The digital future: a look ahead: information management professionals will find new challenges, strategies, and approaches in store with digital preservation. (Tech Trends).

At the Core

This article:

* Examines the `three critical aspects of digital preservation

* Discusses the various approaches to preserving information integrity over time

* Looks at three strategies for a short- or long-term implementation plan

Organizations will increasingly create and receive information in digital form in the 21st century. Indeed, digital preservation will be one of the major challenges facing information managers in the coming years. Information professionals of all experience levels will be expected to steer their organizations through the dangerous digital shoals ahead.

Meeting the digital preservation challenge will involve understanding and addressing issues in three areas: the preservation challenge, preservation approaches, and implementation strategies.

The Preservation Challenge

Part of the challenge with digital preservation is recognizing the full breadth of the tasks facing information managers. In his white paper "Long-term Intellectual Preservation," Peter S. Graham defined three critical aspects of digital preservation:

* Medium preservation: preserving the physical media on which the electronic records reside

* Technology preservation: updating technologies from old to new as they become available

* Intellectual preservation: ensuring the integrity and authenticity of the information as originally recorded. This involves three kinds of change: accidental change, intentional change that is well meant, and intentional change that is not well meant (fraud).

Intellectual preservation is, perhaps, the most unfamiliar area. With information on paper, the authenticity and the integrity of the information, once established at the time of transfer, is seldom modified in the course of preservation custody. Digital records, however, require much greater diligence on this score.

According to Jeff Rothenberg in Avoiding Technological Quicksand: Finding a Viable Technical Foundation for Digital Preservation, digital documents require more diligence because they possess a "unique collection of core digital attributes" that must be retained. These attributes include the ability to be copied perfectly, be accessed without geographic constraint, be disseminated at virtually no incremental cost, and remain machine-readable in all phases of their creation and distribution. In addition, documents that are "born digital" tend to be dynamic, hyperlinked, and interactive -- additional attributes that may need to be preserved.

Preserving the intellectual integrity, in fact, may prove to be more difficult than preserving the physical components. This is because in the digital world, there are five factors that determine information integrity:

* Content: the intellectual substance found in the information objects

* Fixity: The content must be fixed as a discrete object in order to be a record. If a digital object is subject to change without notice, then its integrity may be compromised.

* Reference: For a digital object to maintain its integrity, one must be able to locate it definitively and reliably among other objects over time.

* Provenance: The integrity of an information object is partly embedded in tracing its source.

* Context: Digital objects interact with other elements in the wider digital environment.

Preservation Approaches

Over the last few years, information management professionals have developed a number of approaches to preserving information integrity over time: analog storage, policy formation, standard formats, computer museums, conversion and migration, emulation, and persistent digital archives.

As organizations continue to transition to digital records, they often opt for analog storage: printing a "record copy" on paper or microfilm and entering this copy into an existing recordkeeping system. Such analog records tend to lack key metadata information unless an organization plans the reformatting. For example, just printing the body of an e-mail message is not sufficient; an organization also must preserve transmission and receipt data. Furthermore, preserving records in analog form does not always preserve the core digital attributes previously noted.

Policy formation is a second approach to digital preservation. This involves using external regulatory bodies or internal policies and procedures to establish requirements for electronic recordkeeping. Once the policies are in place, however, either the information manager or another person, such as the internal auditor, must monitor compliance. Having adequate policies is usually necessary, but not sufficient -- it is the foundation for digital preservation, but it is not the entire solution.

Requiring the use of standard formats is another organizational approach. The goal is to avoid reliance on proprietary formats tied to individual vendors. Such proprietary formats always are at risk of abandonment by the vendors, thereby leaving the organization's records stranded. The problem with standard formats, however, is that they seldom contain the latest functions and components so desirable in the digital world. For example, the lowest common denominator standard format for text files, ASCII, does not retain such features as italics and bold-face. The result is a tradeoff: If a standard format is used, it must preserve all of the core digital attributes.

Computer museums have also been suggested as a preservation approach, possibly as a last resort. In this scenario, one or more institutions would, over time, gather and maintain the various versions of computer hardware and software. These institutions would need to maintain sufficient spare parts, manuals, and documentation for the indefinite future. Computer museums raise a number of questions: Can any machine be kept running indefinitely? How many people in the future will have access to the resources of, perhaps, a very small number of computer museums? Doesn't this defeat one of the major advantages of digital technology -- distributed access?

Conversion and migration in some ways are the opposite of the computer museum concept. The goal of conversion and migration is to keep the digital file readable as hardware and software change. Conversion, the less disruptive of the two approaches, involves the regular "rewriting" of digital files from older to newer computing environments. Anyone who has experienced a conversion, however, such as moving from WordPerfect to Microsoft Word, knows that this is not a seamless process. Migration, on the other hand, is usually a last-gasp attempt to move digital information from long-inactive legacy computer systems. Migration almost always requires the writing of special software that could call into question the intellectual integrity of the resulting records.

Emulation is the subject of current research by information professionals. Emulation means that one computer system will act like another. This is regularly experienced when personal computers emulate, or act like "dumb terminals" into a mainframe system. According to Rothenberg, the emulation approach involves developing techniques for specifying emulators that will run on unknown future computers, saving in human-readable form the descriptive data needed to recreate digital documents, and encapsulating documents along with their descriptive data, software, and emulator specifications.

Persistent Digital Archives is the name given to a research project funded by the National Archives and Records Administration (NARA). Working with the San Diego Supercomputer Center, NARA is trying to develop an operational system that captures, maintains, and provides access to electronic records regardless of changes to the computing environment. The goal is to "wrap" into one package the record's contents with enough descriptive data and contextual information to permit its reading on another -- but future -- computer system.

Implementation Strategies

These and other preservation solutions will continue to be developed and refined. In the meantime, information management professionals will need to make some decisions about implementation strategies.

The following three strategies will need to be part of any short- or long-term implementation plan:

* Grand solution versus trench warfare. Will information managers work from the "top down" trying to establish universal rules, or will they work from the "bottom up" applying lessons learned from one case to another? The top-down, deductive approach is exemplified by the InterPARES Project (International Research on Permanent Authentic Records in Electronic Systems). The bottom-up inductive approach is being tested in such places as Indiana University.

* Building recordkeeping functionality into all systems versus transferring records into a recordkeeping system. Once it is acknowledged that most information systems are not automatically designed to capture and maintain records, there is a choice to face: Do information management professionals try to make every information system a recordkeeping system in addition to its other information functions, or do they have a separate recordkeeping system attached to their computing infrastructure? Early efforts to implement the University of Pittsburgh Functional Requirements tended to focus on the former. Some recent initiatives have turned to the latter, particularly by adding a recordkeeping module to the portals that seem to be springing up like weeds in the Internet pastures.

* In-house preservation versus outsourced solution. Some information managers will want to build their own preservation infrastructure while others will contract for these services with outside vendors. The decision will involve more than economics; it must also involve organizational dynamics and institutional priorities.

There is no shortage of issues with digital preservation. In fact, the range of choices that are faced and the options to be considered sometimes lead to an avoidance reaction: Information management professionals hope to ignore digital preservation until they retire and then leave the problem to their successor. A dentist, who has been facing patients with similar avoidance reactions for years, came up with this saying: "Ignore your teeth -- they'll go away!" Unfortunately, the same thing will happen to valuable digital information unless information management professionals start acting now, even on a small scale. Information managers owe it to their institutions to confront head-on what may prove the most difficult professional challenge over the next few decades.

Editor's Note: In the March/April 2002 issue of The Information Management Journal, there will be an article on emulation by Jeff Rothenburg.

Read More About It

"Creative Archiving at Michigan and Leeds: Emulating the Old on the New (CAMiLEON)" at www.si.umich.edu/CAMILEON/top.htm. (provides information about a research project for testing the feasibility and effectiveness of emulation as a preservation strategy)

Dollar, Charles M. Authentic Electronic Records: Strategies for Long-term Access. Chicago: Cohasset Associates, 1999, 26-32, 59-72.

Duranti, Luciana, Terry Eastwood, and Heather MacNeil. "The Preservation of the Integrity of Electronic Records," 1997. Available at www.interpares.org/UBCProject/intro.htm (provides details of a research project by the University of British Columbia to identify and define the requirements for creating, handling, and preserving reliable and authentic electronic records)

Garrett, John and Donald Waters. Preserving Digital Information: Report of the Task Force on Archiving Digital Information. Washington, DC: Commission on Preservation and Access and the Research Libraries Group, 1996, 11-19.

Graham, Peter S., "Long-term Intellectual Preservation," Nancy E. Elkington, ed. Digital Imaging Technology for Preservation: Proceedings from an RLG Symposium Held March 17 and 18, 199,t, Cornell University, Ithaca, New York. Mountain View, CA: The Research Libraries Group, 1994, 41-44.

Hunter, Gregory S. Preserving Digital Information. New York: Neal-Schuman, 2000.

Indiana University at www.indiana.edu/~libarche/index.html (provides information about Indiana University's bottom-up, inductive approach to preservation)

International Research on Permanent Authentic Records in Electronic Systems (interprets) at www.interpares.org (information about a project to develop the theoretical and methodological knowledge required for the permanent preservation of authentic records created in electronic systems)

"Methodologies for Preservation and Access of Software-dependent Electronic Records" available at www.sdsc.edu/NHPRC/main.html (provides information about the National Historical Publications and Records Commission project to conduct research on the long-term preservation of and access to software-dependent data objects and to develop prototypes to preserve and provide access to electronic records over the long-term)

Rothenberg, Jeff. Avoiding Technological Quicksand: Finding a Viable Technical Foundation for Digital Preservation. Washington, DC: Council on Library and Information Resources, 1999, 3, 16, 30.

Gregory S. Hunter, Ph.D., is an Associate Professor in the Palmer School of Library and Information Science at Long Island University. He holds a Ph.D. in American History from New York University and is the author of Preserving Digital Information (Neal-Schuman, 2000). He can be contacted at [e-mail].