By Walter Henry
Having been asked to write about the impact of information technology (IT) on conservation, I find myself reflecting, rather, on why IT has not had a greater impact on our field. Computing and networking have become entwined in our daily practice, yet one can't avoid the sense that we're not getting as much out of the technology as we might—or worse, that it is not delivering on its promises.
Rather than reviewing the current state of play, I'd like to offer a few observations on where we need to go from here. Perhaps because I lack patience and expected our information environment to evolve more thoroughly than it has to date, I will undoubtedly seem more a Luddite Cassandra than the avid technology partisan I've been in the past. The common thread in these reflections is that technologies evolve more quickly than the social and psychological adaptations needed to make effective use of them.
The rate at which a notion decays from being novel and interesting to simply being self-evident, if not downright trite, is a marker of how completely expectations of rapid change now dominate our experience. To have insisted, a decade or so ago, that computers were more significant as communication tools than as computing machines would have been contentious, but today few would argue—and in the field of conservation, especially so. While computation has played an important part in scientific research and analysis, it has—with the notable exception of imaging—made relatively few significant inroads into conservation practice. In communications, however, the impact has been dramatic, especially since 1987, the year that saw the introduction of the Conservation Information Network and the Conservation DistList.
It is useful to distinguish (even if the distinction is blurry and arbitrary) between two major modes of online communication: interpersonal communications (email, online forums, two-way conferencing) and information dissemination/retrieval (databases, most Web sites, online publishing). Both modes are now important parts of the conservation landscape, though with many users, one senses greater comfort with the former mode. To judge from numerous submissions to the Conservation DistList, far too many professionals prefer relying on direct advice from their colleagues to looking into the published literature. Despite enormous efforts that have gone into making access to Art and Archaeology Technical Abstracts simple and affordable and despite the DistList's frequent reminders to "Search AATA before you post," few participants appear to take the advice.
Conservation OnLine and Knowledge Environments
Much of my thinking with regard to technology grows out of my experience with Conservation OnLine (CoOL), a Web server for conservation and allied professionals that I initiated in 1993. CoOL was originally conceived as a site for gathering much of the large body of information that falls outside traditional print literature, and for offering print material in ways that make it more useful (e.g., full-text searching of articles, hypertext dictionaries, etc.). It has, to a very small extent, achieved a portion of that aim, capturing, for example, the message traffic for a number of conservation-related email forums and providing unpublished technical reports (previously published and otherwise), a few online books, and full-text versions of several print-based newsletters and journals.
At the same time—and not entirely by design—CoOL has taken on an odd role as online home for a number of conservation organizations, tying them together in a loose network. These organizations have individual identities (their sites are clearly autonomous), and at the same time they help compose the virtual library that is CoOL as a whole. Searches in CoOL's main indexes will return items from the participant organizations' sites, as well as from CoOL proper. I envision the physical "body" of CoOL eventually dissolving almost entirely, gradually replaced with a virtual aggregation of information about resources everywhere on the network, and supplementing those resources with local documents where there are fillable gaps. This might be achieved through remote site indexing software, though the need to gather metadata sufficient to build a sophisticated information facility complicates matters and will, in most cases, require the participation of remote sites—which itself is a good thing, since we are aiming for collaboration in information sharing. A second, less-attractive approach is that of mirroring—copying entire document webs from remote sites and making them available on CoOL.
There is another direction in which CoOL might move, one that complements the concept of CoOL-as-virtual-library. Looking at CoOL with even the most generous eye, one must perceive that in no subject area is there more than a token offering, enough information to get started, perhaps, but not enough to make treatment decisions.
The answer, I believe, lies in what have come to be called knowledge environments (KE). A KE has been defined as an information service that: "offers structured access to content of all types relevant to a specific user population; includes opportunities for continued learning and the transfer of experiential knowledge; is marketed and sold as an integrated, value-added solution; and is marketed by a credible, authoritative source."
Built by cooperation between technology specialists, subject domain specialists, and librarians, the KE attempts to make available—either directly or by links to remote resources—everything a researcher needs for serious work in a single subject area, organized by people with advanced subject domain knowledge in such a way as to make the knowledge useful to specific user communities. Equally important, it incorporates facilities encouraging ongoing discourse within the user community.
When I first encountered the concept of a KE, I thought that it was exactly what CoOL and similar resources should strive toward—a single locus from which the conservation professional can locate thorough and authoritative information in any format, electronic or print. To an extent, CoOL carries some of the incipient elements of a KE: for example, the inclusion of online forums, especially the integration of the Conservation DistList, fosters the development of ongoing communications within the community. What is lacking is depth and thoroughness of coverage.
In considering knowledge environments for conservation, the question of scope of coverage is a difficult one. How narrow should the focus be? While we might build KEs that coincide with existing conservation specialties, I suspect that narrower coverage will be necessary, perhaps similar in scope to those of AATA's special supplements.
Building a KE is not a trivial task; funding development and maintenance will be a challenge. Existing KEs are principally subscription based, a model for which, given the limited economic resources of our field, I've not much optimism. The most likely avenue for development would seem to be project-based development of isolated components of the KE, which are later joined to form an integrated environment. (An excellent example of a working KE is the Signal Transduction Knowledge Environment, provided by Science [American Association for the Advancement of Science] and Stanford University.)
For as long as I've been in the field, a driving theme has been the need for ongoing training opportunities for conservation professionals. While excellent opportunities for continuing education exist, there remain obstacles for both the provider and the student, especially midcareer professionals. Cost, travel, and time away from the lab are all serious considerations. Of all the applications of IT developed in the past decade or two, none have spurred my optimism more than distance learning. Sitting on the nexus between information dissemination/retrieval and interpersonal communication, distance learning leverages IT to provide instruction to conservation professionals at remote locations. It offers a practical solution to each of the problems above and provides flexibility to both teachers and students, enabling professionals to fit continuing education into their work life.
Significant distance education projects in conservation are already in place. At the University of Western Sydney, the Nepean School of Civic Engineering and Environment offers a master of applied science in material conservation, a three-year part-time program for those entering the field, available via distance learning, as well as through on-site classes. Also in Australia, at the University of New South Wales, the School of Information, Library, and Archive Studies provides courses via distance education, including preservation administration and preservation and conservation of audiovisual materials. In Canada, the Cultural Resource Management Program at the University of Victoria offers distance courses in heritage conservation, conserving historic structures, and museum-related topics.
Most exciting of the current distance education projects is that of Paul Messier and Irene Brückle, who put together a two- way interactive videoconferencing system and used it to teach a course via the Internet on the examination and identification of photographs for students at the Art Conservation Department at Buffalo State College. The topic calls for a great deal of student-teacher interaction and involves subtle visual discrimination, which must be conveyed between the students in Buffalo and Messier in Boston. As such, the project served as a robust proving ground for the concept of using information technology for distance education in conservation. The system appears to have been most effective and offers hope that this technology will be of great significance in conservation education. For subjects that do not require hands-on experience or extensive real-time interaction between students and teachers and that have a reasonably static and well-defined content, Web-based tutorials seem an excellent means of teaching, especially for courses that are (or should be) repeated frequently, as online tutorials can be "replayed" without incremental cost. Topics in which theoretical aspects dominate are ideal candidates for this treatment.
By now, many conservation professionals have had some experience serving on committees or task groups conducting their work via email, and they have experienced both the benefits and the frustrations attending this mode of communication. From the earliest days of electronic mail, users have noted the awkward situations that can arise from email's lack of those nonlinguistic components that make face-to-face communication seem so easy—the kinesic and aural cues that constitute phatic communication, the body language that signals the content of what is not being expressly said (or, in this case, written).
Despite its speed and glibness-encouraging easiness, email is not speech, nor is it quite the same as print. There is a distinct quality to electronic communications, and for the most part our psychological perspective has not yet adapted to the new mode. With our lifelong grounding in telephony and print—almost polar in their sensory and psychological foundations—we've developed shared expectations of how communication works, expectations that the new mode undermines. We're developing new behaviors, maladaptive perhaps, derived from these expectations, which at best lead to wasted time and effort, at worst to failure of the effort, and in any case to a gnawing sense that something about this mode of working just isn't quite right.
Of the problems I've observed working in online task groups, the most vexing one—and perhaps the one easiest remedied—is a tendency for the group never to reach closure or consensus or, more precisely, to fail to realize when consensus has in fact been reached. This would appear to be rooted in the asynchronous nature of email and the lack of kinesic cues; one is never quite sure when the discussion is over. Similarly, there is also a tendency toward false closure, an attitude among participants that having written on a subject once, they are finished, when the essence of discourse is that it runs about, a following point upon point before settling into any resolution.
During the period when the early HTML specifications were developed, members of the Internet Engineering Task Force (an international group of network designers, operators, vendors, and researchers concerned with the evolution and operation of the Internet) created a discursive technology intended to solve this problem, email forum software geared toward the more-or-less formal discussion of a large number of issues—in this case, specific clauses of a proposed specification and voting at various stages in the discussion. While this seemed to work well, most discussions are less structured than those. Nevertheless, the idea showed promise, and other collaborative tools, notably collaborative authoring tools, have since been developed. Most are designed for use within an intranet, but Internet-based systems are entering the market, and some may be helpful for collaboration among conservation professionals.
These technological solutions, however, beg the question. The point is that we have not yet adapted socially and psychologically to the new media. An obvious quick fix is for a leader to declare a deadline and to announce the final consensus, but in practice the oft-noted democratizing proclivity of network discourse seems to militate against that. In practice, more often than I can remember, such discussions are resolved offline, typically face-to-face or by phone, with participants asking each other, "Are we done?" One assumes that with time, our vocabulary of online conventions will grow sufficiently to make such aberrations unnecessary.
Computer communication is a supremely effective information discovery tool, but reading substantial bodies of text from a display screen is for most readers not compatible with careful, considered reading—the kind needed to transform information into knowledge. Alex Pang, a colleague at Stanford, commented that when he assigned an all-Web reading list, he noted a marked superficiality in his students' reading. When he instituted a print-on-demand system, encouraging his students to read from hard copy, the situation improved.
This phenomenon is common, and it is probably a factor in the continuous retreat of the "paperless office." Indeed, when I watch people reading Web pages online, I notice that they tend to scroll quickly, scanning rather than reading. Ease of scrolling encourages this mode of acquisition, rewarding rapid scanning with quickly found answers, an electronic form of speed reading.
The implications for information management in technical fields are clear. I'm no fan of presentation-oriented document formats, but I concede that when presenting complex, hard-to-read materials, online services should offer print-friendly versions in—more or less—platform-independent formats such as PDF, at least until readers better adapt to online presentation. In the longer term, we must relearn to read.
Elsewhere I have written about some of the technical challenges facing those who would construct database and document authoring systems to support conservation treatment and examination documentation. Beyond those technical issues, however, lies a far more intriguing problem. A colleague, Lisa Mibach, pointed out that conservators spend much of their time looking intently at objects, and that having to lift the eyes and hands to use a computer breaks the concentration enough to seriously interfere with the examination.
Over time, humans have learned to integrate handwriting so completely into our behavior that writing while looking does not introduce a cognitive disjuncture. But we have yet to adapt to computer input in the same way. Indeed, with the current configuration of computing devices, it would seem unlikely that we will ever fully adapt, although more recent handheld computers may be moving in the right direction.
In the past year or two, voice recognition technology has made great advances, and it is now possible to buy consumer-level dictation hardware and software that are accurate and convenient enough to suggest that it will not be many more years before conservators can dictate treatment reports at the bench and have them converted to machine-readable text in real time. When that is achieved, computer-based documentation systems will cease being mere record storehouses and will begin, at last, to facilitate the creation of richly detailed examination and treatment records.
Walter Henry, a conservator by training, is acting head of media preservation at Stanford University Libraries. Since 1987 he has served as moderator of Conservation DistList, an electronic forum on the conservation of library, archive, and museum materials. He also continues to administer Conservation OnLine. He has been an associate editor of Journal of the American Institute for Conservation since 1996.
The author wishes to acknowledge the following sources in the preparation of this article: "Creating Knowledge Environments," Information about Information Service Briefing, vol. 1, no. 9, July 31, 1998, Outsell, Inc., quoted in "Leveraging the Intranet in Knowledge Management," by Mary Lee Kennedy, Director, Information Services, Microsoft Corporation; "Photography Conservation Training Via Videoconference: A Project Report [abstract]," by Irene Brückle and Paul Messier, Electronic Media Group, AIC Annual Meeting, St. Louis, June 11, 1999 url:http://aic.stanford.edu/conspec/emg/st_louis_meeting.html
Resources on the Internet
Many resources relating to conservation can be found on the Internet. The National Center for Preservation Technology and Training (NCPTT) maintains a comprehensive electronic clearinghouse of preservation Internet resources on its own Web site, which provides information on other Web sites, list serves, usenet groups, and additional resources related to the field.
To search the NCPTT clearinghouse, go to: http://www.ncptt.nps.gov/