First Monday

Studying special collections and the Web: An analysis of practice by Lorraine Normore

Abstract
Studying special collections and the Web: An analysis of practice by Lorraine Normore

Many digital library collections are the virtual analogs of special collections in libraries, museums, historical societies and archives today. A field study of people responsible for collection maintenance across a variety of institutions was carried out. It aimed at improving our understanding of issues involved in collection description and access. A second study examined the current state of Web access to materials from the previously studied special collections. Data concerning the availability of online finding aids, externally accessible databases for collection content, digitized images and Web exhibits are presented.

Contents

Introduction
What was done
What was learned
Current state of access and digitized content
Implications for the future

 


 

++++++++++

Introduction

Digitization projects create the potential of providing all Web users with a rich variety of high quality primary source materials and with new ways to explore these data. As a result of these efforts, people everywhere can explore life in antebellum America ( http://jefferson.village.virginia.edu/vshadow2/) or see artefacts from classical Greece and Rome (http://www.perseus.tufts.edu/). They can go on field trips with geologists ( http://www.uh.edu/~jbutler/anon/anontrips.html) or archaeologists ( http://catal.arch.cam.ac.uk/catal/catal.html). They can get a view of the solar system (http://space.jpl.nasa.gov/) or explore the U.S. Botanic Garden ( http://www.washingtonpost.com/wp-srv/photo/botanic/frames.htm). The availability of such rich source material affords both opportunity and challenge to their creators. Once you have the digitally encoded content, how are you going to make it accessible? How do you present the material? How should you describe it to enhance its accessibility?

Much of the content that we see in these kinds of Web sites is contained in what are called "special collections" in the traditional library world. Special collections exist not only in libraries, but also in museums, archives and historical societies. To an increasing extent, special collections are providing a Web presence for their institution and content. This phenomenon is seen most strikingly in highly visible digital library projects like the Colorado Digitization Project (Bishoff, 2000). This Project has brought together special collections and unique resources from a wide variety of cultural heritage organizations (archives, historical societies, libraries and museums) throughout the state of Colorado. As Bishoff points out, a key aspect of the experience of consortial projects like this has been a developing understanding of the diverse needs of the associated user communities.

Over the past few years, an OCLC project has been learning about the needs of the holders of special collections, many of which have a Web presence of some sort. The project investigated the organizational, social, and economic characteristics of special collections that are located within a variety of institutions. We wanted to see if there were commonalities across the collections in terms of functional activities, roles and role responsibilities, and organizational values and "drivers". We were interested in learning about how the collections came to be, what kind of metadata they tried to provide for their objects, and how they tried to provide access to their content. We were also interested in finding out just how far along the road to universal access the collections were. To explore this issue, a second data gathering effort was undertaken. The Web sites corresponding to the collections studied were visited and their informational characteristics recorded. The results of the two studies constitute the body of this report.

 

++++++++++

What was done

The initial project aimed at learning about special collections by interviewing people involved in the care and maintenance of collections across a broad range of institutional types. By speaking directly with people in their own institutional environment, we sought insight into what was done rather than what "should" be done or what would be done in the future. In short, we wanted to find out what people were really doing and hoped to get insight into the reasons behind the choices that were being made.

Data collection

A project team was headed by members from OCLC’s Office of Research and included other OCLC staff members with a range of content expertise, from metadata through system development. Contextual design, a field study method developed and widely taught in the human-computer interaction community by Hugh Beyer and Karen Holtblatt (1998), was used to gather and bring together the data.

People involved in the maintenance of 24 special collections were interviewed. These collections were situated in a number of different types of institutions across the United States. The group included collections and archives associated with university libraries, in state libraries and archives, and in museums and in historical societies. The collections were housed in both private and public institutions. The scope ranged from national, through state and regional, down to the local level. The collections included highly diverse types of materials, from personal papers, printed documents, government records, to images (photographs, art works, prints, maps), audio materials and realia (e.g., archaeological dig materials, models, memorabilia, medals).

Unlike a formal structured or semi-structured interview that uses a set of pre-defined questions, contextual design provides a focus for the interview but allows interviewees to structure their comments using the vocabulary and categories natural to them. The method stresses that the interviewee is the local expert, the interviewer merely a lowly apprentice. In this case, interviewees were told that we wanted to learn all about their collection — what types of materials were in the collection, where they came from, what kinds of records were kept for the materials in the collection, who was involved in taking care of the collection, what was the job of the person we talk with, who used the collection, and whatever other topics the interviewee suggested. A typical interview took place in the office of the interviewee, with tours of the site, and lasted between one and two hours. Detailed notes were kept of the conversations, with audio-taped records kept when appropriate.

Data analysis

The information contained in the interviews was extracted by team members to form three separate data sources: (1) interview notes: The notes gathered in the interview are transformed into a series of single concept statements, typically 50-100 statements per interview; (2) a communication model: The people, institutions, and items mentioned in the interview are represented in a bubble diagram with functions annotated and with links between the items represented by lines between the bubbles; and, (3) a culture model: Statements about beliefs, values, and attitudes, are expressed as "messages" between a designated source and receiver.

Upon completion of the interviews and associated data extraction, we entered the consolidation phase of the contextual design process. Consolidation was particularly interesting to us in this project. Model data consolidate if there is commonality in the examined process. Given that we interviewed people from a range of institutional types, it was of interest to see if the data could be described by a single set of abstract properties or if the institutional differences would force the data into a series of separate consolidated categories. In this case, the data from various institutions did, in fact, come together [ 1].

 

++++++++++

What was learned

The goal of this part of the paper is to provide readers with insight into the culture of the special collection. It aims to go beyond the description of technological issues, to promote the development of a shared understanding of the social forces that cause these collections to come to be, that motivate their operation and that underlie their mission. While much can be learned through a detailed inspection of data [ 2], people often understand best from a story. Accordingly, we will proceed by presenting was we learned in a narrative form.

In the beginning...

At some point, a collection is begun because someone decided to keep some of the things that they had created or acquired. They might keep things they thought would be of interest or things that were inherently beautiful or valuable or that provided insight about people, events, or things that were, in turn, of interest. The objects (including papers) might simply have been created in the course of an ordinary life and have come together around a theme or around a common place of origin. This coherent central theme, often referred to as the "fonds" of the collection, is a crucially important feature of a special collection.

At some later time, the collection creators or their heirs decide that they want to involve an institution in the care of the content. These donors seek institutions that will protect and preserve their content. They may give all of the collection or only selected parts to the receiving institution. If they have rights to the intellectual property in the collection, they may or may not choose to give those to the institution. These or other donors may give money or property to support, preserve, and maintain the collection. In exchange, donors often want the tax advantages that accrue when objects are donated to cultural institutions. Many donors seek the prestige that having their content in highly regarded institutions bestows on them. Thus, many donors will seek institutions that are well known, often of national stature. Sometimes, the process goes in the opposite direction. An institution’s curators [ 3] or development officers may seek collections. Rather than waiting to be approached, they may approach individuals known to collect items related to the focus of the receiving institution and "court" these potential donors in the hope that their materials will be given to the institution at some future time.

In these collections, then, we see features that speak of human purpose, of the bringing together of things that are reflective of interest and motive. We also see the need to preserve and protect the content and how that need results in the transformation of the collection from the possession of the individual person or organization to a new role as part of our communal cultural heritage.

In the institution

When either a collection or items to be added to an existing collection come to an institution, a registrar creates a record of the content and then passes the content along to the collection manager. These records are made to help ensure that information about the source of a collection and its ownership history (an important feature of its provenance) are maintained within the institution. It is the collection manager who coordinates the work that needs to be done to take care of the collection. The work to be done falls in three main areas: Content description, content preservation, and resource acquisition and management. All three contribute to the process of providing access to collection content.

Content description

Metadata serve as vehicles that allow descriptions of collection content to be made accessible to the world. The collection manager makes decisions about the kinds of metadata to be created. Three types of collection records were noted. The first type is the catalog record. Catalog records are most often created for the collection as a whole but may include additional records for significant sub-collections. These often use MARC format and standardized vocabularies (LCSH or the Art and Architecture Thesaurus (AAT), for example). Catalog records are most often produced by trained catalogers or archivists. Typically they are designed for inclusion in the catalog of the parent institution rather than as a descriptive tool within the collection itself.

The second type of collection record is known as a finding aid. Finding aids describe the totality of the content of the collection in a structured way, grouping items by common themes or history. The groupings may represent either or both the intellectual organization of the material and the physical arrangement of the collection. In some cases, all of the items in a grouping are themselves identified in the finding aid record. In other cases, only the groupings themselves are described. The creation of the finding aid demands content knowledge and is most often the work of a trained archivist. Copies, typically paper, reside in a public reference area. Often, the finding aid is the only public access point to the detailed content of the collection. Online versions may also appear on the collection’s Web site.

When collections are prepared for online access, a third kind of record may be made. In this case, item level records for collection content are created. Each collection is typically organized into an independent database. The format of the database as a whole, including the record format, is often designed by the collection manager or archivist. Thus the database often reflects the individual collection but does not necessarily employ data elements that come from a widely accepted standard set or that are applied in a standardized way. In this situation, record creation is often delegated to volunteers. These individuals are sometimes well-trained and long term although they may be only be involved with the collection for a short time. Volunteers may come from the technical community that the collection’s content represents — a feature of great value when research skills and knowledge are required. For some collections in which the content requires a high degree of technical knowledge (perhaps linguistic or content-specific), subject matter experts may be sought to create base level descriptive records for items in the collection. Volunteers’ work is often reviewed and regularized (especially with respect to the inclusion of formal subject vocabularies) by the archivist or manuscript cataloger. When these kinds of databases are created, they are often made available to the external world via Web sites associated with the collection.

These three kinds of data provide very different entry points to a collection’s content. The catalog record may provide the most general access to the world at large since it is often included in shared catalogs. However, the level of detail it provides is the most limited. The collection database and finding aid both provide access to detailed features of the collection content. Because both finding aid and database content live "below the surface" of the Web, they may not be easily discovered. The database gives the fullest and easiest access to item-level information. The finding aid has the capability of providing rich contextual information, situating item content in the intellectual framework of the collection as a whole.

Content preservation

If items important to our cultural record are to continue in existence, they must be cared for. In fact, the nexus of the relationship between donors and the receiving special collection is often content preservation: Donors give their collection so that the content will be preserved. The institution agrees, perhaps implicitly, to preserve and protect the content they accept. Collection managers, curators and archivists combine their knowledge and skills to decide which items need to be conserved and to identify preservation specialists who would be able to appropriately preserve physical objects where necessary. Because funds for preservation are not necessarily available within the parent institution, the collection managers, curators and archivists will often join with development officers to generate grant applications to support preservation activities. Institutions may reach out to related institutions to write joint proposals since collaborations are often looked on with favor by granting agencies.

In the minds of many, digitization is associated with preservation. Digitized images of manuscripts and fragile materials, in particular, create content surrogates that support greatly increased access to the materials. Large-scale digitization projects are not undertaken lightly. Cost is a major concern. In addition, the decision to created digitized surrogates is affected by the potential for loss or destruction of the original and by copyright and other ownership issues. Decisions about which objects within a collection to digitize are also often influenced by local interest in a given item or by its public value. Like preservation activities, digitization projects are most often funded by external granting agencies [ 4]. Thus, issues of access are fundamentally linked to the costs of preservation and of digitization.

Resource acquisition and management

A major part of the effort involved in caring for a special collection is spent acquiring resources. Unlike libraries, most special collections do not typically have a regular source of income to support needed acquisitions and technical processing. The collaboration of curators, archivists, collection managers and development officers is needed to identify funding opportunities (e.g., grants, potential donors) and to develop applications. Indeed, much of the need for outreach activities relates directly to the ability of the collection to acquire resources.

Outreach programs are also very important because they guide interested parties to the collection. Collection Web sites may provide access to scholars and researchers who otherwise might have to travel at great expense to learn about the content of the collection. Collection level records, limited to general information about the available content, may be insufficient to meet scholarly needs. Detailed information is very important for this group. However, this level of detail is only available in finding aids or through collection-specific databases.

Exhibits, both physical and virtual, show both parent institutions and potential donors the important content the collection has to offer as well as raising its visibility, within the organization and to the world at large. Developing informative and attractive presentations is a major activity for those who manage the collection and for its curators. Such presentations can even extend access to school children and members of the general public who otherwise would never learn about the existence of this special content. While both physical and virtual exhibits are costly to create and to maintain, they function as a major source of organizational visibility and may serve as an important tool for fund-raising for the collection.

Emergent issues

These data show that special collections across a range of institutions have common features, needs, and functions, that they share a common culture. The common issues we see concern funding, preservation of items in the collection and of digital surrogates for the items, metadata creation, and access. In an ideal world, the interdependence of the four issues might be synergistic. In the real world, resource limitations force restrictions on both preservation/digitization needs and on the creation of needed metadata. In turn, a lack of metadata inhibits access to digitized content. As has been noted, access itself is crucially important to support funding. Thus, the ideal of synergy must confront the reality of resource limitations.

 

++++++++++

Current state of access and digitized content

To better ascertain the current state of affairs of access within this community, we move now to an examination of the Web presence of our sample collections. While we noted common concerns and a shared culture within and across the collections studied, we also noticed differences in the level of technological usage. We wanted to be able to understand the extent to which the collections were able to provide online access to their materials. This will allow us to discuss the current status of the external Web presence of the institutions studied and to enable us to draw lessons about the present and implications for the future.

Data collection

To support this analysis, a data gathering exercise was conducted in March of 2003. The Web sites available for each of the collections described in our interviews were visited and notes made about the forms of online access available. It should be noted that the original study attempted to draw from a variety of institution types and did not aim at collecting a sample of significant size for any type. Thus, because the research method is essentially qualitative, it would not be appropriate to suggest that any given feature is characteristic of other institutions of the same type. The data collected relate to four features of the Web sites: Online finding aids, externally accessible databases and search capabilities, the presence of digitized objects, and the availability of Web "exhibits".

To qualify as an online finding aid, the Web object had to include some overall description of the collection as well as a detailed inventory of the collection’s content. These had to be broken down into categories that were both determined by the categorical structure of the material and related to the physical disposition (e.g., boxes, folders) used to store the collection’s contents. Links to digitized images of collection content were not required.

For this data compilation, the collection or institutional Web site was examined to see if there were externally accessible databases for the collection. We also looked to see if there was a search capability across collections or if search was limited to one collection at a time. The type of search that was available was also recorded. Collections could provide field specific searches or full-text (FT) searching either across the records or within specified fields (e.g., notes).

It could be argued that Web content is, by its very nature, digital. However, for the purposes of this data collection effort, the phrase "digitized images" refers to digital surrogates that depict real-world objects, whether the objects are textual by nature (e.g., a manuscript or hand-written letter) or whether it is a digitized copy of graphic material (e.g., pictures, photos, postcards) or even a three-dimensional virtual representation of realia.

The Web exhibit is among the most interesting and challenging features of the evolving Web presence of cultural heritage materials for a number of reasons. In the early days of the Web, information was presented primarily as static Web pages with textual content. Today we observe the emergence of new forms of user experience. As technology has advanced, there are a wide variety of modalities available for delivery across the Web. The availability of these additional forms of access provides a new challenge to the holder of the special collection. Rich multimedia presentations are entering the world of the special collection. These range from the simplicity of digitized voice to the virtual reality "tours" of both museums and of historical sites. We now find Web sites for special collections that have explanatory and interpretational features. The records and digitized materials that are the content of the special collection serve as the substrate that feed the Web exhibit and give it depth. In this data collection effort, we made qualitative judgments about the richness of the Web presence provided for each collection, noting the extent and complexity of the "exhibit" available.

Results

Overall impressions

The results of our observations are presented in Table 1. The table is organized by institutional type, ranging from academic archives through state-based special collections. Even a careful examination of the table reveals little systematic relationship among institution type and the provision of any particular form of information access. Better endowed institutions are somewhat more likely to offer more access — the least amount of online access is associated with a small public library’s collection, the greatest with a national institution. However, even this generalization is not true of other institutions of the same type in this very limited sample. It is safe to say that almost institutions provide some form of online access to selected parts of their collection or selected collections.

 

Table 1: Online access methods available in the collections studied.

Type of Institution
Online finding aids?
Externally accessible database?
Search?
Digitized objects?
Web "exhibit"?
academic archives
top level outlne
no
no
selected images
yes-moderate
academic special collection
some yes, most collection level records
yes
metadata; collection level records
selected collections
yes-moderate
academic special collection
no
yes
structured database records
selected items (8.6%)
yes-extensive
academic special collection
no
no
no
selected images
yes-limited
academic special collection
detailed, some in progress
yes
FT of records
many images
yes-extensive
academic special collection
for selected collections
selected items; also collection level records in institutional catalog
specific field
selected items
yes-moderate
historic site special collection
no
no catalog; browsable lists
no
selected items
yes-moderate
historical society special collection, regional
for selected collections
yes—both overall & specific collection types
fielded plus FT of notes field
selected collections; selected items
yes-extensive
historical society special collections, state
for selected collections
overall and for selected collections and collection groupings
fielded; database specific
selected collections; selected items in other collections
yes-extensive
museum collection
for selected collections
online search for books
field specific
selected items
yes-limited
museum collection
no
yes, for 6.5% of specimens
FT within fields
selected items
yes-moderate
national library special collection
yes-extensive
yes-extensive
yes-extensive
yes-extensive
yes-extensive
national museum
for selected collections
internal only; borwsable lists of collection level records
 
selected items
yes-moderate
national museum
pathfinders
yes
fielded; FT of selected fields
selected items
yes-extensive
private academic institution
no
no
no
selected scanned documents
yes-moderate
private special collection
no; includes manuscript collection level records
yes
collection level records
selected scanned documents
yes-moderate
public library special collection
no
no
no
selected items
yes-moderate
public library special collection
no
no
no
no
no
state archive
no
yes
FT of records; some field simplicity
no
yes-limited
state archive
for selected collections
yes
FT; field specific
selected collections
yes-moderate
state archive
for selected collections; in progress
overall; for selected collections (more depth)
fielded in collection; FT for selected collections
no
yes-extensive
state digital library
content outline
no
full-text (FT) within documents
some images, FT transcripts
yes-extensive
state special collection
no
yes
FT of records; some field specificity
Web site links
yes-moderate
state special collection
no
yes
field specific
selected items
yes-extensive

Online finding aids

In the interviews, we had noted the importance of the finding aid for both information discovery and for resource description. We knew that there are many paper-based finding aids and that not all of these had been made available online. However, we did not know how prevalent this metadata was online. In this data collection effort, the availability of online finding aids on the external Web site was noted. While fourteen of the twenty-four have begun to make some of their finding aids available online, none of the institutions had as yet completed the task of making all such information digitally accessible. From the interview data, it is clear that conversion is not a trivial matter. Many individual institutions have a large number of paper finding aids. Converting them all to digital form would be costly. It was noted that even electronic versions of older documents were often in formats that were no longer supported and so would themselves require conversion. While paper versions could conceivably be digitized, they would have to be converted to searchable text via OCR to enable searching.

Externally accessible databases and search

In the interviews, many institutions reported that they had created or were creating databases that catalog collection content. In most cases, each collection had its own database, often based on commercial products like Microsoft’s Access or FileMaker Pro. An inspection of Table 1 reveals that collection-level databases accessible to searchers outside the organization were more commonly provided within our sample than were online finding aids. Sixteen of the twenty-four provided online databases and search for at least some of their collections. Many of the collections provided fielded search capabilities although this was often supplemented by having full-text search available, especially for fields like the Notes which tends to contain free text rather than encoded data. Several institutions provided not only the ability to search within a collection but also across collections. This greatly enhances the ability of the user to find where information of interest is located since the retrieved items will usually indicate the collection they came from in addition to item-specific information. The existence of database search, paired with links to finding aid-like context, provides the richest structure for discovery and navigation of collection materials.

Digitized images

While it is textual data (finding aids, databases) that provide access paths to special collection content, the availability of digitized images depicting collection content provides not only information but adds strong visual appeal to the site. Digitized presentation is often necessary when the content is essentially graphic (pictures, photos, postcards, realia). However, even digitized versions of letters and other primary source documents provide a much more powerful impact than would a simple transcription of the text. They make the collection real to the user, whether scholar or school child.

It should not, therefore, be surprising that all but three of the collections we visited have digitized objects on their Web sites. In some cases, the complete set of items in a given collection have been scanned and are available for viewing. In most cases, however, only selected items in the collection have been digitized. Two features reported in the interviews account for this phenomenon. A whole collection is digitized most often as "a project" that may have been externally funded. Thus, selected collections may be digitized, more or less as a whole. Alternatively, institutions with digitization facilities may choose to digitize individual objects for specific purposes. These can be item-specific: a user has requested a copy of a particular item — e.g., a photograph — and in order to create that copy, a digital reproduction is created. The purpose can be more generic: An exhibit is planned to celebrate an occasion or to illustrate some institutional function. Objects are selected for inclusion in the exhibit and may be digitized either as part of a virtual exhibit or to provide documentation (e.g., a guide) for a physical exhibit. Rarely does a collection holder have the resources to digitize all available content. The result is that we see selected collections and selected objects on Web sites.

Web exhibits

There is trend toward the development of Web sites that provide more than a database or a finding aid to make a body of primary source material accessible. Sites like the Valley of the Shadow ( http://jefferson.village.virginia.edu/vshadow2/) weave a story around the data, allowing site visitors to move about, following their interests and get in-depth exposure to the material that lurk below the surface of the story. These sites attract patrons of museums and other institutions, in addition to the traditional special collection user, the scholar. Similar features are found in sites aimed at a diversity of users, from preschoolers through the entire lifespan. Some researchers have experimented with providing a variety of interfaces, based on user characteristics, to items drawn from the same body of materials. A very interesting example of this is the Smart Web Exhibits project that was the joint product of the Carnegie Museum of Natural History, the Carnegie Mellon University Libraries and the Carnegie Mellon University School of Computer Science. The project provided both a series of "packaged tours" to materials drawn from their dinosaur collections, one aimed at children, another at interested adults, while providing an interface for the researcher that allowed search of documents and photographs and links to related publications.

Within the scholarly community, similar Web sites signal the emergence of new forms of expression, using the Web as a communications medium. They provide innovative ways for scholars to share their materials and to collaborate. Rather than publishing papers in journals and accumulating data in isolated datasets that must be downloaded to be shared, new ways for scholars to create and share data are emerging. Chodorow (2000) cogently discusses the new publication opportunities that technological change bring to scholarly discourse, emphasizing the needed concomitant changes in the social structure supporting recognition and in the need for new ways to reference and preserve scholarly discourse. The emerging trend toward institutional repositories like MIT’s DSpace provide another indicator of change.

The importance of the Web exhibit can be seen even in our small sample. While a few of the collections have only limited Web presences — a page or two describing the collection, perhaps with links to a database or finding aid, many more have Web sites ranging in size and complexity. Table 2 reveals that twenty of the twenty-four sites that were visited provide at least a moderately complex Web site to describe their collections.

 

Table 2: Availability of online databases or finding aids by Web exhibit complexity.

Web Exhibit Type (n)
Online database/search?
Online finding aids?
extensive (9)
9
7
moderate (11)
6
6
limited, none (4)
2
1

Table 2 also suggests a relationship between the extent of the Web presence for a site and the availability of the substrate that is access to collection content suggested earlier in this paper. As we see, the more extensive Web sites are more likely to contain both online databases and finding aids than those with fewer Web pages. Later research could test this suggestion.

The kinds of Web sites discussed here will provide significant challenges to a number of communities. The site creators will need to expend a considerable amount of time and energy and to have significant design expertise to create and maintain complex, often multimedia rich, sites. Those who aim to preserve digital content will have to develop ways to preserve not only the information "container" — the Web object that is seen — but to incorporate the underlying data and the mechanism for serving that information into their preservation scheme. Metadata creators will be challenged to create accurate and descriptive metadata for information sources that are inherently dynamic.

 

++++++++++

Implications for the future

When you read stories of large digitization projects, you cannot fail to be impressed by their accomplishments. As you move from these funded, nicely bounded, projects, you are confronted by the magnitude of both the task and opportunity of increasing access to the materials in special collections. Indeed, whenever I try to describe the world of special collection, and especially collections on the Web, a refrain from the title track of an album by the Escape Club (1988), comes to mind. The refrain goes, "living in the wild wild west". There are definite parallels: Boundless riches and opportunity, centers of "civilization", lots of laws, variously applied. And like the "wild wild west", both the cavalry and the settlers are appearing on the horizon.

It is important to realize that what the community interested in this kind of special collection is interested in keeping is not just a set of items that have some commonality that binds them together. It is, in fact, our cultural heritage, the story of our past, reflected in the things that were made by natural or social forces. If that past is to be kept, metadata about the entities need to keep both information about the thing and about the physical and cultural context that it is derived from. Metadata provides both needed information and a way to support access to the material and to the framework of knowledge that it resides in.

As we’ve pointed out several times in this paper, the value of a collection’s content derives from access. If people don’t know the content is available, it won’t be used. If it’s not used, it won’t be valued. Yet, effective access demands both metadata creation and preservation/digitization. Both metadata creation and preservation/digitization are costly. If we are to move forward as the guardians of these resources, mechanisms need to be found to appropriately support the work of the collection holders and their volunteers. Funding agencies can help by both funding and demanding that projects they support create usable metadata and appropriate access mechanisms as an inextricable part of enhancing collection content through preservation and digitization.

But collection holders need help to do this.

While we have argued that there is a common culture within the institutions studied, we didn’t find a solid body of standards or tools that were widely used across institutions. The costs of not having standard procedures and easy to use tools are both duplication of effort and a failure of interoperability of access mechanisms across institutions. Funding agencies need to support the development of these tools and procedures and to demand that funded projects use them.

Finally, the task is too big for any single institution. Large-scale cooperative ventures stand the best chance for success, difficult though cooperation may be. Within the library community, the Research Libraries Group (RLG) and, more recently, OCLC through its Digital & Preservation Resources Division, have been working on efforts to support collaboration among the holders of special collections across institutional types. The settlers are definitely on the way. End of article

 

About the Author

Lorraine Normore was the project leader for the Special Collections Project. She was also a Consulting Research Scientist in the OCLC Office of Research. Her project work on interaction design focused on ways of studying user behavior and on drawing implications from those behaviors that enable the design of better information systems.

 

Acknowledgements

I’d like to thank the institutions that so graciously allowed us to interview members of their staff concerning their truly wonderful collections. I’d also like to acknowledge the invaluable contribution of members of the OCLC Special Collections Project team: Mark Bendig, Leah Houser, Pam Kircher, Ralph Levan, Todd Matola, Sandy McIntyre, Ed O’Neill, Malia Watson, and Susan Westberg. Without each and every one of them, it wouldn’t have been half as much fun.

 

Notes

1. The consolidated data from the notes form a hierarchy of topics organized under the headings "what", "where", "how" and "why". These can be seen in the project’s affinity wall Web page ( http://research.oclc.org/Normore/public/rd/2001/Web/affinity-web.htm).

2. The consolidated communication and cultures models will be described in detail in a future publication.

3. The named individuals (e.g., registrar, development officer) should be regarded as roles (a set of actions that go together) rather than as actual job titles within any of these institutions. Special collections and their parent institutions vary greatly in size and scope. In some cases, a single individual may take on all or most of the roles discussed. In some cases, the roles are dispersed over a number of people. What is consistent is the set of tasks or functions that are carried out.

4. The tie between granting agencies and the funding of digitization and preservation activities is stronger in the United States than in other countries, especially the European Union.

 

References

H. Beyer and K. Holtzblatt, 1998. Contextual Design: Defining Customer-Centered Systems. San Francisco; Morgan Kaufmann.

L. Bishoff, 2000. "Interoperability and Standards in a Museum Library Collaborative: The Colorado Digitization Project," First Monday, volume 5, issue 6 (June), at http://firstmonday.org/issues/issue5_6/bishoff/.

B.G. Callery, 2002. "Smart Web Exhibits: Carnegie Museum of Natural History," presented at Web-Wise 2002, Johns Hopkins University, at http://www.library.cmu.edu/Libraries/IMLS/CalleryFinalPresent.ppt, accessed 28 April 2003.

S. Chodorow, 2000. "Scholarship & Scholarly Communication in the Electronic Age," EDUCAUSE Review (January/February), pp. 86-92.

Escape Club, 1988. "Wild wild West," from the album Wild Wild West. EMI Music Publishing.


Editorial history

Paper received 13 May 2003; accepted 22 September 2003.


Contents Index

Copyright ©2003, First Monday

Copyright ©2003, Lorraine Normore

Studying special collections and the Web: An analysis of practice by Lorraine Normore
First Monday, volume 8, number 10 (October 2003),
URL: http://firstmonday.org/issues/issue8_10/normore/index.html