A New England library

Paper and Digital Preservation

9/27/2004
NELA Annual Conference
Manchester, New Hampshire

Steve Dalton, Director of Field Services, Northeast Document Conservation Center (Andover, Mass.), gave an enthusiastically received talk comparing and contrasting the preservation challenges presented by paper and digital library materials. He framed this "tale of two worlds" by noting that, while the "preservation is access" argument makes sense for books and related materials, in the digital arena, the dominant consideration is different. Here, access to the original physical object (if there is one) is not as important as is preservation of access to the digital files.

Preservation has again become a "hot topic," due largely to two recent events. The September 11 tragedy involved an enormous loss of cultural heritage, raising consciousness about the stakes involved in assembling and maintaining physical collections. The publication of Nicholson Baker's Double Fold, with its sharp criticism of mainstream preservation practices (and including a section on Steve Dalton himself), paradoxically did the preservation community a service by bringing fresh attention to critical issues. One consequence of this change in public awareness is that Simmons College library school has added a third section of its course on preservation management, in addition to a new section at Simmons' Mt. Holyoke College campus.

The basic fact to keep in mind when considered the "analog world" of paper preservation, is that all organic materials (the majority of materials in libraries) deteriorate over time. This is sometimes called an "inherent vice." The collision of molecules causes deterioration in an object. While this process constantly active in all organic substances, heat speeds up deterioration by increasing molecular activity. Mr. Dalton outlined four different types of deterioration:

  • Chemical deterioration occurs when materials are not chemically stable. The changes in acidic paper is probably the best-known example of this.
  • Biological deterioration involves changes from the actions of organisms. The effects of mold , which pose risks to people as well as collections, provide an example.
  • Mechanical deterioration involves changes in the structure of an artifact. For example, bindings may become broken or warped, or stressed through improper handling.
  • Incidental deterioration occurs literally as the result of an "incident," such as fire, a coffee spill, or vandalism.

A "quick history of papermaking" followed, providing a context for changes over time in the stability of paper. In the early days, the rags which formed the basis for paper were shredded by hand. The resulted in long fibers, which were intricately woven to create paper of great strength and longevity. Mechanical trouble began to be introduced by the Hollander Beater (c1690), a machine for shredding rags, which saved much labor but produced shorter fibers, resulting in a less intricate weave. A papermaking machine, invented ca. 1798, allowed for mass production of paper. At the same time, procedures for papermaking began to be compromised to allow for the machine's requirements. Another series of developments brought about increased acidity, and therefore chemical instability, in paper. The process of chlorine bleaching (c1774), left deposits of hydrochloric acid if the chlorine was not rinsed adequately. "In-the-vat sizing" (c1827) caused sizing to be worked through the paper, rather than only affecting the surface. The use of aluminum sulfate as sizing (c1830) created sulfuric acid, and the well-known use of ground wood pulp (c1860) created paper which was inherently acidic.

The magnitude of brittle paper is great. Nearly 80% of all books in research collections in the United States, printed since 1860, are in danger of becoming embrittled. Roughly one-third of all research collection materials have reached this state, with some items being no longer usable. A variety of factors, having largely to do with the storage environment, contribute to deterioration beyond the materials' inherent vice. These include high temperature and relative humidity, fluctuations in both, the action of light (particularly ultraviolet) and environment pollutants, which affect leather especially.

Mr. Dalton demonstrated the Preservation Calculator, available at no charge from the Image Permanence Institute (http://www.imagepermanenceinstitute.org/index.html). This tool calculates the Preservation Index, indicating the combined effect of temperature and relative humidity on the decay rate of organic materials. Besides the PI, the calculator shows potential for mold germination and the "natural aging rate" under different storage conditions. The NEDCC's recommendations for general collections are a maximum temperature of 70° Fahrenheit, with a 30-40% relative humidity range. For collections of rare materials, however, these numbers should be reduced. HVAC systems are ideal if feasible, and should run 24 hours per day, every day of the year, if the system effectively controls both temperature and relative humidity. Establishing a regimen of environmental monitoring is highly desirable, to document existing conditions and help build the case for further improvements.

Monitoring and improving environmental conditions should be one part of a larger plan, which should also include a preservation needs assessment and decisions about levels of care. One useful resource for conducting a needs assessment is Assessing Preservation Needs: A Self-Survey Guide, available for purchase at http://user823621.sf1000.registeredsite.com/selfsurvey/survman.htm. Mr. Dalton provided basic descriptions of three levels of collections care. Level 1 focuses on preventive strategies, such as general facilities maintenance and fire protection. These "best bang for the buck" strategies are not only the most important, but also the most economical. For example, mass deacidification of nonbrittle books presently costs $12-15 per unit. If this preventive step is not taken and microfilming is necessary, the cost could rise to $75 per book. Level 2 reacts to existing problems, similar to the use of medication for specific disorders. Reformatting and basic repairs come into play here. Level 3 resembles "radical surgery," and features intensive treatment of important and valuable objects by a professional conservator. In an active preservation environment, all three levels will be applied to different extents.

To begin developing a preservation regime, one may seek out training opportunities, visit appropriate websites and become familiar with the professional literature (see http://www.nedcc.org/p101cs/furtres.htm#sources). To proceed, Mr. Dalton recommended documenting deterioration in the collection, showing damaged material to other and sharing preservation information in "non-threatening ways" with colleagues. From there, preservation can be incorporated into mission statements and collection development policies, and established as a long-term challenge addressed by a formal program. In conclusion, although this part of the talk focused on paper materials, it is essential to gain knowledge about all organic materials in a library collection.

Steve Dalton then turned our attention to preservation of digital materials. This involves a long-term, permanent commitment, but one that libraries must face. As of 1999, approximately 93% of new information produced is created in digital form, and that percentage is probably higher now. Digital technology has had multiple impacts on libraries. Aside from some of the better-known impacts, such as the redefinition of research and changes in interlibrary loan, adoption of digital technology has enhanced interdisciplinary collaborations. Specifically, librarians, archivists and museum curators have developed increased motivations to collaborate on projects involving digital materials, learning something of each others' vocabularies and aspects of their disciplines.

One important, though narrowly focused, "slice of the digital pie" is the use of digitization for preservation of analog collections. This is an area that is constantly in flux, in terms of policies, techniques and best practices. One absolutely crucial consideration here is that, with digital technology, we can intervene before or during the creation of digital data to build a preservation framework into the data itself. This is not something one can do with analog materials.

It is important to distinguish digitization from reformatting practices, such as microfilming. With microfilm, the aim is to create images on a stable base that will last for hundreds of years. Digitization, by contrast, is primarily an access, not a preservation, tool. While the utility of original artifacts may be enhanced - for instance, one may actually "travel along the brushstroke" of a painting - digitization itself creates new artifacts, with their own preservation challenges, and inherent vices of fragility and technical obsolescence. Similarly, there are important differences between the motivations to preserve and to digitize. Preservation of "analog" cultural materials is important because of their inherent value, and to make possible their continued use. Digitization, by contrast, not only may enhance access, but serve as a source of revenue, allow for contribution to collaborative projects, and so on. Nevertheless, despite these many contrasts, digital artifacts do require preservation, and lessons from the analog world may be applied.

Mr. Dalton pointed us to a sample "selection policy," the Columbia University Libraries' Criteria for Digital Imaging ( http://www.columbia.edu/cu/libraries/digital/criteria.html). These criteria address three basic questions. "Should we?" refers to the potential need for digitization. "May we?" is the legal question, and "Can we?" refers to technical feasibility.

Preservation of digital data raises several questions having to do with overlapping levels of preservation and the different contexts in which data will be (re)presented. At the base level, we must preserve the content: the bits themselves, including metadata bits. At higher levels, we are preserving the form of a document (e.g. sets of bits which represent a page within a chapter) and possibly the style or appearance of the artifact. But even representation of the entire, integral artifact is insufficient, as we must also preserve the data's functionality, so that it will continue to be able to represent the artifact despite changes in hardware and software.

Digitization projects involve a number of considerations that will impact on preservation, depending on how they are addressed. Among these are the quality of the original data files, hazards with use of proprietary formats, provisions for security and plans for migration. The collaborative aspects of a digitization project include clear definition of rights and responsibilities, plans for a project's long-term sustainability, and the creation of good metadata.

Steve Dalton summed up these various considerations in a set of "digital preservation essentials":

  • Approach the task systematically;
  • Define how the program fits the institution's mission;
  • Develop specific goals, considering all stakeholders;
  • Establish an appropriate infrastructure;
  • Draw on existing and emerging resources for planning;
  • Consider that collaboration may be imperative;
  • Choose preservable file formats;
  • Provide for long-term storage and access;
  • Do preservation metadata well!

Reported by
David Miller
Curry College
Milton, Massachusetts
NETSL Writer/Editor