The National Audio-Visual Conservation Center. In 1998, the Library of Congress established the National Audio-Visual Conservation Center to oversee the preservation of and the provision of access to the Motion Picture, Broadcasting, and Recorded Sound Division's (M/B/RS) collections of motion pictures, video recordings, and sound recordings. The Center will open in 2005-06 at a facility in Culpeper, Virginia, and has been generously supported by the Packard Humanities Institute.
The Library preserves works by copying ("reformatting") when the works are deteriorating or are at risk of deterioration, at risk of damage by wear and tear produced by viewing or listening by researchers, or in formats that are obsolete or obsolescent. One of the key functions for the new center will be to implement a new paradigm for preservation reformatting and apply this approach to items identified as preservation priorities. As is explained in more detail in the Appendix, the system that enacts the new paradigm will store, manage, and provide controlled access to digital bitstreams (typically stored as computer files) in a server-based computer system.
Priorities for the Center's Future Preservation Program. The recorded sound and moving image collections in the custody of the M/B/RS Division include materials originally produced in the United States and in other nations. The division has made a preliminary classificaion of 150,000 items as priority items and the need to preserve them will be an important factor driving the Center's digital program. The Appendix characterizes these priority categories for preservation.
Prototyping the Future Preservation System. In order to develop the new approach for preservation and research access, the M/B/RS Division will prototype future digital preservation systems during the years 1999-2004 that will model portions of the systems to be installed in the new Center in Culpeper in 2005-06. The project will define future system requirements, define specific technical formats for the copies to be produced, determine general questions of feasibility, and develop some software elements that will be used in the future system. The planners anticipate that the Prototyping Project will treat a few thousand items, representing from 500-1,000 hours of sound recordings and from 25-40 hours of video recordings.
Listening and viewing access during the prototyping period. Delivery to researchers at the Library. The prototype system will deliver content to research workstations in reading rooms in Library of Congress buildings. From three to six high-quality researcher workstations on Library premises will be established to receive all content in "broadcast quality." Additional moderate-quality researcher or staff workstations on Library premises will receive all content in but at lower levels of quality.
Delivery of selected content off-site. Public domain content or content for which Library receives appropriate authorization by owners or interest holders may be made widely available via the Internet, e.g., through programs like American Memory (the Library's project to provide educational access to historical collections). The system to be developed for the prototyping project will limit access in appropriate ways.
What capabilities will the researcher have? Although researchers will be prevented from making copies of recordings in the reading rooms, they will be provided with very flexible playback capabilities:
The Increased Importance of Security in the Context of Digital Reformatting. As the planning for the prototype proceeded, the M/B/RS Division considered its special responsibility to secure its collections. The responsibility is on behalf of the U.S. Congress and the American people, for whom the collections are acquired and maintained, and to protect the interests of third parties when the materials are protected by copyright, donor conditions, or other restrictions. It is worth noting that the reliability of the Library's film archivists in years past has contributed the motion picture industry's support of the Library's role as the keeper of America's national film collection.
In regard to this responsibility, two issues of long standing emerge with increased force in a digital context: (1) theft or unauthorized copying and (2) the archivists' special control of special collections. For many years, these concerns have been addressed in appropriate ways in the traditional analog environment. For example, patrons of the Library must have a photo identification to use the collections and access to the stacks is forbidden except to authorized individuals. No recording devices are permitted in the reading rooms that provide access to recorded sound and moving image collections. In all Library reading rooms, warnings are posted on all patron photocopiers reminding readers of the applicability of the copyright law of the United States (Title 17 U.S.C.).
Since many observers believe that the digital environment increases the risk of theft or unauthorized use and may dilute the archivist's ability to maintain reliable control of collections, the Library plans to take the steps described below as it develops its digital prototype.
Reference staff will not permit copies of audio and video to be made. Researchers will not be permitted to make copies of the recordings. The prototyping activity will explore the use of tools that can be placed within the systems used to prevent unauthorized copying of digital files or the unauthorized transmission of digital files to another site. Reference staff will be continue to be vigilant (as they are in today's analog environment) to prevent researchers from attaching unauthorized equipment, e.g., audio or video cassette tape recorders, to the workstations they are using for their research.
Although researchers will not be permitted to make electronic copies of still images associated with items in the collection, e.g., record labels, phonograph record jackets, still frames from moving image items, they will be permitted to print paper copies on typical computer printers. As noted in the next section, the workstations and printers will be associated with warning about the rights of owners and with other informational material describing and promoting compliance with the copyright law.
Policy statements, informational material, warnings, and copyright notices. Policy statements, informational material, and warnings pertaining to copyright and other restrictions will be posted in reading rooms and near researcher workstations and will also be incorporated in the online presentations viewed by researchers as they carry out their work. The technology to be developed for the prototype will investigate the feasibility of including a textual statement or legend with each printout that reminds users that the work may be protected by copyright and that (without additional permission), the paper copy is for personal research only.
Any copyright notices that appears in or on an item will be included when the item is reproduced. In many cases, the Library's bibliographic information includes relevant information pertaining to the copyright of an item, e.g., the registration number and identification of the producing organization or publisher. This bibliographic information will be associated with the item in the digital research system.
Computer System Security. The prototyping project will explore a variety of security features for the systems to be used for storage and access. This will include the typical array of elements found in many advanced server systems and firewall structures, and will give special focus to the examination of the exercise of control by specific user groups within the Library. User-group-level control is a significant consideration because it is an expression of archivists' custodial responsibility.
Investigation of Watermarking or Equivalent Technology. During the prototyping project, the Library will investigate and, if feasible, test the application of watermarks or other identification marks to the research-access copies it produces. Because of the risk of degradation of quality, no such marks will be added to preservation copies.
There are two outcomes to the various watermarking schemes available in commerce today. First, some schemes (especially for audio and video) are designed to prevent unauthorized users from recopying an item. Second, some schemes (especially for still images) facilitate the identification of the source of an item and to provide a means to discover information about the item. For example, the coded watermark could signal that this copy or phonorecord came from the collections of the Library of Congress and provide the means to call up the current bibliographic record and/or finding aid information for that item, which would identify the work, and which may provide publication and/or copyright information.
During the prototyping period, researchers will not have the ability to make copies of the new digital recordings. They will be able to make lists of selections. If possible, the playback system will enable researchers to use elapsed time, SMPTE timecode, and the like to capture the numerical representation of start and stop points of segments of interest within programs. In effect, the automated system will provide patrons with the ability to frame an order for duplication of certain files, and will provide them with a list of numerical "pointers" to relevant passages within a selection.
Notwithstanding the proposed prototyping limit on researchers making copies for their personal research, the Library notes its future interest in discovering an approach agreeable to owners that would allow bona fide researchers to make or download their copies for their own study purposes.
The execution of a request for a copy or phonorecord will follow the same procedures that are in place at the present. The procedures include the following elements, as described in handouts currently used by the M/B/RS Division.
From Procedures for Phonoduplication (Recorded Sound Section, M/B/RS, dated 4/99):
What will happen in the future Audio-Visual Center at Culpeper? The M/B/RS Division will manage the production of preservation copies and the provision of access for researchers in keeping with traditional practices and in consideration of the outcome and findings of the prototyping project.
A New Paradigm for Preservation and Priority Categories for Preservation
Background: Why Is a New Approach to Preservation Needed?
Shortcomings of current approaches to preserving audio and video. Current preservation reformatting schemes for audio and video content employ magnetic tape media that lacks desired levels of permanence. Preservation tapes must be made and remade using manual operations. Because of anxieties about digital tape formats like RDAT (a specific type of digital audio tape) and CD-R, the Library's audio preservation copies still take the form of analog recordings. Each time an analog recording is copied, it suffers generation loss, a slight but irreversible degradation in quality. The Library is concerned about the obsolescence of conventional (analog and digital) recording tape stock and the equipment needed to record and play these tapes.
A new paradigm for audio visual preservation . In order to address these shortcomings, the Library plans to develop a computer-file-based mode for the preservation of recorded sound and video moving image collections. For audio today and video in the near future, computer files can contain signals at quality levels high enough to qualify as preservation copies, are susceptible to automated management, and are coming into widespread use in the broadcasting and sound-recording industry. However, computer files-perhaps more accurately the bitstreams contained within files-present a new problem: will the organizations that manage them will be able to ensure their persistence over the long term.
The paradigm for the preservation of computer information has a very low dependency on media. This is in marked contrast to the classic paradigm for library preservation, an approach that features media that endure for long periods, e.g., microfilm. The computer specialist's preservation assumptions include the idea of short-lived systems and media and the correlative idea that content must be migrated to new systems and media in a routine and systematic manner. The computer specialist also understands that some complex data forms may not be migratable and, in order to continue their use, the obsolescent or obsolete systems or environments in which they function must be emulated. In a study sponsored by the Preservation Directorate, the consultant William D. Storm outlined ways in which the migration of audio and video materials into a computer-data environment could address the problems associated with conventional reformatting of these original formats. Storm's report is titled Unified Strategy for the Preservation of Audio and Video Materials (Preservation Research and Testing Series No. 9806; Aug. 1997; rev. Nov. 1998), available from the Research and Testing Office.
It is worth noting that although the new technology represents a paradigm shift, the underlying purpose for making the copies remains constant and unchanged from past practice: to preserve the audio and video collections of the nation held by the Library of Congress.
Protection of archival materials against damage. Research access to motion picture films, phonograph records, and other materials presents its own preservation problem: the act of handling, viewing, or playing the Library's archival copy introduces the risk of scratches or other physical damage. This problem may be compared to the wear and tear suffered by microfilm prints in a reading room, where successive passes through a microfilm reader add numerous scratches, folds, and tears. Although in some cases the Library has been able to produce a viewing copy or phonorecord (like a spare print of a microfilm), the high cost of reproducing, say, a feature film in its entirety rules this out for a significant portion of the collection.
When the Library has a copy of a film in the form of a videocassette, laser videodisc, or a DVD disk, researchers are strongly encouraged to view the video copy instead of the film, in order to protect the film copy from handling. When commercial videocopies of wide-screen titles are cropped, however, researchers may request the film print for viewing in order to permit viewing the full image. Cropping typically consists of the removal of the left and right portions of the original image in order to fill the video screen, whose aspect ratio is not wide screen. Cropping does not occur when the video copy is "letterboxed."
Priority Categories for Preservation
Deterioration priority. Some collections of recorded sound and video materials are a priority for treatment because they are deteriorating. For example, many audio tape recordings from the 1970s and 1980s suffer from what is called "sticky-shed" syndrome. On a reel of tape afflicted by sticky-shed, the signal-bearing coating on one side of the tape has absorbed moisture and sticks to base material on the other. When played, the coating is liable to shed, leading to the loss of information and the clogging of tape player mechanisms.
Obsolescence of format priority. Some collections are a priority because they are recorded on obsolete or obsolescent media. For example, the Library holds approximately 20,000 television programs from the 1960s and 1970s recorded on 2-inch quadruplex tape. Videotape recorders that play 2-inch materials are no longer manufactured and stores of parts are very scarce. Even more recent formats such as 1-inch Type C and 3/4-inch U-matic are now obsolescent. The Library's experience and anxieties are shared by many other organizations, as reported in A Study of the Current State of American Television and Video Preservation (Library of Congress: Washington, 1997; ISBN: 0-8444-0946-4).
Difficult storage priority . Some collections are a priority for treatment because they are challenging to store, handle, and preserve in their original form. Audio cassettes are an example: the thin, narrow tape they contain is liable to stretch or break, and their internal mechanisms (often cheaply made) make them prone to jam. The quantities received make the audio cassettes a special problem: copyright deposits add thousands of ephemeral sound recordings to the Library's collections annually.
Prospective preservation reformatting. In time, all audio-visual materials will deteriorate or become obsolete. For example, newly produced CD-ROM disks have an estimated life of from 5 to 100 years, depending upon the particulars materials and manufacturing methods used and upon the lifespan of CD reading devices in the marketplace. Thus prospective preservation copies will be made of some materials. Generally, this will occur when a researcher comes to the Library's reading rooms on Capitol Hill and requests playback of an item stored at Culpeper. Advance notice will be required and a copy will be made to meet the researcher's requirements. Since the level of effort to make a listening copy is the same as to make a preservation copy, preservation copies will be made on those occasions.
Digital copies as they pertain to motion picture films. Digital copies of motion pictures will be produced to serve as reference surrogates and eliminate the risk of damage to prints, just as videocassettes and disks serve this purpose at present. In order to produce the digital research copy using current technology, the film must first be copied to videotape. When the Library has a videodisc or videocassette copy that is used for research viewing, the digital reference surrogate will be produced from this video copy instead from a new Library-produced video copy.