NARRATIVE

NATIONAL IMPACT
Proposal: The project will bring together historic costume collections from several institutions into a single searchable web accessible database. The partner institutions are libraries and museums from metropolitan Detroit. Their collections number more than 40,000 items and cover 300 years of fashion; clothing from 1800-2000 will be selected for this project. Project partners include:

Detroit Historical Museums
This collection has over 30,000 items of clothing and accessories worn by a broad range of Detroit citizens over the past 200 years. The collection includes examples of occupational, formal, recreational, mourning, masquerade, and everyday wear for men, women, and children. Pieces by European and American designers, local dressmakers and retailers, and homemade garments are included. Special pieces include a wedding gown, worn at St. Anne's Church in 1821, the suit worn by Soapy Williams for his inauguration as Governor, a fireman's parade coat from the 1870's, and a child's satin flag costume from 1904. The true strength of the collection is its wide range of more mundane items like corsets, pajamas, parasols, hats, shoes, bathing suits and shawls. Detroit's role as a major processor of furs throughout the 19th and early 20th centuries is reflected in a very fine collection of coats, scarves, tippets and muffs.

Henry Ford Museum & Greenfield Village
Henry Ford Museum & Greenfield Village's clothing and personal effects collection is recognized by the National Endowment for the Humanities as one of national significance. It contains over 10,000 items and ranges in date from l750 to the present day. The collection includes ordinary folks' clothing as well as expensive, custom-made couture dress. Artifacts are as diverse as a World War II concentration camp survivor's uniform, wool hats made by Christian Dior, and African-Americans' make-up from 1945.

Meadow Brook Hall
Meadow Brook Hall's clothing and accessories collection, consisting of over 500 items, is that of the home's owner, Matilda Dodge Wilson. Born in 1883 and passing away in 1967, Matilda Dodge Wilson experienced her share of fashion trends from Edwardian sheaths and oversized hats to the cinched waists and full skirts of Christian Dior's "New Look." The fashion and costume collection at Meadow Brook Hall spans the first seven decades of the twentieth century and includes clothing belonging not only to Matilda but to her children, in particular her eldest daughter Frances Dodge, and also her first husband John Dodge and her second husband Alfred G. Wilson. Matilda, who had a preference for simply cut yet perfectly detailed clothing, not only purchased clothes from some of the area's leading department stores such as Walton-Pierce, but also from the finest Parisian couturiers including Callot Soeurs. Designers represented in our collection include Paul Poiret, Henri Bendel, Sally Milgrim, Peggy Hoyt, and Harry Collins and retailers such as Bergdorf Goodman, B. Siegel and Saks Fifth Avenue.

Wayne State University Library System, Digital Library Services Unit ()
This team is charged with creating, integrating, structuring, and ensuring access to digitized information services and resources that support teaching, learning, and research at Wayne State University, primarily by delivery on the Web. The unit has experience in digital projects including audio, video, text and images. Currently it is carrying out the work involved in two IMLS projects including:

"Virtual Motor City" ()
"Making of Modern Michigan" ()

Wayne State University, Fashion Merchandising Program ()
The Fashion Merchandising Program provides a professional concentration in the fields of apparel design and fashion merchandising. In addition to contributing approximately 400 images from their historic costume collection, the faculty from the program will also lead the evaluation phases of this project.

Background: In January 2002 Wayne State University began digitizing the Dorothea June Grossbart Historic Costume Collection. The physical collection contains over 400 garments and accessories from the 19th and 20th centuries, and is managed by the Fashion Design and Merchandising program from the College of Fine, Performing, and Communication Arts (CFPCA). The Wayne State University Library System and CFPCA participated jointly in this project and developed digital images and descriptions for the purposes of historic record, observation, instruction, research and analysis.

The Wayne State University Dorothea June Grossbart Historic Costume Collection website () includes links to a search engine, a short video outlining the steps in the process, and a sample three-dimensional rotational image. (See Appendix B: Sample images and records from this and other partner collections.)
Students in the Wayne State University Fashion Design and Merchandising program are currently using this resource in their course of study. Additionally, the collaboration has been accepted as a panel presentation at the Wayne State University Teaching, Learning and Technology Roundtable 2003 Annual Conference. (See Appendix C: Collaborative Approach to Developing Digitized Collections to Enhance Learning.)

This pilot project allowed us to explore options for digitizing historic clothing. Through the work on the Wayne State collection, we were able to clearly identify issues related to digitizing historic clothing. As a result, we are now well positioned to conduct the larger project proposed in this application.

While working on this proposal IMLS awarded a National Leadership grant to Michigan State University and their partners (including Wayne State University) to fund the "Making of Modern Michigan" (). The "Making of Modern Michigan" provides training to librarians at smaller institutions to create digital collections related to Michigan history. While this particular award did not pertain to museums, we believe that our proposal complements the work being carried out as part of that award. (See Appendix D: Making of Modern Michigan.)

Project Goals: The partners intend to create a resource that is not only "research quality" but also sustainable, scalable, and a national model for web-accessible historic costume collections. The partners are committed to using national standards as guidelines for this project and will follow the six core principles from "The NINCH Guide to Good Practice in the Digital Representation and Management of Cultural Heritage Materials." (See Appendix E: NINCH Core Principles.)

Project goals include:

Goal 1: Create a simple yet powerful web portal that will enhance, expand and promote use of our multi-institutional collection by users of all ages including students, researchers, instructors, and other 21st century learners.

Evaluation will be an important aspect to the project throughout all stages. As such, we are planning on using a variety of methods before, during and upon completion of the project to determine if user needs are being met. The PI, Jeffrey Trzeciak, has attended Outcome Based Evaluation (OBE) training at IMLS and will lead all aspects of the evaluation according to the principles for OBE found at,

If, during the course of the project, we discover through the evaluation phase that users are having difficulty using the resource, we will conduct new user surveys, focus groups and interviews. The results will be shared with the Webmaster who will make alterations to the interface. The interface makes use of cascading style sheets and templates that can easily be altered to accommodate a new "look and feel". In the event that the data is problematic, we will map it to new fields and reindex it. Again, the interface allows us to easily make major modifications as it is based on national standards.

Goal 2: Build a substantial and unique cultural resource that "virtually unifies" the individual collections and provides users with a new means of accessing these resources while preserving and protecting the existing collections for future generations.

The partners will select materials following NINCH principles for selection using the following criteria: (See Appendix F: NINCH Matrix -- Selection for Digitizing)

- Is the material intrinsically valuable?
- Will digitizing the material provide greater access?
- What are my goals in digitizing this object?
- Is there a product that already provides this access?
- Do I have the right to digitize it?
- Can I adequately represent this object digitally?
- Can I affordably digitize it?
- Will it meet the needs of my users?
- How will I manage the digital surrogate over time?

Professional photographers will produce 5,000 digital images. The images will be captured using national image standards including those found in the following appendices:

Appendix G: Image Standards
Appendix H: File Naming Conventions
Appendix I - Quality Control Measures

The partners will then build a catalog of images that will be enhanced with descriptive data built upon national standards found in Appendix J: Metadata. Existing data from a variety of databases will be mapped to Dublin Core and then converted to SGML. The SGML will be indexed and made searchable by users through a simple yet powerful web search engine.

The collection will be made available both nationally and internationally through contributions to OCLC and links from the participating institutions' websites.

Goal 3: Serve as a model of library/museum collaboration in the development of a cultural resource.

The partners will demonstrate how collaboration among libraries and museums can preserve and enhance access to the primary resource material that is so vital to the study of local history and to understanding our local cultural heritage. In addition to serving as a local resource, this project will also serve as a model for a broader national or even international historic costume image repository.

IMLS Priorities

Our project will meet IMLS priorities by

- Creating a cultural resource of exceptional value to scholars, researchers, and the 21st century learner,
- Providing unrestricted universal web access to unique and special collections;
- Preserving and protecting the original artifact by reducing unnecessary handling,
- Protecting the ownership rights and copyright held by the partners,
- Demonstrating standards for bringing decentralized resources together,
- Providing a model for all other organizations considering collaborative approaches to expanding access to their collections,
- Increasing the possibility of future collaborations among our organizations.

ADAPTABILITY - Our project will serve as a model of collaboration for other libraries, museums and educational institutions. We will adhere to proven national standards and follow proven procedures for the creating digital images and metadata. This adherence to standards allows the information from different institutions, often captured originally in different ways and using different systems, to be brought together to form a coherent whole. The digital environment and World Wide Web have already shown value in breaking down geographical barriers to information sharing. Hence, through collaboration and adherence to standards, this project will show how smaller collections or previously unrecognized portions of collections, in physically distinct locations, can be brought together to create a valuable and unique new resource that is universally accessible via the Web. In doing so, we will create a model that other organizations may use when planning both individual and collaborative projects to offer digital access to their collections.

Besides being a model of library and museum cooperation, this project is also a model for digitization and representation of three-dimensional objects. By pooling unique institutional resources and expertise, these museums and libraries will build a valuable collection for students, instructors, researchers and community users. In planning this project, each institution has contributed in substantial ways: the Wayne State University Library System is bringing its technical, cataloging and project management skills to the project; Wayne State University, Detroit Historical Museums, Henry Ford Museum & Greenfield Village and Meadow Brook Hall bring their collections and an extensive knowledge of them; and faculty from Wayne State University Fashion Merchandising Program bring expertise in research, instruction and evaluation.

DESIGN - Currently access to these collections requires making a special trip to each of the partners, and only a small percentage of the material can be on display at any given time. The primary goal is to create, convert, or transfer digitized images and data records from each partner to the web and to make these resources universally accessible through a single common search interface.

The database, including a search engine and website, will be based on a set of tools designed specifically to allow libraries to develop their digital collections. This technology is available through Wayne State University's membership in DLXS, the University of Michigan Digital Library Extension Service (). (See Appendix K -- DLXS Overview.)

Existing data will be exported from proprietary databases (Foxpro, MS Access, Argus, or Filemaker), mapped to Dublin Core, converted to SMGL, and indexed. The DLXS software will provide a web interface to the indexes and the images.

While archival quality TIFF images will be stored on the server, web accessible images will be accessible in two different ways: 1) users will be able to browse all of the available images; 2) users can search cataloging records on fields including title, designer, label, general description, date of origin, condition, etc. (See Appendix G: Image Standards.)

Non-commercial users will have free access to jpg images at 72dpi. Users who wish to use the materials for commercial purposes must obtain permission. A copyright statement will be included in each record.

We explored the option of providing three-dimensional, rotational objects through Quicktime or other technologies. However, such a format is not useful for the study of historic clothing. The user must have the ability to identify and examine detail within each image. Therefore, we have selected a tool that will allow the user to pan and zoom. Sample images can be found in the Wayne State University Dorothea June Grossbart Historic Costume Collection website at . Instead of providing rotational images we will create multiple images from a variety of angles and allow the user to zoom in on details of the individual artifacts.

MANAGEMENT PLAN - The proposal evolved from the overall digital library strategic directions of Wayne State University and the needs of the other collaborating partners. The institutions have committed the human and financial resources necessary to carry out the work of the grant. Jeff Trzeciak, the Director of Library Computing and Media Services, Wayne State University Library System, will manage the project. Jeff will have responsibility for overall management including planning, budgeting, monitoring, and reporting. The Wayne State University Library System will also contribute technical and cataloging skills. Detroit Historical Museums, Henry Ford Museum & Greenfield Village, the Wayne State University Fashion Merchandising program and Meadow Brook Hall bring their collections and an extensive knowledge of them; and Experience Design staff from the Henry Ford Museum & Greenfield Village and faculty from Wayne State University bring expertise in research, instruction and evaluation. (See Appendix L -- Partner Contributions)

Specific techniques are tailored to the unique challenges and opportunities of the individual collections. The project design relies on the expertise Wayne State University gained during the Dorothea June Grossbart Historic Costume Collection digitization project including the adoption of international standards for digital resource management.

The project is planned to take place over a two-year period and will include objectives that may be attained relatively quickly and others that will require ongoing study, consultation and experimentation. A website has already been developed at Wayne State University where information about the project will reside. It can be found at . This website currently includes links to this grant proposal, a list of contacts and links to various prototypes. Once the project is funded, the website will include a timeline, technical information such as server specifications, and a description of the software, image standards, and naming conventions. (See Wayne State University's Dorothea June Grossbart Historic Costume Collection website as an example: http://www.lib.wayne.edu/web/hcc/index.html)
The project is divided into seven phases, each with a "Project Lead" from one of the participating institutions. Interns and onsite coordinators will be involved with all phases of the project.

TIMELINE (See accompanying MS Project Timeline)
Phase one activities related to "project design" have already begun. Jeff Trzeciak is the "Project Lead" for phase one. In this phase we have worked to:

Phase 2 will consist of the initial research and preparation of the artifacts. Patience Nauta will be the "Project Lead" for this phase. She will work with Content Developers (See PERSONNEL), which include representatives from each institution including, faculty advisors, content advisors, technical liaisons, and graduate interns.

This phase will consist primarily of identifying, preparing and selecting the individual articles that will be photographed. Emphasis will be placed on those items that have the most historical significance. They will include items worn by men, women, and children and will be broadly representative of various racial, social, cultural and economic groups in Detroit. They will also include items that may, as a result of excessive handling, be fragile or even damaged. Our intention is to create a resource that allows users to access the collections without the need for physical handling. Approximately 5,000 images will be produced.

We will follow the NINCH guidelines for selection which include answering the following questions:

- What are the aims of the digitization project or program and how can these guide selection, in cases where limited resources prevent you from digitizing the whole collection?
- How do these issues apply to material created in digital form?
- Who will be using these resources?
- What are their needs?
- Do the collections you selected for digitization need special handling?
- What are their characteristics and how do these affect the digitization process?

Phase three activities will consist of data preparation. The "Project Lead" for this phase will be Frances Krempasky, Head of Cataloging at Wayne State University Library System. Representatives from the team of Content Developers will also work on this phase of the project.

Wayne State University will create database structures on the server at Wayne State University. Each institution will have it's own database on the server. Users will be able to search individual collections or will be able to search all databases simultaneously using the same interface. (See Appendix N: Sample Multi-Database Search Screen.) The search interface allows the user to select individual databases or any combination of the available databases. It also allows the user to select from multiple fields including title, creator, subject, description and date.

Any existing data at the member institutions will be mapped to Dublin Core, converted to SGML, imported, and indexed. A sample data-mapping table can be found at: . An intern from the Wayne State University ALA-accredited Library and Information Science Program will be hired to map the data to Dublin Core and import it into the DLXS database.
Where data does not exist, we will train the interns hired for the project to establish metadata records for those items that have not been described.

Phase four activities will consist of photography. Matthew Decker, Wayne State University Library System Webmaster will be the "Project Lead" for this phase. The image standards used in the Historic Costume Collection () will be applied to this project as well. (See Appendix G: Image Standards.)

Professional photographers will be used to digitally photograph the items. We estimate that approximately 950 hours of photography time will be necessary to complete the project. Some items such as gowns or full outfits will require more photography time than simple items like shoes, handbags, and hats.

Phase five activities will consist of data and image loading. Matthew Decker, Wayne State University Library System Webmaster will be the "Project Lead" for this phase. Each image will be assigned a unique name based on the standards applied to the Historic Costume Collection. () (See Appendix H: File Naming Standards.) Images will be saved in TIFF format for archival purposes and will be displayed in JPEG format for web users. Users will be able to resize the images on the fly using MrSID software.

Phase six activities will consist of web design. Matthew Decker will be the "Project Lead" for this phase. In order to seamlessly integrate the project into each partner's individual websites, each will have a search interface designed to fit within their existing web sites. These interfaces would allow users to search that institution's portion of the virtual collection. Each site will be consulted in the design of these interfaces.

However, the interface will also allow searches across images from all of the participating collections. In the final product, users will be able to search all of the collections simultaneously using a common interface and a common language, but will also have the ability to limit their search to the collections of a particular institution.

Phase seven will consist of evaluation, promotion and dissemination. Rayneld Johnson will be the "Project Lead" on this phase. Christina Glengary, Henry Ford Museum & Greenfield Village will also participate.

Evaluation will consist of three elements: (1) Front-end analysis -- We will conduct a survey, focus groups and individual interviews of current and potential users to determine their needs prior to embarking on the project. We will use the Wayne State University Dorothea June Grossbart Historic Costume Collection as a mechanism for soliciting input from users. ()

(2) Formative evaluation - Evaluation will be ongoing throughout the process. During each phase, we will work with users through surveys, interviews and focus groups to determine if we are meeting their basic needs for selection, digitization, and access. Should problems occur either in the data or the image display, we will use tools based on national standards for making changes to the data structure and the image display geared towards identified user needs.

(3) Summative evaluation -- At the end of the project we will conduct surveys, interviews and focus groups to determine if we met our user needs.

BUDGET - The partners are seeking a total of $248,333.00 for a two-year project to develop a digital collection of 5000 images, as described in the project statement above. The partners are contributing $237,554.81 or approximately 49% of the total project costs.

There are four major sections of the budget:

1. $261,632.07- Salaries and Wages for selection, preparation and description of the artifacts. Due to the nature of these items (three-dimensional, complex, often fragile), most of the costs are associated with selection and preparation. The budget also includes:

a. $31,219.96 -- Fringe benefits
b. $2,500 -- Consultant fees for the evaluation phase

2. $48,200 - Services - Professional photographers will be used to capture images of the artifacts. The cost is $46,200 for this service. Additionally, $2,000 has been budgeted for promotion.
3. $50,000 -- Materials Supplies and Equipment -- A portion of these funds will be used to purchase a new server for the project and to license software for the indexing and display of the images.
4. $92,335.78 -- Indirect Costs

CONTRIBUTIONS -The project is a collaboration of three Museums (Detroit Historical Museums, Henry Ford Museum & Greenfield Village and Meadow Brook Hall), Wayne State University Fashion Merchandising Program, and Wayne State University Library System, which will provide overall direction. All members are actively involved in all processes. The partners are committed to providing 44% of the overall costs associated with the project. (See Appendix L: Contributions.)

PERSONNEL

Project Director
- Jeff Trzeciak, Wayne State University, will oversee all aspects of the project. He will have administrative responsibility for managing the grant including establishing priorities, goals and objectives, promotion, financial management and reporting to IMLS. He will work with the management team to cover all issues related to the grant such as metadata development, image standards, web access, data conversion, etc. He will contribute 2% of his time over two years.

Project Coordinator - Adriene Lim, Wayne State University will coordinate the daily activities associated with the project such as training and supervision of the interns, data conversion and loading, server administration, image management, website development, etc. She will contribute 3% of her time over two years.

Webmaster -- Matthew Decker, Wayne State University, will serve both as the "webmaster" and "image lead" for this project. He will oversee the photography and interface development. He will contribute 5% of his time over two years.

Metadata Development, Frances Krempasky, Wayne State University will serve as coordinate all activities associated with metadata development, migration, loading and indexing. She will contribute 5% of her time over two years.

Content Developers -- The content developers will select and prepare the costumes for photographing, will identify metadata needs and will assist in the selection of the images for the website. They include:
Jennifer Lehr, Henry Ford Museum & Greenfield Village (12.5% of her time)
Nancy Bryk, Henry Ford Museum & Greenfield Village (5% of her time)
Lisa Korzetz, Henry Ford Museum & Greenfield Village (5% of her time)
Fran Fraile, Henry Ford Museum & Greenfield Village (1000 hours)
Patience Nauta, Detroit Historical Museums, (10% of her time)
Marianne Weldon, Detroit Historical Museums, (10% of her time)
Brandy Hirschlieb, Meadow Brook Hall (25% of her time)
Pauline Sutterfield, Meadow Brook Hall (5% of her time)

Interns - The project will hire interns from Wayne State University's American Library Association accredited library school to assist with all aspects of the project including selection, preparation, photography, metadata development, data migration, and web development. They will benefit through hands-on experience in developing digital resources.

Evaluation -- Christina Glengary, Henry Ford Museum & Greenfield Village and Rayneld Johnson, Ph.D., Wayne State University will design, develop and implement evaluation mechanisms throughout the process.

PROJECT EVALUATION -An integral part of the activities in all of the Wayne State University digital collection activities is evaluation. In all projects, evaluation is ongoing. Each evaluation includes tasks that are common to all projects as well as those that are specific to the particular collection. Specific measurable outcomes are identified for each phase of the project.
Faculty from Wayne State University and from the Experience Design department at Henry Ford Museum & Greenfield Village will lead the evaluation process. Evaluation methods will include interviews, questionnaires, and focus groups. Quantitative and qualitative data analysis will follow all methodologies. Additionally, Jeff Trzeciak has been trained in outcomes assessment through an IMLS workshop in December 2002.

Outcome 1: Instructors, students and free-will learners who use Digital Dress will find resources relevant to their interests and needs.

Measure: Number and percentage of respondents who identify the site and its contents as relevant..
Data sources: Interviews, questionnaires, and focus groups.
Data collection point: Online surveys will be provided at the end of each site visit. Interviews will be conducted quarterly with students, instructors and free-will learners every six months during the project.
Target: 80% of respondents will rank the resource as "relevant"

Outcome 2: Wayne State University Library and Information Science (LIS) students will have an increased understanding of the issues related to digital projects.

Measure: Number and percentage of interns who identify the experience as relevant to understanding the issues related to digital projects.
Data source: Open ended interviews with interns.
Data collection point: End of year 1, end of project.
Target: 100% of respondents will rate the experience as "relevant."

Outcome 3: Other library and museum workers will have an increased understanding of partnerships related to digitization projects

Measure: Number and percentage of library and museum workers who identify the information about the partnership as relevant and useful.
Data source: Questionnaires, interviews and focus groups conducted at conferences and events.
Data collection point: Throughout the project during attendance at conferences and events.
Target: 60% of respondents who rate the information as "relevant and useful"

Outcome 4: Partners find the experience beneficial and seek additional opportunities for collaboration in order to improve services to our users.

Measure: Number and percentage of partners who rank the experience and "beneficial"
Data source: Surveys and interviews of participating grant staff.
Data collection point: End of project
Target: 100% of respondents who rank the project as "relevant and useful"

Throughout the project we will use the data to identify problems and implement any necessary changes.

DISSEMINATION - Information about the project will be disseminated nationally, statewide and locally. Information about the project will be available in print, on the project website, on the partners' websites, on the statewide "Making of Modern Michigan" website and through digital repositories such as the Association of Research Libraries' Digital Initiative Database and through OCLC. Information about the project will be distributed by the partners through press releases, member mailings, and mailings to various museum/library listservs. Presentations and resource materials about the project as a model for library/museum collaboration will be created for delivery at regional, state and local meetings including Detroit Area Library Network, Michigan Library Association, Michigan Archival Association and the Midwest Archives Conference. Presentations will also be made at national meetings, including American Library Association, and American Association of Museums.

SUSTAINABILITY - Our strategy has five components:

1. Software/hardware migration -- The platform we have chosen includes:

a. Sun Server -- While the grant is for a two year period, we are purchasing a server with a three year lifetime and are contributing the 3rd year cost as part of the overall framework for this collection. The server will be replaced every three years as part of an overall strategy for replacement.
b. DLXS -- The software upon which the interface and indexes are built is based on the SGML standard. The images will be available as JPEG or TIFF. All of our files and metadata are based on international standards that can be migrated to new systems.

2. Physical deterioration of digital media -- Digital images will be archived on DVD's and a copy will be given to the Wayne State University Reuther Archives for cold storage. Backups of the server will be housed at the Wayne State University Computing Center and an off-site location.

3. Metadata -- Metadata will be updated frequently and stored as SGML/XML files and will provide a high level of structure and functionality. It will easily migrate to new systems, if necessary.

4. User needs and preferences -- We will continue to discuss the project with our users after the grant deadline in order to meet their basic yet ongoing needs.

5. The partners are committed to maintaining this collection after the grant period. The database will continue to be available, free of charge, to users after the grant period. Updates to the data and images will provided to Wayne State University Library System when available.