of the American Society for Information Science

Go to
 Bulletin Index

Volume 25, No. 6

bookstore2Go to the ASIS Bookstore

August / September 1999



Bringing Fashion Out of the Closet: Classification Structure for the Drexel Historic Costume Collection

by Abby A. Goodrum and Kathi Martin

The Drexel Digital Fashion Project is a joint initiative between the College of Information Science & Technology and the College of Design Arts. It represents the first of several planned projects that will be combined to form the Drexel Digital Museum. The impetus for the project has been the need to provide access to Drexel University's rich collections of art, textiles, clothing, ceramics and artifacts from around the world.

The Drexel Historic Costume Collection had its beginning in the 1890s when members of the Drexel family began assembling a collection of notable garments, accessories and textiles. The collection represents 200 years of historic costume and fabric design. Among the items are eight gowns by Charles Worth. One gown, complete with kneeling pillow, was created for Minnie Drexel Fell Cassatt's* presentation to the Austrian Court. The extensive lace collection has been featured in an outstanding resource book on this textile (Veksler, Bella, Lace: The Poetry of Fashion, 1998.) Shoes, millinery, parasols, gloves and other accessories in the collection present an opportunity to study an entire period ensemble. The collection is estimated to contain approximately 7000 items, although that number may prove to be much greater after a complete inventory of accessories and textiles is completed.

There is currently no place to publicly display this collection and no funding to support a display. This is in keeping with a nationwide downward trend in university resources devoted to costume collections overall, as Arthur noted in a recent article, "Resources for Costume Collections in American Universities" in the Journal of Family and Consumer Sciences (Fall 1997, 57-60). Much of the costume collection for example, is packed away in storage boxes and is only seen by students and faculty in the fashion design program. This is not entirely problematic as the garments are extremely fragile and must not be over-handled. While some documentation of the collection exists on a few 3 x 5 cards, the bulk of the collection is undocumented. The basic need is to bring the Historic Costume Collection out of the closet so that it can be accessed by a variety of users. Creating a searchable database of digitized images and supporting documentation for each piece offers a means by which the collection may be accessed by students, scholars, designers and other interested individuals around the world.

Analysis of Needs and Resources

To begin to address this situation we undertook an analysis of potential users and resources. The purpose of the analysis was to provide a user-centered framework for designing the database and to identify low cost methods for delivering the database. The analysis involved nine steps:

  1. Define our mission and responsibilities.
  2. Identify population to be served (users).
  3. Identify users' needs.
  4. Define access points based on needs of users.
  5. Identify internal as well as external sources of database creation and support.
  6. Identify internal (free) resources for data input (catalogers).
  7. Evaluate consistency of catalogers/identify training needs.
  8. Develop the database using a small sub-set of items from the collection.
  9. Evaluate pilot-test database.

At present, we have completed steps 1- 6 and are moving steadily forward. The greatest challenge, and the step requiring the most effort and time, has been step 4: Define access points based on needs of users. The rest of this paper will describe in more detail the needs of our users and what happened to us when we tried to construct a classification scheme for the collection.

Identifying Users' Needs

 Interviews with students and faculty at the College of Design Arts as well as feedback from fellow fashion design educators at the ITAA (International Textile and Apparel Association) held in Dallas in October, 1998, identified the three main user groups. Myra Walker, curator of the North Texas Fashion Collection, was particularly helpful in determining who would use such a site. For all groups of the user population defined in step 2, a questionnaire was designed to identify areas of interest, needs and uses. Additionally, we sought to identify frequency of Web site use for these interests. The following is a brief summary of the user groups and their needs identified by the questionnaire.

Designers (Fashion) and Designers (Textile/Fabric). These two groups' conceptual models are formed by their design school training and/or by their work experiences. Their search for information on the history of costume is accomplished primarily via books, articles and museum visits. Exposure to visual representations (plates, slides and actual garments) is very important. They cite their own hands-on work with fabric and clothing as being very important to them. Similarly, work done in tailoring, draping, pattern making and construction skills was considered more important than information gleaned from books or museums. Detailed depiction of sundries, finishes, linings and construction of historic costume of all eras is a valuable point of inspiration for the designer of contemporary garments and absolutely necessary for the authentic replication of historic costume. While some members of these groups are using CAD and the WWW for information gathering and work-related activities, the majority are not.

Students. Our mission as a University-based collection is primarily to educate. Responses from students in our own program were therefore extremely important. Our design students shared many of the same information needs and attitudes as the designers groups. They are more comfortable using the Web for research and in using CAD for the mechanical aspects of design and pattern making.  In addition to the questionnaire, we interviewed and observed a fashion design graduate student conducting research on a Charles Worth gown in the collection. This student exhibited a greater need for supporting documentation and published materials. A number of questions were raised for which there were no ready answers.  Why is the kind of information from one garment to the next so inconsistent? How do I find out more about the Drexel family and where the owner of the gown fits into the family hierarchy? What other gowns do we have that were designed by Worth, and for whom were they designed? What other types of garments were being designed at this time?

Scholars (Historians, Archivists and Design Faculty). Scholarly research in historic costume extends the need to examine the actual objects both visually and physically. Photographic surrogates are seen as less useful than the objects themselves. The lack of adequate depiction of minute detail and textures and the inability to see all sides of an object, including the inside, were reasons cited for needing access to physical objects. Scholars required greater need for supporting documentation including journal articles, books, historical information, provenance and exhibition catalogs. These groups used the Web for research less often than libraries and museums, citing a lack of full-text historical manuscripts and exhibition records available.

Access Points for an Historic Costume Collection

In examining the needs of our user populations, several themes emerged. First, all users require access to some form of visual surrogate of the object. While this surrogate is not considered an adequate substitute for visual inspection of the actual object, many see it as an important mechanism for determining if an object is worth the time and effort required to inspect it personally. Second, users require access to multiple views of a garment, including inside seams and details of construction, closures, embellishments, weave, etc. Third, many users require additional supporting documentation and would welcome access to provenance and bibliographies of published sources of information on a garment.

In order to support the visual information needs of our users we have decided to have each object photographed in 360 degree multiple views. These still images will be morphed into a 3-dimensional navigable space using QuickTime VR. As a first step in providing a searchable database, we devised a cataloging form to document each garment in the collection and to link to the visual surrogates. The cataloging form was created by adapting existing classification structures for art images, fashion and textiles to the needs of our users. We borrowed extensively from the Core Categories for Visual Resources (VRA Core), the fields used in the Museum Educational Site License Project and the Longhouse Reserve Fabric Classification Guidelines. A similar project underway at Kent State University has also created a unique hybrid classification scheme based on the VRA, AACR/MARC and the Dublin Core (Zeng, M., "Metadata Elements for Object Description and Representation," Journal of the American Society for Information Science, in press).

 From these existing structures, we initially defined over 200 fields and rules for their data entry. These fields cover three general categories:

  • Accession Information: type of item, primary and secondary designers, storage location, accession number, URL, donor, provenance, etc.
  • Descriptive Information: fabrics, style, genre, region or country of origin, accessories, patterns, embellishments, etc.
  • Additional Documentation: books, articles, photographs, videos etc.

In order to support data entry by untrained students, a Web-based template was designed as a front end to populate the database. The template is still evolving, but it essentially comprises the following elements:

  • Record Creation: simple to use forms-based cataloging that provides pull-down help screens for each field
  • Record Edit/View: provides entire record for editing or further processing
  • Record Export: provides mechanism to export data in comma deleted format for upload to other database.

The entire template may be viewed on the Web at www.drexel.edu/univrel/kpw/dfc/

Although the template itself is simple to use, choosing appropriate terms to describe garments and their construction, fabrics and their construction, and designer or manufacturer names poses numerous challenges. Although sources of controlled vocabulary exist, knowing which source to use and how to use it has been problematic.

Of Bustiers and Corsets

The following sources are not meant to constitute an exhaustive list of resources. They are listed only as an indication of the types of resources available to guide in the selection of terms to aid in the description of garments, designers and fabrics.

Terms used to describe garments and their construction may be taken from the following sources:

  • Art and Architecture Thesaurus
  • Dictionary of Costume
  • ICOM (International Council of Museum's) Guidelines

While these sources taken together provide excellent coverage of costume, knowing which source to use and knowing how to use it consistently has proven difficult. For example, what is the difference between a corset and a bustier? Between a slip and a camisole? Between tap pants and boxer shorts? Is it enough to catalog all of these as underwear? 

Terms used to describe fabrics, textiles and their construction may be taken from the following:

  • The Longhouse Reserve Classification
  • Women's Wear Daily Suppliers Guide
  • The Encyclopedia of Textiles
  • Fabric Glossary

The difficulty here lies in describing multiple ways of creating and embellishing textiles. For example, a fabric may be constructed by weaving together strips of leather and cotton, which may then be dyed and finally embellished with feathers and beads.

Designer's names and the names of design houses and firms may be taken from the following:

  • The Dictionary of Fashion and Fashion Designers
  • The Designer Database
  • Encyclopedia of Fashion

The problem here is in determining an authoritative source for identifying the designer of a garment. For example, do we use the name listed in the label sewn into the garment? What if the label is for a subdivision of a larger design house; do we list the parent company? Should we list the design house as designer although we have supporting documentation that a specific named person was the actual designer? What if there is no label or if there are several labels?

These challenges will be immediately recognizable to anyone engaged in documenting historic fashion collections, and yet we have as yet not been able to locate a single comprehensive set of rules to aid us in our endeavor. The template that we are developing must be capable of providing easy to use drop-down menus to guide the selection of appropriate terms for each field. This is essential if data entry is to be done by student workers with little or no experience in cataloging and costume history. One of the things that we are experimenting with currently is the provision of visual examples for such things as bustiers and corsets, batik and tie-dye, buttons and bows, etc. It is our fondest hope that when this project is completed, we will be able to provide an easy to use Web-based template that others will implement in documenting their collections.

Although the Drexel Digital Fashion Project is a long way from completion, we are encouraged by similar efforts underway at Kent State, the University of North Texas, ICOM and other fashion, textile and costume repositories.

The authors are grateful for the work of the following students in developing the cataloging template and the Web-based forms: Cheryl Olsen, Lydia Javins, Ella Lewin, Salman Malik and Kyle Welliston. Special thanks as well to Renee Chase, Fashion Design Program Director at Drexel.

*Authors' note:  For those of you who might be wondering, the answer is "Yes." Minnie Cassatt was the wife of Robert Kelso Cassatt, Jr., the nephew of Mary Cassatt, the Impressionist painter. The Cassatts, like the Drexels, were a socially prominent Philadelphia family.

One of the many lace fashions featured in Lace: The Poetry of Fashion by Bella Veksler.

Abby Goodrum is assistant professor in the College of Information Science and Technology at Drexel University. She can be reached by mail at 3141 Chestnut Street, Philadelphia, PA 19104-2875; by phone at 215/895-6627; or by e-mail at goodruaa@dunx1.ocs.drexel.edu

Kathi Martin is assistant professor, Design Department, College of Design Arts at Drexel University. She  can be reached there by mail; by phone at 215/895-4941; by fax at 215/895-4917; or by e-mail at martink@drexel.edu



How to Order

@ 1999, American Society for Information Science