clear to me that this wasn't a scalable model for staffing, and I replaced the buddy system with a plan to develop a range of expertise on staff, and apply that expertise as needed in particular projects. We have done that, and at present, IATH has a staff of 8.5 FTEs spread over ten people, with one empty full-time position. Those people and their skills are:

  1. Robert Bingler (full-time, java and database programming)
  2. Regina Carlson (full-time, development officer)
  3. Cynthia Girard (full-time, XML/SGML/XSL and systems administration)
  4. Chris Jessee (full-time, mapping, modeling, and graphics)
  5. Felicia Johnson (half-time, shared with VCDH, web design)
  6. Worthy Martin (half-time, shared with Computer Science, management and technical oversight)
  7. Daniel Pitti (full-time, management and information architecture)
  8. Joy Shifflette (full-time, administrative services)
  9. John Unsworth (half-time, management, systems administration, and long-term planning)
  10. Sarah Wells (full-time, technical writing/documentation)
  11. Empty Position (full-time, programmer in Java, C++, Perl etc.)

IATH's staff is a talented and dedicated group of people, with very different abilities and backgrounds. They are also IATH's greatest asset, and though they do not all come from humanities backgrounds, they do come to understand the vocabulary, the values, and the goals of humanities research, and they bring their own technical and intellectual contributions to achieving those goals. IATH's staff must keep up with rapidly evolving technologies, must be creative and resourceful in applying those technologies to address the needs of the fellows' projects, and--like the fellows themselves--must be able to work in a collaborative environment, where interdependence and mediation is the rule.

Impact on the University

The stated purpose of the Institute is to make information technology an accepted part of faculty culture in the humanities at the University of Virginia, and specifically to do this by changing the way faculty do their research. It was held, by members of the planning committee, that changes in teaching would follow from changes in research methods, and in nearly every project we have worked on, this does seem to have been the case. IATH has also markedly increased the number and size of grant proposals being submitted by humanities faculty to federal and private funders. Many of these proposals have been successful, and over time, that has had a significant impact, especially on the employment and training options open to graduate students. In some cases, IATH research projects have grown into independent entitites, like the Virginia Center for Digital History, or David Germano's Tibetan and Himalyan Digital Library, or the Rossetti Archive. The Institute has also played an important role in the development of the MA in Digital Humanities, a unique and important new graduate program now recruiting faculty, and scheduled to enroll students in the fall of 2004. And finally, IATH has established the University of Virginia as the international leader in humanities computing.

Impact Beyond the University

IATH has played a leading role in two of the major standards efforts in the humanities computing/digital library world, the Text Encoding Initiative (TEI) and Encoded Archival Description (EAD). TEI is used (by editors, linguists, and other researchers, and by libraries) for encoding literary and linguistic texts. EAD is used in libraries and archives for encoding finding aids. Between them, they cover much of the territory of digital primary resources for humanities research. John Unsworth, IATH's director, developed and executed the plan for incorporating TEI as a non-profit membership organization, and (along with the Etext Center) IATH serves as one of its four hosts. Daniel Pitti, IATH's project director, is the original architect of EAD, and has continued to play a central role in the development of that standard. IATH has also been the sole humanities participant in the national supercomputing program, under education, outreach, and training. In 1999, IATH hosted the annual joint conference of the Association for Computers in the Humanities (ACH) and the Association for Literary and Linguistic Computing (ALLC), in Charlottesville. Unsworth has served as an executive board member for ACH, and is currently its president. He has also served on the board of the National Initiative for Networked Cultural Heritage, and as member and then co-chair of the Modern Language Association's Committee on Scholarly Editions (CSE), as it has grappled with the evaluation of electronic scholarly editions. Unsworth is currently co-editing a collection of essays on electronic scholarly editing, co-sponsored by the TEI and the CSE, as well as a Blackwell's Companion to Digital Humanities, which should become a standard textbook for programs like our MA in Digital Humanities. The contributors to these two volumes are a who's who of humanities computing. But IATH's single most important impact outside the University of Virginia has been through the content that comes out of its projects, and that's distributed free of charge on the Web. On average, we get more than a quarter of a million unique visitors each month, and we provide them with over 100 GB of web-based content, for an annual total of over 3 million visitors a year, and over a terabyte of content delivered each year.