This paper examines the concept of metadata — commonly defined as "data about data" — through the frameworks offered by Anne J. Gilliland-Swetland's "Setting the Stage" and Paul Conway's "Preservation in the Digital World." The paper explores how metadata describes the content, context, and structure of information objects, drawing on examples from libraries, museums, and archival institutions. It then connects metadata to preservation theory, discussing Conway's four pillars of preservation: custody, social value, structure, and cooperation. Finally, the paper presents a dual perspective: metadata as an object requiring preservation, and preservation itself as a subject that generates its own metadata.
Metadata basically means "data about data," but the concept itself — along with all its implications and the domains it applies to — makes this explanation too simplistic. A clearer picture begins to emerge when we add that the term is mainly used when discussing information systems and resources, and that it describes "the sum total of what one can say about any information object at any level of aggregation."[1]
As metadata is understood across disciplines, its functions extend well beyond simple labeling. The following sections explore how two foundational essays — Anne J. Gilliland-Swetland's "Setting the Stage" and Paul Conway's "Preservation in the Digital World" — illuminate both the nature of metadata and its relationship to the broader challenge of information preservation.
As Anne J. Gilliland-Swetland writes in her essay "Setting the Stage," metadata can be used to describe the three main features of an information object: content (intrinsic to the object), context (extrinsic to the object), and structure. Consider a library as an example. Even if all information objects possess these three features, many are more strongly characterized by just one or two of them. In the case of a library, the context dimension is relatively unimportant, because it does not have a direct impact on or correlation with the library's purpose. What matters is what the library can provide — its content — and this is precisely where metadata is most useful.
The content of a library encompasses all its books, articles, references, encyclopedias, and other materials — in other words, all the components of the library's content as an information object. Metadata was developed to take the form of indexes, abstracts, and catalog records in order to provide both intellectual and physical access to this content. These are all forms of metadata used to describe content.
When referring to context, the example of museum and archival activities is most instructive. Here, metadata takes the form of accession records and catalog records, following standards that have been developed and refined over time. These examples of how metadata describes both content and context — and, further, how it describes an information object's structure — allow us to better appreciate the definition of metadata as "data about data" and to identify one of its primary characteristics: description.
Metadata describes, following certain rules and algorithms, a series of other data and information. As seen in the library example, a large collection of data and information is organized and made accessible through tools such as catalogs and indexes. This leads to the second major purpose and characteristic of metadata: organization and preservation.
Preservation, as Paul Conway points out in his essay "Preservation in the Digital World," has recently become a "coordinated, conscious management" — a management of data and information. This dimension of metadata deserves careful attention. While Gilliland-Swetland's essay describes several characteristics and classifications of metadata, Conway's essay focuses on the imperative of metadata preservation.
Notably, the durability of recording media has actually declined throughout history. We can still read Egyptian papyrus and Sumerian clay tablets today, yet magnetic tape — an invention of the twentieth century — may become unreadable within as few as thirty days of its production. This paradox underscores the urgency of deliberate preservation strategies.
Conway defines preservation as "acquisition, organization, and distribution of resources to prevent further deterioration or renew the usability of selected groups of materials." While he refers to preservation in its broadest sense, this definition can be directly tied to the concept of metadata preservation — that is, to the preservation of information about information. When we consider the preservation of data about data, we arrive at the insight that there is no fundamental difference between preserving data and preserving metadata. This is why, while Gilliland-Swetland's article addresses the definition of metadata, its operational context, and its classifications, Conway's article addresses how this data can be preserved.
In Conway's view, everything revolves around resource allocation: "people, money, and materials must be acquired, organized, and put to work to ensure that information sources are given adequate protection." He identifies four concepts that define the context in which preservation exists: custody, social value, structure, and cooperation.
"Custody, social value, structure, and cooperation examined"
"Dual relationship between preservation and metadata"
These two articles have therefore been related in two distinct ways. First, metadata can be regarded as another type of data, subject to the same preservation imperatives discussed by Conway. Second, as Gilliland-Swetland's taxonomy makes clear, we can also discuss the metadata of preservation. In this second sense, preservation implies a whole set of rules, a cooperation between institutions, a management of the technological devices on which data is stored, and a comprehensive system of information management and resource allocation — all of which are described in documents that function as metadata for preservation.
These documents record what measures have been taken to properly preserve books or digital files, how different libraries have cooperated to ensure preservation, how resources were managed and allocated, and so on. In one sense, we have the preservation of metadata, with all the elements of preservation that this entails; in another, we have the metadata of preservation — that is, all the documents and records that describe and govern preservation activity itself.
This dual perspective, drawn from reading Gilliland-Swetland and Conway together, reveals that metadata and preservation are not simply parallel concerns but are deeply intertwined. Each informs and depends upon the other, and a full understanding of either concept requires engaging seriously with both. As digital preservation continues to evolve as a discipline, the role of metadata in documenting, organizing, and sustaining information resources will only grow in importance.
You’re 65% through this paper. Sign up to read the remaining 2 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.