Handbook for Digital Projects - Section IV Selection of Materials for Scanning

Many organizations may be tempted to "just digitize it all" because selection seems labor-intensive and expensive. Might not resources be better spent on digitizing more content? Several compelling reasons for selection indicate otherwise.

Web Access as Publication. When mounted on the Web, digitized collections reach the world's largest and most diverse audience. The Web audience includes scholars, students, the general public, news reporters, donors, collection creators, filmmakers, publishers, and discipline specialists, as well as less desirable users, such as potential thieves, vandals, and intellectual property rights infringers. Once alerted to an organization's holdings, this audience is apt to demand high quality reproductions, publication permissions, and access to the originals. Collection security and manageable policies for duplication, publication, and access should be in place long before the enhanced collection visibility generates increased requests and visitation. Effective preplanning can mitigate the impact of increased requests for access, duplicates, and permissions to publish or exhibit digitized items.
High Costs and Limited Budgets. Most organizations lack the resources to digitize their entire holdings, however desirable this might be to enhance access. The initial costs of selection and digitization pale beside the costs of quality control, metadata production, and indexing/cataloging. When scanning the entire collection isn't affordable, selection becomes not only feasible but essential.
The Digital Mortgage. Digital files come with a mortgage. Each organization must budget to transfer old files to new formats as software and hardware change and electronic media reach the end of their relatively short life expectancies (Puglia, 1999). This budget funds not only a substantial initial investment in digitization but also a digital infrastructure including staff, contracts, equipment, and software. Since digital files are significantly more expensive to manage over time than paper files, organizations must identify and program for substantial resources for the management of the large bodies of digital files they have created (Lowry & Troll, 1996; Marcum, 1998). Program costs don't cease when the Web site appears.
Legal Issues. Organizations often lack the intellectual property rights and permissions to the materials they hold. Copyrights; privacy and publicity rights; and issues of obscenity, defamation, and legally protected location information (federally protected caves, sacred sites, wells, and archeological sites), if not dealt with properly, can lead to lawsuits and costly settlements. Digitization done without thoughtful selection may result in the creation of digital files that can't be effectively used due to legal restrictions (Smith Levine, 1995). Determining the legal status of candidate materials is a crucial step in any digital selection process. Permission-seeking for selected materials begins immediately after selection.

Providing broad access to sensitive materials on the Web without appropriate consultation with specialists, affiliated groups, and donors may result in damage to the original resources (if protected locations are divulged) or to the institution's relationship with its donor community. When selecting materials for Web access, careful and accurate cataloging and contextualization and consultation with the stakeholder group(s) are both wise policies and good manners. Publishing materials on the Web is not equivalent to providing access to the material for a few hundred people annually in a reading room but is closer in impact to a television documentary or newspaper coverage. If there are qualms about publishing something on the front page of a daily newspaper, then the materials should not be put on the Web.

Stakeholder Concerns. Publishing culturally or ethically sensitive materials on the Web may raise a firestorm of protest, even if the collections are not legally restricted. Exercise temperate judgement that balances the need to respect valid stakeholder concerns against a wish to avoid arbitrary censorship. Consult with stakeholders when digitizing the following types of materials:
Cultural information that is normally limited to members of special groups or cultures, such as sacred ceremony, burial, sacred and/or subsistence food-gathering site locations, or similar knowledge that can lead to damage to the resources if widely published.
Donor-restricted materials, such as a request to avoid data release for 25 years after the donor's death.
Documentation. During selection, the committee determines if candidate materials are well documented in terms of individual captions or complete and accurate item-level catalog data. If not, the organization is committing itself to significant additional expense to research the materials, fact-check information obtained, and write appropriate accompanying descriptions that provide a clear and accurate context. In fact, the National Archives staff estimates that up to two-thirds of the actual cost of digitization during the first ten years of a digital project is the creation of metadata and quality control work (Puglia, 1998). Exhibit text is even more expensive, more closely resembling traditional book or exhibit preparation costs.
Institutional Credibility. During selection, the committee checks the accuracy and authority of all information to be included with the electronic resources. The institution corrects inaccurate information to protect its credibility and reputation. Increasingly, audiences using the Web evaluate electronic resources for quality. Common criteria include the authority of an organization, the background and reputation of the individual creator, and the verifiable footnotes or citations included with the materials.

Statistical Sampling

Regardless of how systematic the selection process and how clearly defined the criteria, no two committees select materials in precisely the same way, as no two groups have the same experience, skills, training, and viewpoints. Some historians and scientists prefer to digitize a random percentage of collections, as this sample allows a statistical analysis of collections that is apt to more accurately reflect the full universe of materials. If a significant portion of the expected audience includes scholars who work with sampling techniques, consider incorporating sampling into the selection process.

Although less effective at preserving scholarly context and narrative flow, sampling can be:

More objective
More effective at presenting a cross-section of the whole corpus of materials
Less driven by scholarly trends.

When factoring a statistical component into a digitization project, base sampling on a random number table. Use the random number table to select particular containers and items or combine sampling with a more traditional selection approach. For example, select 5% of a collection using sampling techniques to give researchers a flavor of the whole. Since probably 2/3rds of the sampled materials will be weeded out due to their condition, restrictions, or other concerns, select many more items than are actually required.

The Selection Committee

By its nature, selection is very knowledge- and skill-intensive. A single individual cannot master all the necessary knowledge of subject disciplines, law, conservation, education, technology, and so forth. Most well-managed organizations form Selection Committees to avoid making costly mistakes. Committee members, acting in concert, can evaluate candidate materials judiciously from a variety of viewpoints, disciplines, and perspectives.

Candidates for the Selection Committee include:

Discipline specialists related to the themes being covered in the project
Education specialists, appropriate to the level of the project audience
Digitizing specialists such as photographic and digital laboratory staff
Librarians, archivists, and curators
Researchers experienced in working with online resources
Conservators and preservation specialists
Lawyers.

Even a large Selection Committee may need additional help. Therefore, a wise organization figures out how to tap the knowledge and abilities of the stakeholding communities during the nomination process. (Note: Particularly effective representatives of those communities can also be included on the Selection Committee.)

Why Involve Stakeholders in Nomination for Selection?

If funded by public monies, the collection-holding repository is responsible to each of these groups for how the institution manages, preserves, and makes collections accessible. Private repositories that take no federal or state funds legally may be able to disregard some of these communities. However, ignoring a repository's stakeholder community during selection is profoundly unwise as it affects the institution's reputation and may affect future donations and community cooperation.

Evaluation criteria can be adapted to various institutional settings. The criteria used in this chapter are geared primarily for museum environments.

Stakeholders involved in nominations can include:

Collection creators and donors who have in-depth knowledge of the resources to be digitized
Institutional staff who manage the collection and know the history of usage of the originals and their cataloging and physical condition status
Individuals and groups represented in the collections and their heirs and communities who can decipher hidden meanings and explain cultural concerns about collection access
User communities of scholars, students, educators, filmmakers, reporters, and scholarly publishers who know the market, audience, and secondary users of the project files and can help plan for them
Members of the general public who can articulate community concerns and standards about sharing local knowledge internationally
The middle people, including digital publishers, online order fulfillment services, search engine and technology companies, public television stations, and edutainment companies that have developed innovative strategies for distributing files and leveraging content to obtain funding
Professional colleagues, discipline specialists, and funders who know the content and the institution's role and goals.

The Selection Process

Those involved in selection must familiarize themselves with the goals of the scanning project and discover how the project supports the organization's mission, collections focus, audience, and cooperative strategies. Earlier chapters discuss these important issues.

Pre-Selection Tasks

At the outset, Selection Committee members should meet with other organizations, consortiums, or groups that are managing digital projects. Obtain their project goals and plans. Ask for copies of their audience studies and evaluations. Investigate the selection criteria used by partners and cooperators (e.g., Library of Congress National Digital Library Collection Evaluation Criteria). Examine other projects of the organization, such as exhibits, publications, and courses to determine if their goals and plans suggest a digital component. Building on the expertise of others will speed the development of strong and systematic internal guidelines, policies, and procedures.

Nomination

During the nomination phase, stakeholders and staff members will use forms A and B to nominate materials for evaluation by the Selection Committee. Staff -- particularly librarians, archivists, and curators -- should be involved in nominating related groups of materials for evaluation. When nominating materials, staff should consider the following questions:

How much of the collection is well and accurately documented at the item level in reliable and complete indices and finding aids, and where are these well-documented items?
How much of the collection is in stable or good condition, and where are these stable materials?
What portion of the collection is standard and consistently sized, normal contrast, black-and-white and/or printed materials, and where do these materials fall? Note: Avoid oversized, unusual or varying format, long-tonal range, color, and handwritten materials for start-up projects.
What materials are easy to provide to researchers because of their size, format, or viewing requirements (for example, an 8" by 10" typed document or high resolution and high contrast photographic print original, but not a circuit camera negative or microfilm), and where are they in the collection?
What percentage of the materials does the institution have the copyrights to or licenses for, and where are the public domain materials?
What percentage of the materials has no restrictions or sensitivities of any sort (such as privacy, publicity, defamation, obscenity, and sensitivity, or donor restrictions), and where is this unrestricted and nonsensitive material?
What materials are of highest monetary value and well secured, and where are they in the collections?
What materials are judged to be at highest risk and why, and where are they located in the collections? Of these, which are stable enough to be scanned without damage or which have already been well photographed?
What materials are used most frequently, how are they used, and where are they located?
What materials are unique to the institution, and where are they located?

As noted above, capturing a variety of viewpoints in nominations and deselection at this stage ensures a more balanced and equitable selection process that reflects the full range of user and scholarly interests. Encourage staff, researchers, scholars, committee members, and other stakeholders to use Forms A and B to nominate candidate materials or identify materials that shouldn't be digitized. The forms should be available in the research room for all interested parties who are familiar with the collections. However, inform nominators that the Selection Committee evaluates all nominations and deselection recommendations and makes the final decisions.

Evaluation

Evaluation criteria can be adapted to various institutional settings. The criteria used in this chapter are geared primarily for museum environments.

The Selection Committee will first compare the nomination (Form A) and deselection (Form B) forms to see if any materials appear on both. Mark any materials that were both nominated and recommended for deselection as requiring special attention. The committee works through the questions on Form C for each of the nominated groups of items. If the answer to any one of the questions on Form C becomes, "no, don't digitize," remove the candidate materials being considered until the problem can be resolved. Some materials may qualify for digitization later when certain conditions can be met -- for example, when permission can be obtained from copyright holders, or when copyright protection lapses, or when models or interviewees provide releases. Keep a record of these candidates so that the Selection Committee can reconsider them in the future.

If the quantity of selected items is inadequate for the size of the planned project, consider adding a random sample of the collection, as noted above (Statistical Sampling). For example, if 4,000 items have been selected and there is funding for 5,000, use a random number table to select the final 1,000. Review sampled materials for legal, preservation, sensitivity and other problems just as for other materials. When sampling, select an extra quantity (three times the number needed) to allow for materials that will be weeded out.

Once the Selection Committee has completed this evaluation process all inappropriate materials should be weeded out, leaving only excellent candidates for digitization.

Sample How to Evaluate Collections

Look at each group of nominated items and answer the questions on Form C.

Name: the Quonsethut Academy of Fine Arts in New York

Mission and Collections Focus: Serves Art Historians, Publishers, and Educators in the United States in documenting art history, with an emphasis on landscape architecture and sculpture in upstate New York.

After the nomination period is over, the Academy receives three digitization nominations. The chart below describes the collections and how they were evaluated using Form C. The Status line indicates whether the materials were nominated (Form A) or suggested for deselection (Form B).

Collections Description	Status	Evaluation Decisions
Collection 1: Historic Photographs of the New York Branch of the American Atheneum of Art, 1870-1956, including images by Julia Margaret Cameron, Ansel Adams, and Alfred Stieglitz that document the institutional history. No model releases and the institution holds no copyrights.	Nominated (Form A) by visiting photo historian and staff curators	Selected: Almost 90% of this collection fits the Academy Scope of Collections and Mission. Copyrights of most of the images have lapsed. Permissions will be sought from materials still under copyright protection. As models are no longer living, no model releases are necessary. No publicity rights, E-FOIA, or sensitivities, except for one nude, which won't be digitized. Images are authentic, have associational, evidential, and artifactual values and can be scanned accurately. Collection is well cataloged and can easily be well contextualized due to existing exhibit catalog. Will add value by making very valuable images accessible internationally in conjunction with searchable indices. May avoid digitizing deteriorating glass plates and platinum print components of collection. Originals are well secured and placed in cold storage. Although no one else has digitized these images yet, the Academy's Consortium is interested in sharing the costs since other consortia members have portions of this collection and they can be combined into a virtual museum online
Collection 2: Matias Martin Papers, 1890-1979, contains correspondence with major worldpoets and artists, including six letters (<2% of collection) to the Director of the Quonsethut and Academy of Fine Arts. Most of the pages are very brittle, and many letters done with iron gall ink, leading to rips and tears.	Suggested for deselection (Form B) by lawyers and con-servators	Deselected (Weeded Out): The collection contains high quantities of materials with defamation issues relating to living private individuals. The Academy lacks copyrights, and permissions will be expensive and time consuming to obtain. Recommended for deselection by lawyers and conservators. Reconsider in future (around 2024, when copyrights lapse and poets included are no longer living), as the interest from scholars will be high. In the meantime, funding will be sought to stabilize the collection.
Collection 3: Sylvia Hands-off Collection, 1930-1990, contains the office and personal papers of first curator Hands-off, plus records of the first 50 exhibitions held at the Quonsethut Academy of Fine Arts, including significant scholarly documentation and interpretive notes on exhibitiontopics that relate to the organization's mission. Copyright restrictions on 50% of collection. The papers have active mold in about 35% of the collection, but the estate left funding for conservation treatment. Researchers other than staff have not used it.	Suggested for deselection by an heir, as is felt to cast a bad light on the family	Selected with reservations, to be digitized once the collection has beenstabilized. Precisely fits the Academy's Mission and Scope of Collections. The stakeholder's deselection request is judged to be frivolous and can be renegotiated to address the issues. Most of the collection was done as work-for-hire according to the Academy lawyer, so Hands-off's letters are not covered by copyright. Permissions can be sought from correspondents. No privacy, publicity, E-FOIA, sensitivity issues. Materials have evidential, associational, and informational value, authenticity, and visual accuracy. Materials are well documented and contextualized and will add significant value to already digitized exhibition catalogs and collections documentation. Audience is largely art scholars. The technology will work wellfor these items. Slightly over a third of the collection will require stabilization and cleaning. All items will require item-level cataloging to make them accessible to the international art community. Materials are currently inaccessible and digitizing duplicates no effort elsewhere. This would not qualify as a cooperative project, although it would link to other collections-related work in the Academy itself.

Prioritization by Value, Use, and Risk

If there are too many materials to digitize, prioritize the remaining items by value, use, and risk (Vogt-O'Connor, 1995). These evaluation criteria have been used for several years by many organizations. The committee assigns values to materials and computes totals to assist in this final phase of selection. However, each of these factors must be determined in relationship to the organization's mission and collecting statement, not in isolation.

Value

Materials to be digitized must have one or more of the following values in relationship to the organization's approved scope of collection statement.

Informational value refers to the material's topical content in relation to the organization's scope of collection statement and mission.

High-value collections offer significant information on the key people, places, events, objects, periods, activities, projects, and processes (both natural and cultural) reflected in the collecting statement, thematic framework (if the organization has one), and mission.
Moderate-value collections tell something of the topics and themes (such as the "who, what, where, when, why, and how") reflected in the mission statement and collecting policy.
Low-value collections provide little information about key factors reflected in the mission and collecting statement.

Administrative value refers to the material's functional usefulness to the creating organization on a regular basis, such as the need for architectural drawings for building renovations or vital records for operation purposes.

High-value collections are constantly being used for organization management.
Moderate-value collections are occasionally used.
Low-value collections are rarely reviewed.

Artifactual value, as used by archivists, is the same as intrinsic value and refers to original materials that have value due to their nature.

High-value materials include items in good condition that are rare or interesting objects of material culture. For example, high-value materials include well-composed visual materials, holographic letters with unusual letterheads, or unique diaries; documents in rare historic processes such as platinum prints; materials in unusual genres and formats such as psychic photos or half-plate daguerreotypes.
Moderate-value materials are widely used processes, such as albumen photographic prints, library bound books, or typed letters, and formats, such as stereographs, that are in good condition.
Low-value materials are items in poor condition or copies or duplicates.

Associational value refers to original materials that have a relationship to an eminent individual, place, event, or group, such as letters created, owned, or signed by Thomas Edison or photos taken by or of Civil War soldiers.

High-value materials include such items as the personal papers of a notable individual or group, or those associated with a project like an archeological excavation.
Moderate-value collections might contain some correspondence or portraits of a notable individual.
Low-value materials include copies or duplicates.

Evidential value refers to the documents' ability to serve as legal or historical proof of an activity, event, or occupation.

High-value materials are the originals in an unmodified form.
Moderate-value collections might include some records of legal value, such as birth certificates or legal copies of land records.
Low-value materials are modified records or copies.

Monetary value refers to the current market value of an item. This value may change daily.

How to Score Value

When prioritizing different groups of materials, such as several manuscript collections or series within a single archival collection, or major donations:

Score 6 points if a group of objects has high value in any of the above categories for a significant portion of a collection, i.e., 10% or more (high value).
Score 3 points if a group of materials has less than 10% or no high-value materials, but does have a moderate value in any of the above categories (moderate value).
Score 1 point if the collection has no high or moderate value (low value).

Risk

Risk comes in several forms: legal, social, and preservation. Since legal and social risk will be weeded out during the selection process, this prioritization focuses on preservation.

High Risk. The highest risk materials are primarily chemically unstable, which results in their self-destructing and damaging or contaminating nearby materials, as well as posing health hazards to staff and researchers who use them. Classic examples of high-risk materials are cellulose nitrate negatives and film and materials with biological or chemical contamination, such as mold, insect, and vermin that pose risks of information loss and health hazards. Examples of health and safety risks include materials contaminated with asbestos, Aspergillum mold, and Hantavirus. Other high-risk materials may be self-destructing due to inherent fault (such as iron gall ink, leather bindings with red rot, very acidic and brittle paper, and cellulose acetate film) and those items that may be causing damage to nearby materials (such as materials that have oozing tape).
Moderate Risk. Moderate-risk materials are experiencing primarily mechanical or physical damage due to their housing and handling and the characteristics of their material (e.g., folding strength). Materials that are deteriorating and losing their informational content naturally or gradually due to their component processes and materials are moderate risks. Some examples include electronic and digital data carriers such as CD-ROMs and diskettes; most color slides, negatives, and prints and cellulose-ester based materials (acetate, diacetate, and triacetate); all flaking, retouched, friable, or handcolored images; letterpress books, particularly those with copy pencil inks; carbon copy correspondence; and some tracing paper drawings. Other factors being equal, smaller format materials, such as microforms, should be given top priority as more information is being lost. Also included in the moderate risk category, but of lesser priority, are items with holes, cracks, broken or ripped off pieces, rips, tears, punctures, losses, or those that are warped, folded, creased, wrinkled, cockled, buckled, scratched, abraded, stained, discolored, or otherwise structurally damaged or changing appearance (e.g., color balance shift).
Low Risk. Low-risk materials tend to be the more long-lived processes in undamaged condition and adequate storage conditions. Examples include items with freckle-like stains called foxing; dusty or dirty documents; and slightly faded blueprints and cyanotypes that are well housed in neutral pH materials. Some additional low-risk items might include visual materials that are separating from a mount or support and loose or friable media (such as easily smearable conte crayon, pastel, graphite, or charcoal) that are correctly housed.

How to Score Risk

When comparing groups of materials (for example, manuscript collections):

Score six points if 10% or more of a collection is at high risk per the criteria above. Consider the entire collection as high risk.
Score three points if less than 10% of the collection is high risk. Consider the entire collection as moderate risk.
Score three points if there are no high-risk materials and 10% or more of the materials are at moderate risk. Consider the entire collection as moderate risk.
Score one point if there are less than 1% high-risk materials in the collection and less than 10% moderate-risk materials. Consider the entire collection as low risk.

High-risk collections that are also high value merit digitization when the risk can be minimized or eliminated. The institution may choose not to digitize high-risk collections of low value.

Use

The third factor in determining a collection's priority for digitization is use. High-use materials are those that are requested most frequently for reference purposes by staff and/or outside researchers. If the digital project is geared toward a new audience, past use statistics will not be of much assistance. Determinations will need to be based on predictions of expected use. Talk to repositories experienced in working with the desired new audience, as well as to members of the audience when trying to predict usage. Consider a small pilot project to test audience response before committing to a major new initiative.

Generally, high-use materials have high value. On some occasions, materials of no perceivable value may suddenly become popular because of a particular charm of expression -- for example, a turn-of-phrase in a letter, a quirky angle in a snapshot, or linkage to a previously uncelebrated event or activity. As scholarship changes, the values placed on materials also change. When high use can be predicted and risk minimized, digitizing is a wise access solution.

How to Score Use

Each repository must set its own values for this field based upon reference statistics and visitor logs. To do this, know the institution's usage statistics. Then establish median usage values for a collection. For example, if 10 were the median number of uses annually per collection, then a low use for a collection would be 1-6, moderate use would be 7-13, and high use would be 14-20+.

Putting it All Together:
How to Score and Rank Collections

Value, risk, and use, when considered together and assigned scores (based upon numerical values of high=6, moderate=3, and low=1), indicate the collections requiring digitizing. The key is that each of these factors must be determined in relationship to the organization's mission and collecting statement, not in isolation.

After assigning numerical values to the ratings of value, risk, and use, the committee prioritizes the collections by their numerical scores. In the case of an identical score, compare the usage figures to determine which actually is higher. Also compare the actual types of deterioration to see which is the more threatening to the life of the collection.

In the example following there are two collections with a 15 score. Collection 5 has a larger scale problem with nitrate than Collection 1 has with mold; also the usage of Collection 5 was higher, making it the clear winner in terms of priorities.

Example of Ranking Based on Value, Use, and Risk Name: the Quonsethut Academy of Fine Arts in New York Mission and Collections Focus: Serves Art Historians, Publishers, and Educators in the United States in documenting art history, with an emphasis on landscape architecture and sculpture in upstate New York.
Collections Description	Value	Risk	Use	Score	Priority
Collection 1: Historic Photographs of the American Atheneum of Art, 1870-1956, including images by Julia Margaret Cameron, Ansel Adams, and Alfred Stieglitz that document the institutional history; some with mold (12%), and moderate (or 70) uses annually.	High (6)	High (6)	Moderate (3)	15	3 (has less use and risk than Collection #5)
Collection 2: Matin Femwit Papers, 1900-1989, contains correspondence with major world scientists in the fields of entomology, physics, and mammalogy and two letters (<1% of collection) to the Director of the Quonsethut Academy of Fine Arts. In good condition with a little foxing; relatively little usage as not yet cataloged. Note: Might be given higher priority if linked to a consortia need or outside funding priority that includes funding to do the work.	Low (1)- Does not fit mission or scope	Low (1)	Low (1)	3	5
Collection 3: Sylvia Hands-off Collection, 1930-1990, contains the office and personal papers of first curator Hands-off, plus records of the first 50 exhibitions held at the Quonsethut Academy of Fine Arts, including significant scholarly documentation and interpretive notes on exhibition topics that relate to the organization's mission. Copyright restrictions on 50% of collection. The papers have active mold in about 35% of the collection, but the estate left funding for conservation treatment. Researchers other than staff haven't used them.	High (6)	High (6)	Low (1)	13	4
Collection 4: Quonsethut Academy of Fine Arts Oral History Collection, 1950-1999, includes oral and video histories, transcripts, and release forms documenting major artists, particularly sculptors and landscape architects of the American Northeast. Tapes have not been migrated or refreshed and many are on acetate bases or have flaking binder. Receives high usage, generally above 300 uses a year. Many requests for digital copies.	High (6)	High (6)	High (6)	18	1
Collection 5: Sculptor Tom McMakeitup, 1935-1999, includes the personal papers of New York sculptor and landscape architect Tom McMakeitup, a famed relativist. The collection contains his family, personal, and business papers, including correspondence, films and videotapes, and photographic documentation of his work for the RockePont-Mellon Family at the Kitchie Estate in the Adirondacks. The collection contains about 6,000 nitrate negatives; 4,000 feet of nitrate film; and about 5,000 color dye coupler slides on cellulose ester film. Usage is moderate among filmmakers, art historians, and landscape architects, with about 96 uses annually.	High (6)	High (6)	Moderate (3)	15	2 (Has more use and risk than Collection #1)

Summary of Key Points

The three stages for selection are:

Nominating materials for selection and deselection (stakeholders, the public, staff, and scholars);
Evaluating material s and weeding out materials that aren't appropriate for digitization using uniform deselection criteria (Selection Committee);
Prioritizing the remaining materials based upon the criteria of value, use, and risk to ensure that the most important materials are digitized first (Selection Committee).

Following this process ensures that the digital project is responsive to the individuals who will care about it most and have the biggest stake in its success. By thinking through major issues as materials are reviewed, the committee avoids potential problems from halting the project in mid-stream. Finally, thoughtful selection ensures that the organization spends its funding on the most important, useful, and at-risk items first, and that wise decisions are made.

Form A, Nomination Form for Selection

X Institution Digital Project Nomination Form

1. Materials Being Nominated for Digitization (Please indicate collection number, series, number, box number, folder number, item control number or equivalent and the creator; caption of the item or a bibliographic citation to the fullest extent possible.)

_______________________________________________________________________________________

2. Reason for Nomination (Describe why the materials are important, who might want to use them in a digital form, and what usages are likely if they are digitized.)

_______________________________________________________________________________________

3. Potential Assistance Sources (Please indicate if you have any special knowledge or skills that might be shared with the X repository during the selection process. For example, can you provide caption information, historical background, or are you aware of potential funding sources or digital projects that are covering similar materials to those you are nominating?)

_______________________________________________________________________________________

4. Restrictions (Indicate if you are aware of any reason why the specified materials should not be digitized, such as legal, ethical, or cultural sensitivities. Please be as specific as possible citing a source, such as a law or culture group and a contact name if necessary.)

_______________________________________________________________________________________

5. Your Name: _____________________________________________________________________

6. Your Address: ___________________________________________________________________

________________________________________________________________________________

7. Tel:______________________________________ Fax:__________________________________

8. E-Mail: _____________________________________________

Note:_The Selection Committee will make all final decisions on what will or will not be included in the digital project. If you have any special information you would like to share with the committee, please write it below.

Form B, Nomination Form for Deselection

X Institution Digital Project Deselection Form

1. Identify the Materials That Shouldn't be Digitized (Please indicate collection number, series, number, box number, folder number, item control number or equivalent, and the creator, or caption, of the item to the fullest extent possible

_______________________________________________________________________________________

2.Reason for Deselection (Describe why the materials shouldn't be digitized or shared electronically. Identify problems or concerns that would arise, including legal, cultural, social, or ethical concerns. Identify who might be affected if the materials are available electronically.)

_______________________________________________________________________________________

3. Specific Restrictions (Indicate if you are aware of any reason why the materials should not be digitized by citing specific laws, policies, or equivalent documentation. Please be as specific as possible citing a source, such as a law or culture group, and a contact name if necessary

_______________________________________________________________________________________

5. Your Name: _____________________________________________________________________

6. Your Address: ___________________________________________________________________

________________________________________________________________________________

7. Tel:__________________________________ Fax:______________________________________

8. E-Mail: _____________________________________________

Note: The Selection Committee will make all final decisions on what will or will not be included in the digital project. If you have any special information you would like to share with the committee, please write it below.

Form C, Checklist for Evaluation

Answer each question yes or no. Evaluation Factors:	Yes Digitize	No Don't Digitize
Mission Statement: Does the project fall within the repository or institution mission statement? If not, don't digitize.
Scope of Collections Statement:Do the candidate materials fall within the repository's Scope of Collections Statement (Collecting Policy). If not, don't digitize unless the repository will redefine the policy to include the materials.
Stakeholders' Deselection Requests: Has the repository received requests to select the materials for digitization from a stakeholder or reputable source? If so, are the requests challenged by equivalent requests not to digitize the materials? If so, don't digitize the materials. Note: If you have requests not to digitize that are judged frivolous or insubstantial by the Selection Committee, ignore them.
Donor Restrictions: Is the candidate material unrestricted? If so, digitize. Has the donor or creator of the materials placed substantial and nonnegotiable restrictions on their usage that would prevent them from being digitized? If so, don't digitize the materials. Note: On occasion donor restrictions can be renegotiated.
Copyrights: Is the material either in the public domain or covered by copyright protections that your organization has obtained? If so, digitize. If not, do you have reason to believe that you will be unable to obtain a license to use the materials? If yes, don't digitize until you have obtained copyrights or licenses/permissions.
Privacy Rights: Does the material contain images of living individuals for which you have release forms (particularly for oral and video histories, medical records, personnel records, psychiatric counseling records, or photographs in which the individual is recognizable)? If yes, digitize. If no, do you have reason to believe you can't locate these individuals to obtain permissions or that they won't grant permissions? If yes, don't digitize unless and until you have the permissions.
Publicity Rights: Does your state have a publicity law (e.g., California, Tennessee, New York)? If yes, does your material include images or recordings of famous individuals such as motion picture or recording stars, scientists, artists, or authors (living or dead)? If yes, do you have permissions or licenses to use the images from the individuals or their estates? If yes, digitize. If no, don't digitize until you have permission or licenses.
E-FOIA and State and Local Equivalents: Are you required by law to digitize the candidate materials to meet the electronic Freedom of Information Act, the Paperwork Reduction Act, or similar initiatives? If so, digitize as long as there is no compelling reason why you may not digitize the items that can't be resolved at this time. If there is a compelling reason, don't digitize the materials.
Sensitivity: Does the candidate material contain sensitive information (such as locations of sacred sites, burials, endangered species, fossils, threatened cultural resources [such as petroglyphs], or subsistence food gathering sites), or do the materials nominated present an unbalanced point of view or lack counterpoint perspectives? If so, are the project schedule and staffing adequate to seek consultations and permission-gathering activities from those groups affected and to consult with scholars of various viewpoints? If not, don't digitize the materials or digitize only materials that the committee is fully equipped to evaluate and put into context. Involve stakeholders on the Selection Committee or project staff.
Evidential Value: Is the primary value of the materials evidential, or as legal or historical proof of an action or event? Does the material also have substantial informational and/or associational content of interest to a key audience? If so, digitize. If not, will translating the item from the analog realm to the digital realm so erode the value of the item that it will no longer serve its primary purpose? If so, or if the value is seriously eroded or there is no audience, don't digitize.
Authenticity: Is the item to be digitized authentic and not faked, forged, or altered substantially? If so, digitize. If not, will digitizing the material lend a false authenticity to an inauthentic document or object? Is it impossible to correct the misconceptions through careful contextual documentation, captioning, and metadata? If so, don't digitize. Note: If the project involves substantial altering or retouching of a visual work for purposes other than parody or satire in potential violation of the Visual Artists' Rights Act, don't digitize the materials.
Visual Accuracy: Will the proposed scanning technique be able to capture the appearance of the item accurately? If so, digitize. If not, can the project move to a more sophisticated scanning technique such as color scanning to capture the information? If you can't capture the image accurately, don't digitize the materials.
Documentation:Are the candidate materials well captioned? If so, digitize. If wrongly, poorly, or incompletely captioned, described, and labeled, are the project staffing and budget adequate to provide good documentation within the project timeline? If not, don't digitize the materials.
Contextualization: Does the candidate material require substantial research and a sophisticated and expensive context in order to be useful? If so, can the project provide this context? If so, digitize. If not, will the ability to view the materials serially, but not side-by-side, decrease the value of the files to the audience significantly? If so, can the project provide a way to view materials side-by-side? If not, are there other items within the collection that can be selected instead on this topic? If the files are to be used, must a whole archaeological dig be reconstructed or must an archival finding aid be placed in the Encoded Archival Description (EAD) format or an equivalent effort? If so, are the project staffing and budget adequate to produce this contextualized treatment? If you can't provide the necessary context and the context is judged essential by the Selection Committee, don't digitize.
*Added Value:* Are the candidate materials both valuable and available for the first time? If so, digitize. Does the project add value to candidate materials? If so, digitize. Does the project simply repeat work already in existence in an analog or paper publication (as shovelware)? If so, can the project be reworked so as to add value to the materials by improving access by creating: new audiences for rare or unique materials currently accessible to only a few? linkages to separated materials via HTML, SGML, or XML coding? virtual collections of materials by the same creator; in the same process, media, technique, or format; or other linkage that are otherwise physically separated in real life on a single Web site or CD-ROM? new indices and finding aids that are electronically searchable? new searchability through post-scan processing via OCR or rekeying so textual files are fully searchable? new ways to analyze the originals by techniques, such as microscopic scans, 3-D scans, or similar techniques? usable files for research when the originals are too stained, deteriorated, or damaged for use by retouching or other treatment? If not, are the project staffing expertise and budget currently adequate to producing this new treatment of the material? If not, don't digitize the material until the digitization provides some added value.
Audience: Is the expected new audience for the digital images the same as the existing audience for the originals? If so, will the repository consider recontextualizing the digital product to reach a broader audience? If so, digitize. Will the digital project help reach the same audience more effectively? If not, don't digitize the materials.
Supplementary Selection Criteria: Has the audience set up supplementary evaluation criteria that must be factored into the evaluation process, such as the Teacher Usefulness Criteria developed for the Library of Congress? (EDC) Does this selection accommodate these additional criteria? If so, digitize; if not, don't digitize.
Technology: Does the expected project audience require complex or sophisticated scanning techniques and viewing equipment to use the digitized images as envisioned? If so, is it likely that a sufficient percentage of the audience has this level of viewing technology? If not, replan the project. If so, will textual materials digitized require postscan processing, such as OCR processing or rekeying? Do images require retouching, very high resolution copying, color capture, or extensive coding to maintain linkages and hierarchies? If so, can the work be done within the project budget and timeline, using the project staff? If not, don't digitize.
Condition: Are the candidate items either in stable condition or available as duplicates or copies for use in digitization? If so, digitize. If not, are the candidate items so deteriorated or at risk that it would be difficult or damaging to originals to digitize or copy them? For example, is there a need to disbind a unique scrapbook or rare book, remove items from frames and mats, or place pressure on a cockled and brittle image? If so, is stabilizing the originals too expensive and time consuming to do within the scope of the project budget and timeline? If so, don't digitize.
Control:Are the original items accessioned, described, and placed in secure storage? If so, digitize. If not, would digitizing them place the originals at risk by alerting potential thieves of valuable and vulnerable originals? If so, don't digitize.
Duplication of Effort: Have you checked to see if the items have already been duplicated well elsewhere? If not well duplicated elsewhere, digitize. If digitized elsewhere, is the digital copy made of adequate quality? If so, obtain a copy from the other source and don't digitize the materials.
Accessibility: Are the candidate items inaccessible, such as in cold storage? If so, digitize. If already easily accessible in multiple locations Ñ such as through widely distributed microfilm copies or in many published exhibition catalogs Ñ is there some special reason why digital copies are necessary? If not, don't digitize.
Project: Are the candidate items given priority due to some thematic, cooperative, or grant funding priority? If so, do these priorities fit the institutional mission and collecting statements? If so, digitize. If not, don't digitize.
Cumulation: Is the candidate material a grouped and linked body of materials that draw additional value by being related to other materials held by the repository? Are they already digitized, already selected for digitization, or related to materials already well digitized by other organizations? If so digitize. If a single item and the effort are not for public relations alone or in response to E-FOIA, or are a request by a stakeholder, don't digitize.

Sources

Atkinson, Ross. "Selection for Preservation: A Materialistic Approach." Library Resources and Technical Services 30 (October/December 1986): 344-53.

Bagnell, Roger S. and Carolyn L. Harris. "Involving Scholars in Preservation Decisions: The Case of Classicists." The Journal of Academic Librarianship 13:3 (1987).

Besser, Howard and Jennifer Trant. Introduction to Imaging. Santa Monica, CA: The Getty Art History Information Program, 1995.

Child, Margaret S. "Further Thoughts on Selection for Preservation: A Materialistic Approach." Library Resources and Technical Services 30 (October/December 1996): 354-62.

----. "Selection for Preservation" in Advances in Preservation and Access. Edited by Barbara Buckner Higginbotham and Mary E. Jackson. Westport, CT: Meckler Publishing, 1992.

Council on Library and Information Resources. Scholarship, Instruction, and Libraries at the Turn of the Century: Results from Five Task Forces Appointed by the American Council of Learned Societies and the Council on Library and Information Resources. Washington, DC: Council on Library and Information Resources, 1999.

DeStefano, Paula. "User-Based Selection for Preservation Microfilming." College and Research Libraries, September 1995.

Ellis, Judith, ed. Keeping Archives, 2nd edition. Melbourne, Australia: Thorpe and the Australian Society of Archivists, 1993.

Ester, Michael. Digital Image Collections: Issues and Practice. Washington, DC: Commission on Preservation and Access, 1996.

----. "Digital Images in the Context of Visual Collections and Scholarship." Visual Resources 10 (1994): 11-24.

----. "Image Quality and Viewer Perception." Leonardo:Journal of the International Society for the Arts, Sciences, and Technology, Supplemental Issues (1990): 51-63. Reprinted in Visual Resources 7:4 (1991): 327-52.

----. "Specifics of Imaging Practice." Archives and Museum Informatics, Hands on Hypermedia and Interactivity in Museums. Selected Papers from the Third International Conference (ICHIM/MCN95), San Diego, CA (October 1995): 147-58.

George, Gerald. Difficult Choices: How Can Scholars Help Save Endangered Research Resources? A Report to the Commission on Preservation and Access. Washington, DC: Commission on Preservation and Access, 1995.

Gertz, Janet. Oversize Color Images Project, 1994-1995. Final Report on Phase I. Washington, DC: Commission on Preservation and Access, 1995.

Gonzales, Pedro. Computerization of the Archivo General de Indias: Strategies and Results. Washington, DC: Council on Library and Information Resources, 1998.

Hazen, Dan, Jeffrey Horrell, and Jan Merrill-Oldham. Selecting Research Collections for Digitization. Washington, DC: Council on Library and Information Resources, 1998.

Kenney, Anne R. and Stephen Chapman. Digital Imaging for Libraries and Archives. Ithaca, NY: Department of Preservation and Conservation, Cornell University Library, 1996.

----. Digital Resolution Requirements for Replacing Text-Based Materials: Methods for Benchmarking Image Quality. Washington, DC: Commission on Preservation and Access, 1995.

Lesk, Michael. Preservation of New Technology: A Report of the Technology Assessment Advisory Committee to the Commission on Preservation and Access. Washington, DC: Commission on Preservation and Access, 1996.

Library of Congress National Digital Library Collection Evaluation Criteria prepared by EDC Center for Children and Technology, 96 Morton Street, 7th Floor, New York, NY 10014 (no date, 58 pages, plus bibliography and forms).

Lowry, Charles and Denise Troll. "The Virtual Library Project," in Serials Librarian. ASIG Proceedings: Tradition, Technology, and Transformation Part I. 28:1/2 (1996): 143-70.

McClung,Patricia A., ed. RLG Digital Image Access Project. Palo Alto, CA: Proceedings from an RLG Symposium held March 31 and April 1, 1995.

Marcum, Deanna. CLIR Issues 2 (March/April 1998): 1-3.

National Research Council Committee of Preservation of Historical Records. Preservation of Historical Records. Washington, DC: National Academy Press, 1986.

Ostrow, Stephen E. Digitizing Historical Pictorial Collections for the Internet. Washington, DC: Council on Library and Information Resources, 1998.

Puglia, Steven. "Cost Benefit Analysis for Reformatting Options," Speech Afterimages: Reformatting Visual Materials in a Digital World. Offered by NEDCC and the National Park Service September 17, 1998, at National Archives and Records Administration, College Park, MD.

----. "Creating Permanent and Durable Information: Physical Media and Storage Requirements." CRM 22:2 (February 1999): 25-27. On the Web at http://www.cr.nps.gov/crm.

Smith, Abby. Future of the Past: Preservation in American Research Libraries. Washington, DC: Council on Library and Information Resources, 1999.

----. Why Digitize? Washington, DC: Council on Library and Information Resources, 1999.

Smith, Levine, Melissa. "Electronic Publishing: A Legal and Practical Primer." CRM 18:9 (1995).

Vogt-O'Conner, Diane L. "Chapter 2: Legal Issues." Museum Handbook, Part III: Access and Use. Washington, DC: National Park Service and GPO, 1998.

----. "Is the Record of the 20th Century at Risk?" CRM 22:2 (February 1999): 21-25. On the Web at http://www.cr.nps.gov/crm.

----. "Preservation Reformatting: Selecting a Copy Technology." Conserve O Gram 19/11. Washington, DC: National Park Service and GPO, 1995.

----. "Reformatting for Preservation and Access: Prioritizing Materials for Duplication." Conserve O Gram 19/10. Washington, DC: National Park Service and GPO, 1995.

Vogt-O'Conner, Diane L., Virginia Kilby, and Joan Bacharach. "Chapter 1: Evaluating Museum Collections for Use." Museum Handbook, Part III: Access and Use. Washington, DC: National Park Service and GPO, 1998.

----. "Chapter 3: Publications" Museum Handbook, Part III: Access and Use. Washington, DC: National Park Service and GPO, 1998.

Waters, Donald J. "Archiving Digital Information" in Tomorrow's Access--Today's Decisions: Ensuring Access to Today's Electronic Resources Tomorrow. Proceedings of the Fourteenth Annual Conference of Research Library Directors. OCLC and the OCLC Research Libraries Advisory Committee, Dublin, OH: 1996.

----. From Microfilm to Digital Imagery: On the Feasibility of a Project to Study the Means, Costs, and Benefits of Converting Large Quantities of Preserved Library Materials from Microfilm to Digital Images. Washington, DC: Commission on Preservation and Access, 1996.

Waters, Donald J. and J. Garrett. Preserving Digital Information: Final Report and Recommendations of the Task Force on Archiving Digital Information. Washington, DC: Commission on Preservation and Access, 1996.

Weber, Harmut and Marianne Dorr. Digitization as a Method of Preservation? Washington, DC: Commission on Preservation and Access, 1997.

Table of Contents

Northeast Document Conservation Center
100 Brickstone Square
Andover, MA 01810-1494
Telephone: (978) 470-1010
Fax: (978) 475-6021
http://www.nedcc.org

Last Modified: September 12, 2005

Copyright 2000. Northeast Document Conservation Center

The Selection Process	Who does it?	When?
1. Nomination of Materials: A. Nomination of What to Digitize (Form A)B. Nomination of What Not to Digitize (Form B)	Collection Stakeholders and Staff	At the start
2. Evaluation of Nominated Materials: Review Selection, and Weeding of Materials from the Nominated Materials (Form C)	Selection Committee	During formal review
3.Prioritization of Remaining Materials: Ranking of Materials based on Value, Use, and Risk	Selection Committee	During formal review

Handbook for Digital Projects:
A Management Tool for Preservation and Access

IV

Selection of Materials for Scanning

Introduction

Why Select for Scanning?

Statistical Sampling

The Selection Committee

Why Involve Stakeholders in Nomination for Selection?

The Selection Process

Pre-Selection Tasks

Nomination

Evaluation

Sample How to Evaluate Collections

Prioritization by Value, Use, and Risk

Value

How to Score Value

Risk

How to Score Risk

Use

How to Score Use

Putting it All Together:
How to Score and Rank Collections

Example of Ranking Based on
Value, Use, and Risk

Summary of Key Points

Form A, Nomination Form for Selection

X Institution Digital Project Nomination Form

Form B, Nomination Form for Deselection

X Institution Digital Project Deselection Form

Form C, Checklist for Evaluation

Sources

Handbook for Digital Projects: A Management Tool for Preservation and Access

IV

Selection of Materials for Scanning

Introduction

Why Select for Scanning?

Statistical Sampling

The Selection Committee

Why Involve Stakeholders in Nomination for Selection?

The Selection Process

Pre-Selection Tasks

Nomination

Evaluation

Sample How to Evaluate Collections

Prioritization by Value, Use, and Risk

Value

How to Score Value

Risk

How to Score Risk

Use

How to Score Use

Putting it All Together: How to Score and Rank Collections

Example of Ranking Based on Value, Use, and Risk

Summary of Key Points

Form A, Nomination Form for Selection

X Institution Digital Project Nomination Form

Form B, Nomination Form for Deselection

X Institution Digital Project Deselection Form

Form C, Checklist for Evaluation

Sources

Handbook for Digital Projects:
A Management Tool for Preservation and Access

Putting it All Together:
How to Score and Rank Collections

Example of Ranking Based on
Value, Use, and Risk