The NII Agenda for Action identified the following five basic components of a national information infrastructure (NII Agenda for Action, September 15, 1993, p. 5):
- The physical facilities used to transmit, process, display and store data (voice, text, images).
- The information itself, in the form of scientific, scholarly or business databases, video programming, images, sound recordings, library archives and other media.
- The software programs (also known as applications) that allows users to access, manipulate, organize and digest proliferating masses of information.
- The network standards and transmission codes that allow networks to connect with each other, and that also ensure reliability, user privacy and the security of information.
- The people who create the information, develop applications and services, construct facilities and train others.
These components, or requirements, apply to all participants in the information infrastructure, whether in the sciences or the arts and humanities, whether in manufacturing, health care or electronic commerce. The section that follows assesses the current progress made by humanities and arts computing in these five areas (making allowances for the great disparities that currently exist among institutions, disciplines and individuals in these fields).
University- and college-based programs in the humanities and arts have invested substantially in acquiring and installing all kinds of computer equipment. Faculty offices usually have a desktop workstation as virtually standard equipment (though not necessarily connected to the Internet); students typically have access to computer labs and computerized library catalogs, perhaps from their dormitory rooms. Together, these university-based investments in equipment have laid a basic, if low-powered, foundation of facilities for an information infrastructure serving the humanities and the arts.
Nevertheless, the promise of universal access is far from a reality. While the investment in physical facilities has been significant, many inequities exist, both between campuses and within any given campus. These gaps are likely to continue, if not worsen, as technological improvements in the quality and multimedia capabilities of equipment make access to more than basic facilities necessary. Now, as technological advances improve capacity, expand applications and link computers to other forms of communication, universities and colleges will need to continue to upgrade their physical facilities. Indeed, the acquisition, maintenance and upgrading of physical computing facilities will be particularly crucial for the arts and humanities, where the complex form of such information requires sophisticated equipment and technical improvements in visual, audio and text representation.
In addition, what has been missing until very recently is the consciousness of the paramount importance of interconnectivity: each college, university and university system has proceeded independently in acquiring its computer facilities and equipment, which were operated primarily on a stand-alone basis. Now, in the age of networks, institutions of higher education face a set of “last mile” or even “last foot” interconnectivity problems: that is, while a campus, or a library, or a single department may have the necessary equipment or network connection, not all faculty, administrators or students will necessarily have full access to these resources. In an era of constrained budgets, many institutions of higher education will be hard pressed to find the financial resources to meet these evolving needs.
Cultural institutions outside higher education often lack the most basic computer facilities. By contrast, university-affiliated museums, libraries, arts centers, and archives can take advantage of their institution’s investments in information facilities and access to networks. For example, according to the Museum Computer Network (MCN), it is not uncommon for museums to regard non-administrative computing facilities as unnecessary unless funded by outside grants or required to meet a specific project or legal need. As a consequence, many museums have not established institutional computing facilities (SPECTRA, Vol. 21, No. 4). While exhibition planners could use interactive multimedia tools to reach and engage more museum visitors, and provide better resource materials for scholarship and classroom use, the necessary equipment is likely to be beyond the means of most such institutions.
Even for fully automated cultural institutions, interconnection to the Internet may be unavailable or costly if acquired through commercial service providers. Indeed, such large museums as the Metropolitan Museum of Art and the National Gallery of Art in Washington have no access to Internet services on a gallery-wide basis (SPECTRA, Vol. 21, No. 4). This situation is probably true of virtually all of the 15,000 museums, historical societies and archives in the United States as well.
The same deficiencies characterize many local arts and performing arts organizations, relatively few of which are fully computerized, and some of which lack even basic computer equipment. Communication networks in the arts are still uncommon, though participation in services such as Arts Wire have risen dramatically in the past few years. Certainly many artists, particularly those who are not institutionally affiliated, do not own or have access to networked personal computers, and are thus excluded from a medium that they might find has intense creative potential for them
While electronic networks are certainly communication-rich, at present they are relatively content-poor, especially for researchers in the humanities and arts: an enormous amount of work remains to be done to convert the riches of our arts and cultural heritage information to electronic form. Without a critical mass of information, technological capacity is a hollow structure, like a library without books.
In the sciences, the newest information is the most valuable; historical information plays a distinctly secondary role to current documentation. In the arts and humanities, ancient archival materials are as valuable as modern. Materials accumulated over centuries — manuscripts, texts, plays, maps, dance notation, sound and video recordings, drawings, paintings, sculpture, and artifacts of all kinds, as well as catalogs of all these materials — are awaiting transformation into digital form. Such conversions will be enormously costly, because they must be undertaken at the highest possible quality levels so that the expense of successive re-digitizing can be avoided as technology improves.
The Brittle Books initiative of the National Endowment for the Humanities aims to preserve and improve access to 3 million brittle books through a nationwide effort over the next 20 years. The value to scholars and others of such improved access would be even further enhanced were the content of this significant corpus made available across the global Internet. Other projects aiming to increase access to humanities and arts resources include bibliographic, indexing, and object registration databases of long standing, which have been online for several years, such as the MLA Bibliography, the Avery Index to Architectural Periodicals, and the National Museum of American Art’s Inventories of American Painting and Sculpture. Still others, such as the Art & Architecture Thesaurus (AAT) and the Library of Congress’ National Coordinated Cataloging Operation (NACO), standardize the vocabulary prevalent in humanities and arts information. If such standards efforts were extended and adequately funded, they could become the building blocks for national cultural heritage databases. They would provide the integrating terminology needed to enable hundreds of individual and institutional projects to combine in a fully accessible digital resource of popular interest and educational value.
While the list of such projects is long and varied — with generous support from NEH and NEA manifest in many cases — it is still a cacophony of individual efforts, unguided by any systematic plan. What is needed are the vision and the funds to convert these separate projects into broadly based national data sets in the humanities and arts. A few model projects that have established a coordinated data collection process include the following:
- Inter-University Consortium for Political and Social Research (ICPSR). Hosted at the University of Michigan, the ICPSR is supported by funds from the NSF and by over 300 American and Canadian universities as well as national memberships in Europe, Latin America, and elsewhere. It maintains an archive of more than 30,000 data sets in the social sciences derived from surveys, censuses and administrative records. Of these, 1,000 are available over the Internet, and others will be added as additional disk space and documentation is prepared.
- American and French Research in the Treasury on the French Language (ARTFL), a database containing the French national literary corpus the Trésor de la Langue Française with works from the 17th to the 20th century.
It is not that the federal government fails to recognize the value of national data collection efforts; substantial support is currently provided for the creation of national data sets in the sciences. In fiscal year 1995, for example, the federal government plans to spend $152 million on the Human Genome Project, the creation of a database of our biological heritage, and (including investment from the private sector) $231 million for the Global Climate Change and Biological Diversity Documentation initiatives. Over the next decade, the federal government will commit many billions of dollars to these and other scientific projects. It is not unreasonable to ask for a similar level of commitment toward building national data sets recording our cultural heritage.
Government support for national data sets in other countries might provide models for similar support in this country. Many cultural databases, such as The Network of European Reference Corpora (NERC); the Oxford Text Archive; EuropArt; and the Network of Art Research Computer Image Systems in Europe (NARCISSE) are being developed in Europe, evidently because Europeans have come to recognize the research value of cultural databases and are discovering that their cultural information can have economic value as well.
Digitizing the centuries of existing humanities and arts information will be a long, expensive process that will only be worthwhile if it is built with an eye toward anticipated use and accompanied by software applications and tools that make possible a new dimension of learning and experience.
Burgeoning numbers of users can, for example, already participate in dialogues carried on within listservs; circulate journal articles for peer review with ease; and collaborate on research projects with geographically separated colleagues. Examples of promising tools include the following:
- Mosaic, a browsing mechanism containing graphics and hypertext links (developed at the National Center for Supercomputer Applications at the University of Illinois at Champaign-Urbana), which allows a user to navigate the Internet with relative ease.
- Textual Analysis Computing Tools (TACT), a collection of programs for textual analysis which allow scholars to examine their materials in diverse ways.
At the same time, the field is wide open for the development of new tools geared to the needs of network users of cultural heritage information. Some examples of needed new tools include:
- Authoring tools suited to the production and exploration of content in the humanities and arts.
- Shareable libraries of software tools.
- More sophisticated navigational tools to replace today’s rudimentary ones.
Needs in this area are detailed further in the document of the Working Group on Technical Requirements, in the second section of this report.
Network Standards and Transmission Codes
Standards and transmission codes are the most daunting of all the challenges the developers of global information networks face. The sheer newness and immense size of electronic networks means that standards for many different functions had to be invented nearly overnight, among them encryption for data security; log-on permission; recordkeeping to permit billing for the use of copyrighted material; and network management. This will continue to be true of such standards as data structures, coordinated vocabularies and many more.
Since electronic networks are global communication structures, network standards are by definition problems of the whole. Although different groups of organizations (computer firms, utility companies, federal contractors) assume responsibility for developing discrete parts of the total set of standards/transmission protocols, all work must be informed by a keen awareness of what all other groups are doing if the network is to function at all.
Technical committees of the International Organization for Standardization (ISO) and, in this country, the Institute of Electrical and Electronics Engineers (IEEE) have responsibility for regulating the development of network standards and transmission codes. Because the approval process is cumbersome, companies may publish a new standard and begin using it as they wait for it to make its way through the process. Yet the imperatives of global communication are so great that no proprietary standard could survive if it ignored the interconnectivity requirements of the network as a whole.
Another characteristic of network standards and transmission codes is that standards are in constant evolution as technological advances make new applications possible. Twenty years ago, as the earliest network protocols were being fashioned for use on the ARPANET, it was not envisioned that computer networks would ever be capable of carrying real-time video and that therefore a new protocol (asynchronous transfer mode, or ATM) based on entirely new principles would be required for such transmission to occur satisfactorily.
The network standards discussed above are defined as information interchange standards. Data standards are other criteria which, though not peculiar to the networked environment, protect the long-term value of the data stored in electronic databases, and make the transitions from one form of hardware or software to another easier when technological improvements dictate changes in equipment.
Standards for the content of our cultural heritage must accommodate the special characteristics of humanities and arts information manifested in all media of expression. The global nature of the humanities and arts favors the definition of open standards arrived at by broad consensus of an international community, reflected in ongoing efforts such as:
- The Text Encoding Initiative, an international project sponsored by the Association for Computers and the Humanities, the Association for Computational Linguistics, and the Association for Literary and Linguistic Computing with support from the NEH, the University of Illinois at Chicago and Oxford University, which has created guidelines for Standard Generalized Markup Language (SGML)-based encoding and interchange of machine-readable texts.
- The Art Information Task Force (AITF), a group representing the concerns of scholars, museum professionals and information specialists that has established a comprehensive set of categories for the description of works of art. Their work has in part been funded by an NEH grant to the College Art Association.
- The Consortium for the Computer Interchange of Museum Information (CIMI), a not-for-profit and for-profit private sector consortium that is working with museums and information networks in the Unites States, Canada and Europe to define application protocols for the interchange of museum data. CIMI grew out of a NEH-funded project to enable museums to exchange data between systems.
- The Image and Information Standards Initiative, sponsored by the Getty Art History Information Program, which is identifying issues in imaging that require collective solutions and standard approaches. Among these issues are the vexing problems of developing guidelines for access to intellectual property and frameworks for project management for the ever-growing universe of digitization projects.
Many similar large-scale standards development efforts, and much education and assistance in the implementation of common methods, will be required before humanities and arts information will be widely available. Nevertheless, it is only through such content and application standards that the information so much desired for cultural enrichment can be made available and usable.
People create information, develop applications and services, train others to navigate available data resources, and are the ultimate contributors to and users of the electronic network. These crucial processes were first begun as individual scholars, librarians and artists became interested in demonstrating the potential of networked information to their communities. Now professional organizations have begun efforts not only to educate their members but also to represent their interests in policy debates. Professional associations in the humanities and the arts have an important role to play, by:
- Providing forums for discussing standards and priorities, as well as identifying items for policy action.
- Disseminating information (e.g., by producing directories), collecting statistics and evaluating current programs and practices.
- Articulating how improvements in electronic resources would serve the public.
- Sponsoring research into user behavior and the development of educational materials.
Just as professional associations have focused the efforts of individuals wishing to provide leadership in demonstrating the potential of networked information to their audiences, a center created to coordinate the efforts of a large number of professional associations could in turn provide an even greater leveraging effect.
Clearly, training in the use of the information highway will be a persistent concern. Not only will every American soon need basic training in electronic literacy, but specialized training and periodic refreshers will be needed as new resources and technologies appear. One model for such specialized training and software evaluation is currently provided by the Center for Electronic Texts in the Humanities (CETH) based at Rutgers and Princeton Universities. To say that such training will be the joint responsibility of schools and government, parents and teachers, business and the nonprofit sectors means that each segment of the economy will undoubtedly pursue its own agenda: employers will provide training in the hopes of increasing employee productivity; profit-making firms will educate potential users about the benefits of their electronic products. Coordination (or at least acknowledgment) of the aims of the various types of training available will ensure that training in the humanities and the arts is not neglected because these fields offer less obvious or immediate economic benefits.