How Do We Show You What We've Got? Access to Archival Collections in the Digital Age

Kate Cruikshank

Political Papers Specialist

The Lilly Library

Indiana University

[email protected]

Caroline Daniels

Automated Systems Archivist

University Archives and Record Center

University of Louisville

[email protected]

Dennis Meissner

Archival Processing Manager

Minnesota Historical Society

[email protected]

Naomi L. Nelson

Coordinator for Research Services

Special Collections and Archives

Emory University,

[email protected]

Mark Shelstad

Associate Archivist/Information Manager

University of Wyoming

[email protected]

.01 Abstract

Computers have transformed the ways in which researchers gain access to archival material, starting with inclusion of collection descriptions in online catalogs, and moving to detailed online guides to the contents of archival collections.  With each new stage in this development has come the necessity of standardization, which has in turn opened the door to new possibilities, such as digitization and sharing of manuscripts, images, and audio-visual materials online.  Resource constraints require individual institutional choices as to what is possible and desirable. At the same time standardization has proved to be a powerful tool for increasing collaboration that will enable users to access ever more material ever more easily.  This paper traces the transformation in describing and presenting archival materials, and presents the state of the art in both description and digitization.  The paper poses questions as to whether current trends in archival description are maximizing the potential of the online environment to increase access for researchers, and whether both descriptive and digitization initiatives are really providing what users want and need.

.02 Introduction

The central problem for archivists, as for librarians, is bringing users together with the materials that they need and want.  Over the last two centuries librarians have taken the lead in developing tools to make that happen, beginning with classification systems such as the Cutter classification scheme, the Dewey Decimal System, and Library of Congress classification. Each resulted in books on the same subject being shelved together so that researchers could confidently browse the stacks for new materials.

The adoption of a few systems by many libraries was a first step toward setting common standards in the library world and worked smoothly for books that might be considered to be about only one subject.  Such books are rare, however, and classifying a book under one subject did not assist in its being discovered by an interested patron seeking books on other subjects to which it might relate.  The remedy to this problem was the creation of subject headings—words or phrases describing what the book was "about."  These were added to the basic catalog information so that a catalog record would not only have author, title, and publication information but also a list of subjects—people, places, historical periods, concepts prominent in the work—in which a researcher might have an interest.  In the days of the card catalog, the result was that for any given book there were cards filed under the author's last name, the title, and each of the subject headings, so that a researcher coming from any of those directions would find the book.  In order to make the system work, librarians had to decide which of the many ways to express a subject they would consider official or "authorized" (e.g., “moving pictures” versus “motion pictures” versus “films”) and then use that consistently. They also needed to create cross-references from all the other possibilities that they could imagine a normal person might expect to be able to use.  That was a second step toward setting common standards in the library world.

The creation of all these cards was very labor-intensive work, and it soon became obvious that there were thousands of librarians all over the country working with the same books to create essentially the same cards.  When the Library of Congress offered to make its printed cards available to other libraries, it revolutionized the work of librarians creating catalog records, and standardization across libraries became accepted practice.

In the archival world, things were very different.  Archival collections are completely unique—there is only one of each of them—and they are unified first and foremost by whomever or whatever created or generated them: a person, an organization, or a governmental body.  Within a given collection there can be material "about" a wide variety of subjects, and there can be boxes and boxes of it.  The traditional way of providing access to that material was to create finding aids. These could take a variety of forms, and depending on the time and energy of the processing archivist, could include:

• a narrative description of the collection
• a list of boxes with a general indication of contents
• a list of file folder labels
• a list of letters with details as to date, addressee, and subject, or
• an index of subjects covered in the collection, keyed to folder location.

Library practices simply did not work here, although occasional attempts were made to apply them.  In at least one situation, a librarian taking over a university archives decided to physically arrange all correspondence by subject, pulling letters from many collections and placing them in file folders with subject labels.  But such practice violates the two basic principles of archival work, namely that provenance is respected—meaning  that the papers or records of one creator are not mixed with those of another—and that where there is a discernible original order, that order will be maintained, being in itself of value to researchers.  Archival collections thus could not be subjected to a classification scheme that collocated materials in the way that books could.   Subject indexes were a possibility, as were descriptions of collections that brought subject content to the forefront.  These tools and others were available to researchers in archives reading rooms, as was the best tool of all, the archivist her/himself.  In reality, there was always a great deal more knowledge in the archivist's head than ever got into the finding aids, and the wise researcher knew to plumb the depths of that knowledge.

As archival practice developed, archivists shared knowledge and traditions emerged, but standardization was not considered a particular virtue because of the uniqueness of every collection.  Finding aids varied from institution to institution in both form and content, but there seemed to be no compelling reason to consider that a matter of concern as long as there was an archivist to assist the researcher with the idiosyncrasies of the tools at hand.  All of that changed, however, when computer technology offered librarians the possibility of freeing their card catalogs from their physical location and making information about their books available to users anywhere there was a computer. It quickly became clear that archival collections could be presented to potential users in the same way.  Both cost and labor considerations in the leap to online catalogs pointed to the advantages of institutions working together.  With that came more concern for establishing common standards and the need to determine how best to use computers to present the information as institutions wished users to receive it.

.03 Describing Archival Collections in Online Catalogs

Archivists first ventured into this computerized realm by creating MARC (MAchine Readable Cataloging) records for their materials. These records, which form a part of many archives’ resource discovery systems, are relatively brief descriptions that are available to researchers via an online public access catalog. Using fields that allow narrative description, as well as the "subject heading" type of access used to find other kinds of materials, archivists are able to use this technology to make the existence of relevant materials known to researchers. While their records necessarily share many characteristics with those describing other kinds of materials, they also differ significantly as a result of the uniqueness of archival materials.

Many catalogers who create records for archival collections add them to online catalogs that are shared with other libraries, whether locally (as in the case of a multi-library campus) or more broadly (for example, via a union catalog such as the Online Computer Library Center’s [OCLC] WorldCat). Thus, archival catalogers seek not only to create useful records of their materials, but also to create these records so that they will be compatible with those describing books, videos, maps, and other materials. To achieve this goal, archival catalogers share fundamental principles with catalogers of these other materials—creating records that must coexist meaningfully with these other records. Unlike other kinds of archival description, catalog records cannot take idiosyncratic forms, or adhere to entirely locally-generated formats. If researchers cannot rely on finding certain kinds of information in certain fields (for example, if there is no title field at all), then they cannot perform a meaningful and reliable search. If records followed vastly different rules, the online catalog would be chaotic and less useful to researchers.

There are adaptations of cataloging rules, however, that allow archivists to create more useful records for their collections, even though these materials differ significantly from most books. Most books have a more or less clear title, for example, but most collections of papers do not.  The primary source of rules for catalogers in general is Anglo-American Cataloging Rules, 2nd edition, Revised  (AACR2R).  While AACR2R provides guidelines for describing archival materials, these guidelines have been found by many to be less than satisfactory. [1]  For example, under AACR2R, titles that are supplied—not explicitly given in or on the work itself—must be placed in brackets. But if archival catalogers provide titles for nearly all of their materials, why not take it as a given and omit the brackets? [2]  As a community, archivists have developed standards related to AACR2R, but designed them to meet the needs of archival materials. These standards (first Archives, Personal Papers, and Manuscripts, or APPM, and now Describing Archives: a Content Standard, or DACS) allow archivists to bridge this apparent gap between the uniqueness of each collection and the need for standardization in a multi-format, multi-library context.

Within this standardized context, two components of archival cataloging are particularly important for describing collections and leading researchers to those collections: entries for subject headings and authors, and descriptive notes. Other elements, such as descriptions of the size of the collection in physical and chronological terms, and descriptions of how the collection is organized, are also helpful, but the first two means of conveying information are particularly important.

The first of these, entries for subject headings and authors, is the most standardized type of information. Archivists make entries for authors or creators of collections, whether they are people or organizations, and listing other individuals, organizations, and geographic entities that are covered in the collection. Interestingly, correspondents are frequently considered subjects rather than authors, which may affect the outcome of an “author browse.” Topical subject headings are phrases and words, or strings of phrases and words, that briefly describe a topic, and archivists, like other catalogers, use them to describe the subjects that are covered in the collection. These entries—whether people, places or ideas—should be tremendously useful for searching, especially once the researcher knows how a person, place or idea is described in the catalog.

This is where standardization becomes particularly important. For example, catalogers use standardized formats for names of people and places. European names are given last name first and cities in the U.S. are referred to by names followed by the state. Catalogers also use controlled vocabularies, which are essentially listings of specified words or phrases used to describe such specific ideas, people, or events. This means that a cataloger will consistently refer to general athletic shoes as “sneakers” rather than tennis shoes, and once an information seeker knows that, he or she can find all the materials in the catalog—primary and secondary—that deal to any degree with sneakers, tennis shoes, or trainers by doing a subject browse for that heading.  As previously noted, these standardized forms are considered the “authorized” forms. Archivists also share controlled vocabularies with other libraries, so that once a researcher knows the authorized version, she or he could, ideally, search any library catalog using that heading.

The Library of Congress deserves special mention here, as it provides the controlled vocabulary that researchers in U.S. repositories are most likely to encounter. [3]  Since its vocabulary is applied by so many different institutions, the Library of Congress makes it possible for researchers to use headings found in one repository's catalog to search in another. Thus, in a collective database such as OCLC’s WorldCat, a user can search records created by a large number of archives and libraries from across the country and overseas.

Of course, this hinges on the researcher being able to identify the “authorized” version. As noted above, in many cases, catalogs contain references from common variations to the authorized form. Researchers browsing the subject “tennis shoes” are referred to the heading “sneakers.” In an online environment one should be able to click on a link to search under this heading.

While the forms of subject headings that are used in archival cataloging are identical to the forms used in cataloging other materials, there are important differences in the way subject headings are used in describing archival collections. First, records for archival collections may contain a larger number of subject headings than descriptions of other materials. With book cataloging, the cataloger is generally encouraged to keep subject headings to a minimum. (One cataloging instructor known to the writers of this paper trains her students to aim to cover everything in three subject headings.)  One leading cataloging text contends that "the ideal situation is one in which one heading will suffice to express the subject of the work being cataloged." [4]  Archival collections, unless extremely small, can rarely be described using three, let alone one, subject heading. In recognition of this fact, archival catalogers may use as many headings as needed to accurately describe a given collection.

However, there are certain limitations to the number of headings applied. An exhaustive list of topics is unlikely. In some cases, the library catalog may have limits. Some bibliographic utilities, such as OCLC, have limited the number of subject headings in the past due to limitations on field length or record size. [5]  Catalogers also try not to oversell the materials, to only include topics that are covered to a reasonable extent, and to include only individuals who are represented in an important way. For example, a collection containing one signed photo of John F. Kennedy, with no other references to him, would not get a heading for JFK. To include such a heading would imply that there was a meaningful amount of material relevant to our 35th President in the collection. Some collections, or parts of collections, may not get as much analysis as others. In addition, while trying not to make judgments about what is of "historical value," archival catalogers have to determine what parts of a collection are significant, and highlight those portions of the material. [6]  Human resources also present limits, since it takes time to evaluate the collection in depth and identify the appropriate authorized headings.

In addition to subject headings, archival catalogers are able to provide descriptive information in narrative form to help researchers determine if a particular collection might be useful. A catalog record indicates that the collection covers a particular topic. The summary (or scope and contents) note and the biographical and historical notes provide more detailed (although still brief) descriptions of the particular collection. These notes are not standardized beyond being assigned a particular location in the record. The text may be composed entirely by the cataloger, who determines what points will be covered and how they will be expressed. While these are often less detailed than similar information provided in a longer finding aid (discussed in detail in the next section of this paper), these notes should help the researcher determine if the collection is worth investigating further.

The "scope and contents" or "summary" note provides a brief description of the collection. It may include the formats of the materials included, such as films, correspondence and computer files, as well as information on the subjects, people, events and localities that are touched upon in the collection. For example, Figure 1 consists of the catalog entry for the Louisville and Nashville Railroad Company’s records at the University of Louisville.

This summary note provides information that will allow the researcher to make at least an initial assessment of whether the collection will contain relevant information or not. If the researcher is looking for pictures of steam engines, this collection may be of use to her or him. Further information can often be obtained in the finding aid, which in some cases is linked to the catalog record. The summary note field is indexed, which means that it can be searched via a keyword search, providing a "backdoor" into the collection.

The biographical/historical note seen in Figure 2 similarly provides either a brief history of an organization or agency or a brief biography of an individual or family. Again, this is just a brief summary, not as detailed as what may be available in the finding aid, but it may enable the researcher to make an initial assessment as to whether the collection is indeed relevant.

For example, this record of the L & N Railroad indicates it was chartered in 1850. A researcher interested in Civil War railroads might determine that this collection is worth learning more about. Once again, these fields are searchable, so that someone searching for the L&N under the alternate name "Old Reliable" will find this record.

Keyword searchability makes it possible for catalogers to intentionally provide keyword access to materials via common alternative terms. Once the researcher has found a useful record via keyword search, he or she may make use of one or more of the subject headings in that record to locate other useful records. It is not clear how common a practice this is among catalogers, however. [7]

While catalog records provide valuable information about collections, they have shortcomings. They are necessarily brief. They are also typically collection-level descriptions. While it is possible to describe different levels of the collection (e.g., providing a record of a particular series), and to indicate the relationship between different records, it is difficult to communicate the context of individual series, folders, or items in this manner. [8]  But what MARC records can do well is provide a succinct but meaningful description of the collection that allows researchers to discover the existence of the collection, and make an initial assessment as to its potential relevance.

.04 Benchmarks, Standards, Finding Aids

Having looked at the structure and content of the catalog record, and at its evolution in the context of archival description, the next step is to take a more general look at the whole concept of archival finding aids.  In fact, why do archivists have finding aids in the first place?  Why has the archival community invested so much thought and effort in designing and creating these descriptive tools?  Archival repositories have formed and grown as the result of a two-pronged impetus: preservation of historically and administratively valuable documents, and public access to those preserved materials. But more and more, the latter impetus, the use of these materials by researchers, has really come to be seen as the only acceptable justification for the high costs associated with preservation.  And if these collections are to be used effectively they have to be described—both to publicize their existence and to provide a guide for using them.  These are the roles played by archival finding aids.

The system of finding aids that has become the current model for most archives is a two-stage system of consciously linked information. The first stage is the catalog record that has already been discussed in some detail.  This is the more generalized of the two stages and provides collection information at a level that is less detailed.  The catalog record may be thought of as a very high-level finding aid, whose first purpose is resource discovery.  Within a catalog of resources—it might be an institution’s local catalog or a union catalog for a large consortium—the catalog record (expressed in the MARC21 format) permits the researcher to identify and distinguish a particular collection from among all the other resources (books, periodicals, audiotapes, artifacts, or other archival collections) comprising the library’s holdings. Following discovery of the collection, the catalog record's next purpose is to describe, in very general terms, the collection as a whole.  The objective is to characterize the collection well enough to allow the researcher to decide whether to investigate it further.  In this sense, the catalog record has low granularity in an informational sense.  The catalog record can also be seen to serve as a bridge from resource discovery to the overall collection description.

The second stage in the system of archival description is the detailed finding aid.  In archival repositories this finding aid is variously known as an inventory, a register, or a collection guide, and though its appearance differs somewhat from institution to institution, it always serves a definite set of roles, which both follow from and contrast with those played by the catalog record.  First of all, whereas the first objective of the catalog record is to permit the researcher to find the collection as a whole, the initial purpose of the detailed finding aid is to help the researcher to find the individual components (series, files, folders, items) that comprise the collection.  So, while the catalog record demonstrates low granularity (describing the collection as a whole in quite general terms), the detailed finding aid shows high granularity, focusing as it does on all the structural and informational pieces that make up the collection.  And, whereas the catalog record is a bridge from resource discovery to collection-level description, the detailed finding aid acts as a bridge from that collection-level description to component-level description.

From this start, a picture develops of how this two-stage finding aid system works in practice.  The researcher initially discovers the collection in an online catalog via access points (controlled vocabulary terms) embedded in the catalog record, the text of which record then provides a very encapsulated narrative summary of the entire collection.  The catalog record then refers the researcher to (or directly links to) the detailed finding aid, which begins with a greatly amplified description of the collection as a whole, but then presents the reader with an enumeration and description of the informational components that comprise the collection, so that the researcher can decide which specific pieces to study.

The general model depicted in Figure 3 has been adopted by the great majority of archival repositories and is used in local catalogs, as well as in union catalogs mounted by various consortia and by large bibliographic utilities like OCLC and RLG (Research Library Group).  But the ability of finding aids to intermix successfully and intelligently in these catalogs, and to share a helpful presentational uniformity, requires a huge amount of standardization on the back end.  Archivists have been actively developing and building a head spinning array of conceptual and technical standards over the past 20 years.  And they have been doing so for the primary goal of maximizing researcher access to archival collections.  Archivists, rightly or wrongly, have come to believe that the application of these descriptive standards will afford the most effective access to the greatest number of users.  They have come to see the enforcement of descriptive standards as being key to researcher success in several areas.  First, it reinforces predictability, in that users will, after a few attempts, become adept at navigating and understanding the finding aids because of the common elements in each.  Second, it will enable interoperability in the sense that finding aids can co-exist intelligibly in a union or interdisciplinary environment if they share a common structure and presentation.  Third, it creates efficiency for archivists by permitting them to develop templates and automation tools that would not be possible in an idiosyncratic environment.  And finally, because standards make use of recognized best practices, they help elevate the quality of description throughout the archival community.

There are many current and emerging archival descriptive standards, and they break down into four basic types, although many of them cross over typological boundaries.  There are data structurestandards that identify essential information categories and the relationships existing between those categories.  These are similar to the tables in a relational database and the keys and superkeys that relate the tables to each other.  The most prominent structural standard in the archival universe is known as ISAD(G) and is the International Council of Archives’ codification of the fundamental data elements that are essential to describe archival materials of any type or scope, and the manner in which they relate to the natural hierarchical structure of archival collections.  (See Figures 4 and 5 for representations of two important aspects of this fundamental standard.)  Then there are data content standards that serve as guides in deciding what specific types of information to include in these structural categories and how that information should be formatted.  This is the role that the Anglo-American Cataloging Rules play in the library world, and there are analogous publications in archives, as mentioned earlier.  A third type are data value standards, which actually provide controlled vocabulary terms (e.g., personal, corporate, and geographic names) that are to be used in defined situations.  The Library of Congress Subject Headings are a good example.  A final type are data interchange standards, which provide detailed technical specifications for encoding and transmitting data that will be loaded into (primarily) electronic information systems.  The MARC21 format for the exchange of bibliographic information is perhaps the most familiar example.

These interrelated standards together suggest a few key points for archival finding aids, and have greatly shaped their current iterations.  First, description is multi-level. Like the natural arrangement of the materials, it should proceed in a hierarchical fashion, first describing the whole, and then describing all of the component parts.  Second, the key data elements (See Figure 5) are repeatable at all hierarchical levels.  The data elements that are important to describing the whole collection, such as scope and content or extent or arrangement, are just as important in describing the smaller component parts.  Third, at any hierarchical level, only the information appropriate to that level is included.  This prevents redundancy and verbosity that can frustrate researchers.  Fourth, the same principles of description apply regardless of format.  So, the same data elements and structure should be used to describe motion pictures as to describe letters and diaries. And finally, it is just as important to describe the creators of the material as it is to describe the materials themselves.  Archivists believe that the circumstances of the records’ creation and subsequent administration shape and contextualize the topical content of the materials, so the finding aids should also reflect their creation and use.

Acceptance of these key points, and the standards that underlie them, have led to the model for archival finding aids—both catalog records and detailed finding aids—that is currently in use.  The model identifies several basic parts, each of which has applicability at every hierarchical level in the description.

In adopting these descriptive benchmarks and practices, many repositories have been investing considerable time reengineering their finding aid models.  This reengineering effort continues as archival educators conduct further user studies to test the usability and effectiveness of the current models.  There are several principles that continue to drive and shape the model.  They include: distinguishing between information about context and information about content, applying information at the appropriate level, and not repeating information at multiple levels.  A few additional principles are emerging and should be noted. Researchers do not enjoy reading finding aids and want to cut to the chase. Uniformity of structure and presentation from finding aid to finding aid is extremely valuable. And finally, finding aids are not transparent to users. Adequate navigational and way-finding tools must be built into them.

.05 Archival Finding Aids Online

The emergence of standards for what kinds of information should be included in descriptions of archival collections, and the development of Encoded Archival Description (EAD) to reproduce those descriptions in computer-readable form, have led to a rapidly increasing number of online finding aids employing the new standards.  Yet there is a good deal of variety, reflecting not only differences in computer sophistication and institutional resources but also differences of understanding about how the online environment can best be used for access to the contents of archival collections.  At one end of the spectrum are websites that present a finding aid in its word-processed form, whatever that form may be, with little to no internal navigability.   Navigation consists largely of scrolling through the document, perhaps with internal links to a few of the major portions of the finding aid from a brief table of contents list.  Search capabilities are limited to those of the browser, in Microsoft Windows the "Find" function under "Edit" on the Task Toolbar.  For small collections these are not major considerations, but for large collections, they can represent a barrier to usability.

Congressional papers collections provide a prime example of such large collections.  Routinely consisting of hundreds of archival cartons, their finding aids also tend to be hundreds of pages long in their typescript form.  While putting the finding aid online in this form is certainly better than not doing so, the challenge is how one uses the online environment to advantage to bring to potential users the information that they need and want when they need it and want it, and to make the other information available as they choose to seek it.  Navigability and searchability are primary considerations, with the recognition that experienced researchers may want to be able to go to the complete finding aid itself in its traditional form.

Examples are helpful.  The finding aid for the University of Connecticut's Thomas J. Dodd Papers transforms a basic typescript finding aid for a collection of 220 linear feet into a navigable document while still retaining the basic finding aid structure.  The opening page presents repository and collection overview information, with a table of contents at the upper left from which navigation to the major parts of the finding aid can be undertaken should the user not wish to scroll through the document to get to them.  Access to the folder lists is through the arrangement section, where descriptive information for each major division (series) of the collection is given, ending in a link to the folder list for each.  Each of those links leads to an opening page identical in format to the collection opening page except for a distinctive series heading and a new table of contents offering "Detailed Description," which leads to the folder list for that series.   Search capability depends on the browser. [9]

A finding aid that breaks with linear presentation and explores the potential of the online environment a bit more extensively is that for the William S. Cohen Papers at the University of Maine.  Here the opening page offers a color display with a photo of Cohen at the center and at the left five prominent links and one less prominent one.  Repository and institutional information is present at the bottom as links but minimized visually.  The five major links at the left are to the introduction and finding aid, a lengthy biographical note with photos and internal navigability via buttons across the top, a chart of Cohen's Committees, the Cohen Center, and links that lead to a list of Congressional and Other Resources.  The link to the introduction and finding aid leads to a single page that summarizes the collection, with buttons across the top leading to a lengthier collection overview and each of the main records groups in the collection.   Each of these record groups is presented with a single page overview narrative containing links to series or subdivisions within that group, supplemented by a tree outline at the upper left with the same subdivisions as links.  With each click moving down through the hierarchy, the tree changes, and descriptive information, including date spans and extent of material, is presented at each level.  The folder lists are not produced in this sequence, however.  Rather, they are accessible through a separate link entitled "Search the finding aid and sample digital material,"—the less prominent link mentioned above as being on the opening page.  That link also appears at the top right of the introduction and collection overview pages, and at the bottom of the opening pages for each of the major record groups in the collection.  It opens a search screen where a keyword can be entered and one of several choices made as to where to search, including for digital objects.  The results for keyword entries are highlighted in aqua and must be retrieved by scrolling through the entire document.  Also on the search screen is the option to "Browse the full finding aid,"—leading to the complete document from which the online guide to the collection was created. [10]

The site for the George J. Mitchell Papers at Bowdoin College similarly transforms the basic finding aid.  As for the Cohen Papers, the opening page presents a limited number of options to the left, including a biography with photos and five buttons leading to portions of the collection such as photographs, memorabilia, videos, speeches, and press releases, some of which offer digital images.  The first link, to the finding aid, leads to an overview summary of the collection with date spans, extent, cataloging number, background note, description, arrangement, access restrictions, and citation information.  Along the top, as on the opening page for the collection, are button links for the major record groups, and at the upper left is a tree outline which, as with the Cohen Papers, provides links in an ever-descending sequence through the hierarchy of the collection's arrangement.  At each link, the page format is the same but with the information geared specifically to that level, and potentially useful supplementary links are provided below the tree outline.  When one reaches the last level, a "container list" provides a list of the contents of individual cartons.  The major distinction between this finding aid and that for the Cohen Papers is that from the Mitchell site it is not possible to bring up the finding aid in its original linear form.   There is also no internal search mechanism, although the browser search capability is available. [11]

These examples by no means exhaust the possibilities of variation in online finding aids, but they do point the way to questions that might be raised.  The first of those questions might address the distinction among issues of content, issues of format, and issues of display.  A second line of inquiry might explore the related issues of navigability and retaining one's bearings in the online environment.  A third might address what the optimum search capability would be, and where and how that would be offered within a site.  All of these questions and more beg for further discussion, pending more extensive study of how researchers approach online resources, and what they want and need from online access to archival collections.

.06 Collaboration and Best Practices

Archivists have not been alone in devising new ways to deliver and share scholarly information.  Other cultural heritage organizations such as libraries, museums, and historical societies have had to develop means to manage and provide access to their paper and digital collections.  The current hybrid environment has led cultural heritage organizations to examine their organizational practices and develop new models for providing broad access to digital objects through collaboration.

The differing characteristics of the newer collections and increased user expectations have brought about examinations of traditional access methods.   Developing better management and access methods has to take into account several notions about the collections and users.  Collections can be distributed. That is, they are owned or managed jointly between institutions and associated organizations, and the user is unaware of these relationships.  For example, publishers make available electronic journals through a service provider for a subscription to a library for local use.  The availability of information on the Web has also affected user search strategies and interactions among reference staff.  Users are asking why they cannot conduct federated searching between collections in notjust one institution but also between the archives, libraries and museums among many institutions.  Users are also having less interaction with reference or curatorial staff, and are being directed straight to the materials via the Web.  Institutions have little data on how users locate a web site or online catalog, their search strategy, the results, and if they are satisfied with those results. Furthermore, cultural heritage organizations may be surprised that their user might be a ten-year-old or a professional historian. Archives, museums and libraries need to make their materials understandable to many audiences they neither see nor expected. [12]

.07 Collection Management

Archives and libraries generally have stronger traditions in using shared collection management systems that provide access to the public, as well as sharing resources for group purchases of journals and sharing their holdings with other institutions.  Museums have tended to focus more on their stewardship of collections—stewardship taking precedence over sharing through loans.  Nevertheless, many organizations have realized that comparing numerous works via the Web would have value to their users.

These changes, however, have not meant that the archival professions have had to abandon their tenets of providing access, protecting privacy, or using best practices and standards.  The opportunity to collaboratively manage their digital resources does mean that they can bring their resources to a broader audience and overcome limits of geography and exhibition space. A wide range of opportunity for collaboration can result in increased access to digitized collections by bringing new users to different institutions and sharing resources by reducing costs associated with equipment and expertise.  Staff can also learn new practices through training programs and materials available to all participants, which capitalizes on the expertise of their partners in both subject and technical knowledge. [13]

There are, however, increasing circumstances of groups collectively building something bigger than individual efforts, such as the Online Archive of California, the Northwest Digital Archives, North Carolina’s Exploring Cultural Heritage Online, and the Collaborative Digitization Program, not to mention several other state-wide initiatives.  In addition to dealing with management issues of staffing, resources, and priorities for digital conversion, collaboratives may find themselves dealing with matters of organizational culture on how

• to best provide physical security,
• to address property rights and confidentiality,
• to create metadata or standards for describing the digital objects,
• to image the original items to the highest possible resolution so that the item will not have to be rescanned and the digital surrogate can accommodate a number of uses,
• to manage rights (such the right to create and distribute derivatives of the original donation), and
• to initiate preservation activities, for arguably the materials receive their greatest use during the digitization process.

Collaborative efforts, while they involve more time being spent on coordination, do provide users with a single place to search for and locate materials for several institutions. [14]

.08 Metadata

Adoption of standards is key to effective sharing of resources and inter-institutional cooperation. The term “metadata” has become particularly common with the popularity of the World Wide Web. But the underlying concepts have been in use for as long as collections of information have been organized. Library catalogs represent a well-established variety of metadata that has served for decades as collection management and resource discovery tools.  Literally, “data about data,” metadata includes data associated with either an information system or an information object for purposes of description, administration, legal requirements, technical functionality, use and usage, and preservation. [15]

The creation of quality metadata is critical for the best management of information and digital files.  For many projects, metadata are created using existing catalogs, finding aids, or accession records that are used in combination with information about the digital files.  Some projects and programs begin creating their metadata from scratch.  There is no one standard for metadata creation that meets all the needs of all materials types or repositories.  Some common metadata schemas, or descriptions of the structure given to a set of metadata elements, include the following:

• Dublin Core [16] is an internationally recognized metadata standard comprised of fifteen basic elements used to describe a variety of networked resources. Although originally developed with an eye to describing document-like objects (because traditional text resources are fairly well understood), Dublin Core metadata can be applied to other resources as well.
• Text Encoding Initiative  [17] is an international cooperative effort to develop generic guidelines for a standard encoding schemes for scholarly texts, chiefly in the humanities, social sciences and linguistics. Since 1994, these guidelines have been a widely used standard for text materials for performing online research and teaching.
• Visual Resources Association Core Categories [18] are designed specifically to share information about works and their associated images.  The VRA Core Categories consist of a single element set that can be applied as many times as necessary to create records to describe works of visual culture as well as the images documenting them.
• Metadata Encoding Transmission Standard  [19] is an XML-based encoding standard for digital object metadata. It brings together structural, descriptive, and administrative metadata about a digital object and provides an XML document format for encoding that metadata for management of the digital object within a repository, and for the exchange of such objects between repositories.

.09 Metadata & Search Engines

The two major national bibliographic cooperatives, OCLC, the Online Computer Library Center, and RLG, the Research Library Group, with 50,540 and 150 members respectively, have initiated efforts to increase access to their databases through Internet search engines.  OCLC’s WorldCat database currently consists of approximately 57 million records [20] of differing media, and RLG’s Union Catalog consists of 45 million book titles. [21]  OCLC’s Open WorldCat program began as a pilot in September 2003 with 2 million records reflecting the most popularly held items without advanced searching.  It includes a distance indicator to the libraries and book cover images.  Beginning in January 2005, OCLC began making available all WorldCat records for indexing by Google and Yahoo!.  A search can be initiated by typing in the phrase, “Find in Library” in the Google or Yahoo! search boxes.  (See Figure 6.)  OCLC reported 2 million click-throughs during the testing phase [22] and individual institutions will be able to view statistics for the web traffic generated through this search function as more items are made available this year.

In September 2003, RLG began its RedLightGreen.com search page, making available 120 million book titles from its union database. (See Figure 7.) Similar to OCLC’s functionality, search results have an icon for “Libraries Near You” and a link to the local library catalog.  The program is aimed at college undergraduates and does not include archival materials. [23]  OCLC has worked with Yahoo! to create a custom Internet browser toolbar and RedLightGreen has a plug-in available for the Firefox toolbar search box.  To date, RLG’s materials are limited to books. Both RLG and OCLC’s search mechanisms lack the ability to limit by format.  While a step forward for simplified searching—some users might find it useful to know that nearly every local public library has a copy of The Da Vinci Code—primary source materials found in archives and special collections will eventually be made available through OpenWorldCat if they are held by OCLC members.

December 14th, 2004 was described as “the day the world changes” [24] with Google’s announcement that it would scan 15 million titles of the collections at the Universities of Michigan, Stanford, Harvard, Oxford, and the New York Public Library with the intent to make the results searchable.  In a non-exclusive agreement, Google is paying 150 million dollars for the scanning of materials and the labor involved in pulling books off the shelves.  Google has also apparently created a scanner that is non-intrusive—so much so that Michigan is allowing the scanning of a Bible from the eighth century.  It will take an estimated seven years to digitize the Michigan volumes. The project is worth an estimated one billion dollars to Michigan and it would have taken 1600 years without Google’s assistance.  The institutions will receive a digital copy of the books and for now, there will be limited advertising.  The citations come as part of a regular Google search preceded by the word “Book” and the search results are indicated with a book icon. See Figure 8 for a sample Google search.

For Google this marks an entry beyond the current search services market, valued at upwards of half a trillion dollars over the next several decades.  Indeed, the searchable web contains approximately 167 terabytes of data compared to the deep web’s 92,000 terabytes of private databases, and controlled access sites. [25]  While the scanning will provide Google with an unmatched critical mass of important materials for research, do people know what it means to search across such a large quantity of information? [26]  Librarians and archivists might accept the results because they know the scope of available data and limits of web searching capabilities.  Perhaps the best information professionals can do for patrons, as suggested by Stanley Wilder, is to connect them with information in a manner that reduces the complexity of the retrieval process. [27]

.010 Meeting User Needs

As archivists, we are proud of the improvements in catalog records, finding aids, and digital collections.  But are they meeting the needs of scholars?  A number of studies have been done to determine how scholars search for information and to assess how well finding aids and other descriptive standards meet their needs.

The archival community is now at a point of transition in archival research.  Jill Tatem has described this transition as the move from “traditional mediated services in archives to new unmediated, networked services.”  Under the traditional model, the archivist created a “ ‘precisely packaged’ response that met individual needs on a case-by-case basis … interpreting subject or name-oriented requests and deriving answers from provenance and context driven descriptions.” [28]  Online access to finding aids and to digital versions of the collections takes the archivist out of the mix.  How then might archivists change  finding aids to make them more usable in this unmediated environment?

.011 How Do Scholars Use Print Finding Aids?

Wendy M. Duff and Catherine Johnson conducted a survey of 11 mid-career historians at two Canadian universities.  They found that finding aids served several purposes for these historians:

• as a first step to gaining familiarity with an unfamiliar collection or archives
• as a summary source to gauge the extent or size of the collection and the types of materials it contained
• as secondary source of information about people or organizations
• as a source of names related to their topic. [29]

These historians continued to rely on archivists to locate materials that may not be listed explicitly in the finding aid, and to help them find their way through an unfamiliar finding aid structure. [30]  One member of the group noted that archivists are easier to use than finding aids and can take the researcher’s particular interests into consideration. [31]  On the other hand, online finding aids with folder or item level description allow the researcher to search more comprehensively, and they offer the potential for searching finding aids at several institutions at the same time. [32]

In 2003, the Primarily History project sought to determine how historians teaching American History at 68 top-ranked institutions locate primary sources. [33]  In an article summarizing the project’s findings to date, Tibbo noted that by now there is enough of a critical mass of finding aids available on the web that “it is reasonable to expect at least some historians to be using them.” [34]  But how are scholars using finding aids?  The results of user studies reveal findings that archivists might not expect.

.012 The Challenges and Promises of Electronic Finding Aids

Archivists tend to assume that scholars are familiar with traditional archives operations and archival access tools. [35]  Elizabeth Yakel’s 2002 study of common ground between researchers and archivists is one of many studies that have concluded that this simply is not true.  She found that the concept of a finding aid was difficult to explain to survey participants without giving away the answer—indicating just how far the concept is outside of the context of research elsewhere in the library. [36]  She reported that “researcher after researcher” commented on “the intricacies” of archival access systems, leading her to conclude that “finding aids are not the transparent tools for users that archivists intend.” [37]

Yakel’s study participants identified several problems:  1) getting lost in the hierarchy of the finding aid, 2) the length of the finding aid, and 3) unfamiliar terms used to label the parts of the finding aid.  Given the possibilities offered by technology to manipulate the ways in which information is displayed, Yakel suggested that archives should be able to address these difficulties in the online environment using EAD, through exploding box lists and navigation tools, for example. [38]  A study compiled by Burt Altman and John R. Nemmers on the Pepper OnLine Archival Retrieval and Information System (POLARIS), and Kim Rosenbusch’s study of 12 archival descriptive systems also found that navigation within a finding aid was one of the chief concerns voiced by study participants. [39]

Elizabeth Yakel used a focus group to test the usability of a group of finding aids from the Historic Pittsburgh Project, and she found that because archival terminology was confusing to the survey participants their use of the more sophisticated aspects of the search function was impeded. [40]

How a finding aid is formatted or displayed can also affect how useful it is.  Wendy Duff and Peka Stoyanova did a study of display options to see which appealed most to their test group (a range of types of archives users).  They found that in addition to formatting conventions (such as right justified or bolded labels), other display options affected the usability of finding aids.  Lists, for example, make it easier to browse a finding aid quickly.  Long descriptive paragraphs, on the other hand, can discourage some researchers from browsing if the amount of information seems overwhelming. [41]  Other researchers, however, found that detailed narrative descriptions provided useful contextual information—“photographs of Joe Smith and his family” rather than just “photographs,” for example. [42]

Archivist Chris Prom has pointed to the importance of visual clues.  He noted, for example, that dividing finding aids up between two windows made his study participants uncomfortable and tended to confuse them. [43]

Dennis Meissner, who participated in a finding aid reengineering project at the Minnesota Historical Society (MHS), usefully summarized these difficulties with finding aids as problems related to the structure, order, and presentation of the information in the finding aids. [44]  The order of the information was particularly important for researchers who need to quickly determine whether or not a collection would be relevant for their research. [45]  The group at the MHS also noted that finding aids should include explicit instructions for researchers on how the information in the finding aid should be used to request materials or to cite them in a publication.  In other words, it is not enough to provide information—there must also be understanding of how to use it. [46]

In a review of the implementation of EAD, James Roth noted that despite these problems with electronic finding aids evaluations of EAD indicate that researchers are more interested in the content of the finding aid than in its structure or format. [47]  As archivists seek to make the structure and presentation of finding aids more transparent and intuitive, they will also need to pay attention to which kinds of information interest scholars.

Studies done over the past ten years indicate that scholarly researchers tend to use chronological, geographic, individual or group names, and topical subject terms to locate relevant archival collections. [48]  But do scholars search for collections in these ways because they are adapting their research strategies to match the strengths of existing finding aids and catalog records?  And have changes in the subjects of scholarly research affected the ways in which researchers search for primary sources?

Duff and Johnson found that historians are conscious that archival collections are organized according to the person or organization who collected or created the records, and that they therefore tailor their search strategies to focus on names.  They purposefully keep track of the people and organizations related to their topic.  “As one historian put it, ‘Well, you need to know where things are going to end up.’” [49]

The Primarily History project not surprisingly found that the traditional method of describing archival collections—based on the person or group that created or collected the collection—does a good job of supporting research on individuals.  These descriptive practices, including finding aids and repository guides, are less helpful to social historians.  Historians working on biographical projects were found to have used a “wider variety of methods” to locate materials than did historians pursuing other types of research (cultural, economic, political, social).  While this may indicate that they were having trouble finding what they were looking for, it also might reflect the fact that most bibliographic systems do a good job of capturing personal names.  Social historians, on the other hand, may use fewer search methods because the kinds of topics they are looking for—“the immigrant experience,” for example—are not easily captured in bibliographic systems.  For these types of topics, it may be more efficient for the researcher to consult with an archivist. [50]  Jackie Dooley has pointed out that, “Many historians have shifted their interests from elites to masses: in subject access terms, this means a change from people of known name and title to those of anonymous name and no title at all.” [51]

Another study by Diane L. Beattie reached similar conclusions.  Beattie surveyed researchers in Women’s History and found that they were more likely to use informal channels to locate primary sources.  Although two-thirds of the researchers in the study could associate personal or corporate names with their topics, fewer than one-fifth of them used indexes or finding aids to locate relevant materials, relying instead on archivists to help them.  Interestingly, this same group of researchers ranked finding aids as more helpful than other formal and informal research aids. [52]  Duff and Johnson similarly found that social historians found collections organized around creators difficult to use for their purposes.  They were looking for a way to search for subjects across collections. [53]  EAD finding aids have made such searches practical, perhaps helping to bridge this gap.

.013 Searching Finding Aids

Data has also been gathered on the ways in which researchers search online finding aids.  In general, studies have found that full-text searches retrieve a higher percentage of all relevant items and that for abstract or index records a higher proportion of the records retrieved are relevant. [54] Researchers more interested in seeing everything that might be relevant, therefore, would be better served by searching the full text of the finding aid rather than the abstracted MARC record.  Researchers looking for only the most relevant records might be better served searching the MARC record—assuming that they are looking for a name or other subject term that is likely to have been used in these collection level records. [55]  Researchers who are more familiar with their library catalog than with traditional archival finding aids may find it helpful to be able to search the online catalog and then link to the full text finding aid. [56]

Susan Hamburger found that 78 percent of the respondents to her study of researcher search strategies used keyword searches, while 31 percent used personal names and 23 percent used subject terms. [57]  The predominance of keyword searching reflected both the researchers’ familiarity with keyword searching on the web and the fact that they were searching for terms not likely to have been indexed by archivists.

In 2003 Chris Prom of the University of Illinois conducted a study contrasting the search strategies of researchers with experience using archives, researchers with experience using computers, and researchers who were novices in both areas.  He asked researchers to locate known information in finding aids using the websites and finding aids developed at several archives across the country.  Like Hamburger, Prom found that many researchers started with a keyword search if a search box was provided. [58]  Websites that provided researchers with several searching options confused researchers, particularly the novices. [59]

For researchers searching for particular collections, keyword search boxes (like those made popular by internet search engines like Google) and alphabetized lists of collections proved the most efficient and straightforward ways to finding relevant finding aids. [60]  As was noted earlier, however, many historians need to find materials related to subjects that do not fit well within this model.  Prom concluded his report with the observation that “the well-known role of the archivist in mediating access to collections has been enhanced, not diminished by the internet.” [61]

.014 Conclusion

The archivist’s task is to present information that is meaningful for researchers in way that is easy to understand.   User studies to date have concentrated on the structure and display of the finding aid more than on its content.  These surveys have indicated that the archival community needs

• to replace archival jargon with common terms,
• to improve navigation within finding aids so that researchers do not become lost in the hierarchy,
• to redesign the look of online finding aids to make them easier to read,
• to simplify options for searching and browsing, and
• to add instructions about how to use the information found in the finding aids.

Information about what kinds of content would be helpful to researchers is more limited and suggests that different kinds of information may be useful to different groups of researchers.  An experienced historian, for example, may wish to see extensive and detailed descriptions of the kinds of records contained in the collection, while a genealogist may find that level of detail intimidating.  The online environment offers an opportunity to make finding aids more flexible and to customize the presentation of information to meet the needs of different groups of researchers.  As archivists continue to develop online finding aids, they will need to continue to seek feedback from historians and other scholars.

.015 Notes

1. Lyn M. Martin, “Viewing the Field: A Literature Review and Survey of the Use of U.S. MARC AMC in U.S. Academic Archives,” American Archivist 57 (Summer 1994): 492.

2. Robert L. Maxwell & Margaret F. Maxwell, Maxwell’s Handbook for AACR2R: Explaining and Illustrating the Anglo-American Cataloguing Rules and the 1993 Amendments (Chicago: American Library Association, 1997), 154.

3. Richard P. Smiraglia, "Subject Access to Archival Materials Using LCSH."  In Describing Archival Materials: The Use of the MARC AMC Format, ed. Richard P. Smiraglia, 81, (Binghamton, NY: Haworth Press, 1990).

4. Lois Mai Chen, Cataloging and Classification: An Introduction, 2nd ed.  (New York: McGraw-Hill, 1994),168.

5. Lisa B. Weber, "Record Formatting: MARC AMC."   In Describing Archival Materials: The Use of the MARC AMC Format, ed. Richard P. Smiraglia, 130, 136 (Binghamton, NY: Haworth Press, 1990); Peter Carini and Kelcy Shepherd, "The MARC Standard and Encoded Archival Description," Library Hi Tech 22: 1 (2004): 19.

6. Smiraglia, "Subject Access," 69, 74; Helen Tibbo, “The Epic Struggle: Subject Retrieval from Large Bibliographic Databases,” American Archivist 57 (Spring 1994): 321-323.

7. Susan Hamburger makes a similar recommendation in “How Researchers Search for Manuscript and Archival Collections,” Journal of Archival Organization 2:1/2 (2004): 91.

8. Carini and Shepherd, "The MARC Standard," 19.

9. See http://www.lib.uconn.edu/online/research/speclib/.../MSS19940065.html#d0e38

10. See http://www.library.umaine.edu/cohen/default.htm

11. See (http://library.bowdoin.edu/arch/mitchell/index.html)

12. For a discussion of the shift between archival users and reference services see Elizabeth Yakel, “Thinking Inside and Outside the Boxes:  Archival Reference Services at the Turn of the Century,” Archivaria 49 (2000):  140-160 and Elizabeth Yakel, “Information Literacy for Primary Sources:  Creating a New Paradigm for Archival Researcher Education,” OCLC Systems & Services 20:2 (2004): 61-64.

13. For an excellent review of collaboration’s benefits to organizations and users, see Nancy Allen and Liz Bishoff.  “Collaborative Digitization:  Libraries and Museums Working Together,” Advances in Librarianship, 26 (2002):  43-81.

14. Allen and Bishoff, “Collaborative Digitization,” 64-73.

15. Diane Hillman, “Using Dublin Core” http://dublincore.org/documents/usageguide/ (accessed February 16, 2005).

16. Dublin Core Metadata Initiative, http://dublincore.org/ (accessed February 16, 2005).

17. Text Encoding Initiative, http://www.tei-c.org/ (accessed February 16, 2005).

18. VRA Core Categories, Version 3.0,http://www.vraweb.org/vracore3.htm (accessed February 16, 2005).

19. Metadata Transmission and Encoding Standard (METS), http://www.loc.gov/standards/mets/ (accessed February 17, 2005).

20. WorldCat [OCLC - Home], http://www.oclc.org/worldcat/default.htm (accessed February 17, 2005).

21. RLG Union Catalog, http://www.rlg.org/en/page.php?Page_ID=174 (accessed February 17, 2005).

22. OCLC, 2003/2004 Annual Report, http://www.oclc.org/news/publications/annualreports/2004.pdf: 17 (accessed February 17, 2005); OpenWorldCat Program, http://www.oclc.org/worldcat/open/default.htm (accessed February 17, 2005).

23. RedLightGreen, http://www.rlg.org/en/page.php?Page_ID=435 (accessed February 17, 2005).

24. Barbara Quinn, “Google’s Library Project: Questions, Questions, Questions,”  Information Today, December 27, 2004, http://www.infotoday.com/newsbreaks/nb041227-2.shtml (accessed February 17, 2005).

25. Charles H. Ferguson, “What’s Next for Google,” Technology Review.com.  January 2005, http://www2.technologyreview.com/ (accessed February 17, 2005).

26. Walter Kohler, “Digital Libraries, Digital Containers, ‘Library Patrons’ and Visions of the Future,” Electronic Library 22:5 (2004):  401-407;  Roy Tennant, “Google Out of Print,” Library Journal, February 15, 2005, http://www.libraryjournal.com/article/CA50214 (accessed February 17, 2005).

27. Stanley Wilder, “Information Literacy Makes All the Wrong Assumptions,”  Chronicle of Higher Education, January 7, 2005: B13.

28. Matthew Young Eidson, “Describing Anything that Walks: The Problem Behind the Problem of EAD,” Journal of Archival Organization 1:4 (2002): 16.

29. Wendy M. Duff and Catherine A. Johnson, “Accidentally Found on Purpose: Information-Seeking Behavior of Historians in Archives,” Library Quarterly 72:4 (October 2002): 483, 493.

30. Duff and Johnson, 483.

31. Duff and Johnson, 485.

32. Duff and Johnson, 486.

33. Helen Tibbo, “Primarily History in America: How U.S. Historians Search for Primary Materials at the Dawn of the Digital Age.”  American Archivist 66:1 (Spring/Summer 2003): 9.

34. Tibbo, "Primarily History," 13.

35. Elizabeth Yakel, “Listening to Users,” Archival Issues: Journal of the Midwest Archives Conference 26:2 (2002): 111.

36. Yakel, "Listening," 116-117.

37. Yakel, "Listening," 122; Christopher J. Prom, “User Interactions with Electronic Finding Aids in a Controlled Setting,”  American Archivist 67 (Fall/Winter 2004): 246.

38. Lisa R. Coats, “Users of EAD Finding Aids:  Who Are They and Are They Satisfied?”  Journal of Archival Organization 2:3 (2004): 31.

39. Jihyun Kim, “EAD Encoding and Display: A Content Analysis,” Journal of Archival Organization 2:3 (2004): 42-43.

40. Elizabeth Yakel, “Encoded Archival Description: Are Finding Aids Boundary Spanners or Barriers for Users?”  Journal of Archival Organization 2:1/2 (2004): 74-75.

41. Wendy M. Duff and Penka Stoyanova, “Transforming the Crazy Quilt:  Archival Displays from a User’s Point of View,” Archivaria 45 (Spring 1998): 51-52.

42. Duff and Stoyanova, 55.

43. Christopher J. Prom, “User Interactions with Electronic Finding Aids in a Controlled Setting,” American Archivist 67 (Fall/Winter 2004): 262.

44. Dennis Meissner, “First Things First:  Reengineering Finding Aids for Implementation of EAD,” American Archivist 60 (Fall 1997): 375.

45. Meissner, 379.

46. Meissner, 381.

47. James M. Roth, “Serving Up EAD: an Exploratory Study on the Deployment and Utilization of Encoded Archival Description Finding Aids,” American Archivist 64:2 (Fall/Winter 2001): 234.

48. Rita L. H. Czeck, “Archival MARC Records and Finding Aids in the Context of End-User Subject Access to Archival Collections,” American Archivist 61 (Fall 1998): 428; Kim, 50; Tibbo, "Primarily History," 21, 24.

49. Duff and Johnson, 493.

50. Tibbo, "Primarily History," 21, 24.

51. Coats, 29.

52. Duff and Johnson, 477.

53. Duff and Johnson, 493.

54. Czeck, 428.

55. Czeck, 430.

56. Yakel, "Listening," 119.

57. Hamburger, 84.

58. Prom, 251.

59. Prom, 250.

60. Prom, 263.

61. Prom, 265.

.016 Bibliography

Allen, Nancy and Liz Bishoff.  “Collaborative Digitization:  Libraries and Museums Working Together.”  Advances in Librarianship, 26 (2002):  43-81.

Carini, Peter, and Kelcy Shepherd. "The MARC Standard and Encoded Archival Description," Library Hi Tech 22: 1 (2004): 18-27.  Library Literature and Information Full-Text, Wilson Web (accessed 8 February 2005).

Chen, Lois Mai.  Cataloging and Classification: An Introduction, 2nd ed.  New York: McGraw-Hill, 1994.

Cohen, William S.  Papers.  Raymond H. Fogler Library, University of Maine.  http://www.library.umaine.edu/cohen/default.htm.

Coats, Lisa R.  “Users of EAD Finding Aids:  Who Are They and Are They Satisfied?”  Journal of Archival Organization 2:3 (2004): 25-39.

Czeck, Rita L. H.  “Archival MARC Records and Finding Aids in the Context of End-User Subject Access to Archival Collections.”  American Archivist 61 (Fall 1998): 426-440.

Dodd, Thomas J.  Papers.  Archives and Special Collections, Thomas J. Dodd Center, University of Connecticut.  http://www.lib.uconn.edu/online/research/speclib/ASC/findaids/Dodd_Thomas/MSS19940065.html.

Dublin Core Metadata Initiative. http://dublincore.org/ (accessed February 16, 2005).

Duff, Wendy M. and Catherine A. Johnson.  “Accidentally Found on Purpose: Information-Seeking Behavior of Historians in Archives.”  Library Quarterly 72:4 (October 2002): 472-96.

Duff, Wendy M. and Penka Stoyanova.  “Transforming the Crazy Quilt:  Archival Displays from a User’s Point of View.”  Archivaria 45 (Spring 1998): 44-79.

Eidson, Matthew Young.  “Describing Anything that Walks:  The Problem Behind the Problem of EAD.”  Journal of Archival Organization 1:4 (2002): 5-28.

Ferguson, Charles H.  “What’s Next for Google.”  Technology Review.com.  (January 2005). http://www2.technologyreview.com/ (accessed February 17, 2005).

Hamburger, Susan.  “How Researchers Search for Manuscript and Archival Collections.”  Journal of Archival Organization 2:1/2 (2004): 79-102.

Hillman, Diane. “Using Dublin Core” http://dublincore.org/documents/usageguide/ (accessed February 16, 2005).

Kim, Jihyun.  “EAD Encoding and Display: A Content Analysis.”  Journal of Archival Organization 2:3 (2004): 41-55.

Kohler, Walter.  “Digital Libraries, Digital Containers, ‘Library Patrons’ and Visions of the Future.” Electronic Library 22:5 (2004):  401-407.

Martin, Lyn M.  “Viewing the Field: A Literature Review and Survey of the Use of U.S. MARC AMC in U.S. Academic Archives,” American Archivist 57 (Summer 1994): 482-496.

Maxwell, Robert L.  & Margaret F. Maxwell, Maxwell’s Handbook for AACR2R: Explaining and Illustrating the Anglo-American Cataloguing Rules and the 1993 Amendments.  Chicago: American Library Association, 1997.

Meissner, Dennis.  “First Things First:  Reengineering Finding Aids for Implementation of EAD.”  American Archivist 60 (Fall 1997): 372-387.

Metadata Transmission and Encoding Standard (METS). http://www.loc.gov/standards/mets/ (accessed February 17, 2005).

Mitchell, George J.  Papers.  Bowdoin College Library.  http://library.bowdoin.edu/arch/mitchell/index.html.

OCLC. 2003/2004 Annual Report. http://www.oclc.org/news/publications/annualreports/2004.pdf (accessed February 17, 2005).

OpenWorldCat Program.http://www.oclc.org/worldcat/open/default.htm  (accessed February 17, 2005).

Prom, Christopher J.  “User Interactions with Electronic Finding Aids in a Controlled Setting.”  American Archivist 67 (Fall/Winter 2004): 234-268.

Quinn, Barbara.  “Google’s Library Project: Questions, Questions, Questions.”  Information Today, December 27, 2004. http://www.infotoday.com/newsbreaks/nb041227-2.shtml (accessed February 17, 2005).

RedLightGreen. http://www.rlg.org/en/page.php?Page_ID=435 (accessed February 17, 2005).

RLG Union Catalog. http://www.rlg.org/en/page.php?Page_ID=174 (accessed February 17, 2005).

Roth, James M.  “Serving Up EAD: an Exploratory Study on the Deployment and Utilization of Encoded Archival Description Finding Aids.”  American Archivist 64:2 (Fall/Winter 2001): 214-37.

Smiraglia, Richard P.  "Subject Access to Archival Materials Using LCSH."  In Describing Archival Materials: The Use of the MARC AMC Format, edited by Richard P. Smiraglia, 63-90.  Binghamton, NY: Haworth Press, 1990.

Szary, Richard.  “Encoding Finding Aids as a Transforming Technology in Archival Reference Service.”  In Encoded Archival Description on the Internet, edited by Daniel V. Pitti and Wendy M. Duff, 187-197. Binghamton, NY: Haworth Information Press, 2001.

Tennant, Roy. “Google Out of Print.” Library Journal, February 15, 2005. http://www.libraryjournal.com/article/CA50214 (accessed February 17, 2005).

Text Encoding Initiative. http://www.tei-c.org/ (accessed February 16, 2005).

Tibbo, Helen.  “The Epic Struggle: Subject Retrieval from Large Bibliographic Databases.”  American Archivist 57 (Spring 1994): 310-326.

Tibbo, Helen.  “Primarily History in America: How U.S. Historians Search for Primary Materials at the Dawn of the Digital Age.”  American Archivist 66:1 (Spring/Summer 2003): 9-50.

Tibbo, Helen and Lockman I. Meho.  “Finding Finding Aids on the World Wide Web.”  American Archivist 64 (Spring/Summer 2001): 61-77.

VRA Core Categories, Version 3.0. http://www.vraweb.org/vracore3.htm (accessed February 16, 2005).

Weber, Lisa B.  "Record Formatting: MARC AMC."   In Describing Archival Materials: The Use of the MARC AMC Format, edited by Richard P. Smiraglia, 117-143.  Binghamton, NY: Haworth Press, 1990.

Wilder, Stanley.  “Information Literacy Makes All the Wrong Assumptions.”  The Chronicle of Higher Education, January 7, 2005: B13.

WorldCat [OCLC - Home]. http://www.oclc.org/worldcat/default.htm (accessed February 17, 2005).

Yakel, Elizabeth.  “Encoded Archival Description: Are Finding Aids Boundary Spanners or Barriers for Users?”  Journal of Archival Organization 2:1/2 (2004): 63-77.

Yakel, Elizabeth. “Information Literacy for Primary Sources:  Creating a New Paradigm for Archival Researcher Education.”  OCLC Systems & Services 20:2 (2004): 61-64.

Yakel, Elizabeth.  “Listening to Users.”  Archival Issues: Journal of the Midwest Archives Conference 26:2 (2002): 111-27.

Yakel, Elizabeth. “Thinking Inside and Outside the Boxes:  Archival Reference Services at the Turn of the Century,” Archivaria 49 (2000):  140-160.