Association of Research Libraries (ARL®)

Alan F. Lewis
Subject Area Expert—Audiovisual Preservation
Special Media Archives Services Division
National Archives and Records Administration

When I think about risk to a collection of machine-based AV archival materials, I think about:

• loss, damage, or destruction of the physical item, and also
• loss of access to the content of the recording, the information content.

Hence, our risk reduction thinking needs to have a wider scope than just the stuff on the shelf in the back room. As I view risk reduction, then, I try to think about the recording media as parts of an AV recording system, because without the system, the item on the shelf is of little use. Reducing risk, then, means reducing the risk to all the components of the system.

My assigned topic is to discuss care, handling, and storage:

• To me, care and handling means understanding the various recording media and their matching equipment in a collection and using appropriate techniques to eliminate or minimize potential damage to those media and to the equipment when the former, the recorded media, is off-the-shelf and being used and the latter, the equipment, is turned on and in operation.

• Likewise to me, storage means understanding the various recording media and their matching equipment in a collection and using appropriate techniques to eliminate or minimize potential damage to those media and to the equipment when the former, the recording media, are in containers and on the shelf and the latter, the equipment, is turned off and dormant.

So to understand these things, permit me to do some basic training first and then briefly review my “19 Conservation Concerns.”

Machine-based audiovisual recording systems are composed of three elements. The first and most obvious, because it’s the “stuff” on the shelf, is the recording media. It is the physical item that has been used to fix in some [hopefully] permanent way the sounds that are the collection, whether those sounds are music, spoken words, or natural sounds. For audio recordings, the media have had a variety of shapes: cylinders, flat discs of various thicknesses, endless belts, wires on spools, and ribbons of material running from reel-to-reel in the open or enclosed in housings.

The second element of a recording system is the equipment, the devices that initially captured and later can retrieve the sounds that have been fixed on the recording medium. Over the century plus that sound recordings have been in existence, the equipment has used transponders to convert sound waves into either mechanical energy or electrical energy. That energy, in turn, drove styli connected to diaphragms or electromagnets. Edison’s original acoustical cylinder machines, for example, used a horn-diaphragm-stylus-soft cylinder as the technology.

The third element is the standards that were developed as a part of the invention of the system. They specify all the details of how the signal passes through the technology, onto the recording medium and later, how they are retrieved in usable form. Again, using the example of a cylinder machine, the standards involved the dimensions of the cylinder, its speed of rotation, the relative softness of the surface to be incised, the number of grooves per inch that the device would cut across the surface of the cylinder, etc., etc.

Now, because I think it is very important to understand the recording media, let me get into a bit more detail. A typical audio recording medium is likely to have at least two of the following components:

• base or substrate
• a signal capturer
• a linker

It is important to understand that each of these components may have natural enemies which would include their own built-in seeds of deterioration, sometimes called inherent vice in the archives field, as well as unnatural enemies, the conditions we subject the media to over time: its use, overuse, and misuse.

The first component is the base or substrate. It is the physical foundation of the recording medium. It typically has physical size (dimensions) and a shape. It is a cylinder, a disc, a wire, or a tape and generally the term we use to describe the medium is the physical form of that substrate. (“How many discs do you have in your collection?”)

The second component is the information capturer, the technical system by which the information is transformed from sound energy in the air into whatever means is used by the system to fix it on the medium: variable grooves for cylinders and disc recordings and electromagnetic impulses for the magnetic media.

The third component, lacking in some recording media, is a linker, the physical or chemical means by which the information capturer is secured to the substrate. In original phonograph records, one-off instantaneous discs, the linker is the bond between the cellulose lacquer layer into which the grooves are cut and the base material which might be cardboard, aluminum, glass, etc. With legacy magnetic tape materials, the linker is typically a urethane plastic binder. (Modern mass-produced phonograph records have no linker because the grooves are impressed right into the substrate. Likewise, magnetic wire has no linker because the wire itself, the substrate, becomes magnetized.)

Bear in mind that because sound recordings come in so many types, I can’t go into any real detail about any of them in the short period of time I have. However, they include:

a. acoustically made mechanical recordings
b. electrically made mechanical recordings
c. magnetic wire recordings
d. magnetic recordings on paper, cellulose, and polyester base tape

With all of these, knowledge may get more difficult because there are original, one-off recordings and also mass-manufactured products. The bottom line is that if humans created these technologies, the standards and the recordings, something can and will go wrong with them over time. What can go wrong are what I call the “Attacks Against Recording Systems.”

First, with the recording media, there is their own inherent vice, the deterioration factors that are basically manufactured into the media because they weren’t manufactured for the long term. One might think of it as natural deterioration or natural aging because most media were designed for short-term, commercial use with little or no thought for their long-term keeping qualities. Secondly, because these media are handled by people and by machines, there is the wear and tear of overuse and mishandling by the uninitiated, the careless, or the mean-spirited.

Attacks against the technology begin with the real life fact of commercial obsolescence of technologies. Technologies give way as newer systems come on the market that are more desirable and therefore more acceptable. (Consider, for example, how computer removable storage technology has gone from 8” floppy discs to 5-1/4 inch floppies to 3-1/2” floppies to CDs.) Consider, too, that these technologies we use were more often designed for production and/or distribution purposes without thought to their long-term availability. Finally, there are the costs and ultimately the impracticality of equipment upkeep as technicians and spare parts become unavailable and as skilled operators retire or die.

Standards (and software if we’re in a digital domain) are also subject to obsolescence as new versions are developed and the old ones are consigned to the audiovisual scrap heap. In cases where new versions are marketed, the question of backward compatibility arises as version replaces version. Consider with grooved discs alone, how standards have changed with cutting inside-out to outside-in tracks, rotation speeds from 16-2/3 rpm and 78 rpm to 45 and 33-1/3 rpm (to say nothing of half-speed mastering!), and groove pitch from old standard (100 grooves/inch) to microgroove’s (200 grooves/inch).

Finally, there is the legitimate concern about quality loss through signal compression that discards information (lossey compression) and even what can happen with compression that doesn’t dispose of material (lossless compression) but may leave audible artifacts as a result of the electronic processes.

In summary, in my view, risk reductions starts by knowing the details about the media, the technology, the standards, and the software, and paying attention to those details.

Now, let me move on to my “19 Conservation Concerns” because whether handling the media or storing it, these are relevant. Again, because of the complexity of the audio recording field, much of this needs to be general and not medium- or system-specific.

+Environment. Conservators in all fields, not just AV media, cite the temperature and humidity conditions in which heritage media are stored as the single most critical factor in their long-term survivability. If materials held by an archive are not to be at-risk, proper levels of temperature and humidity are required. Because of the diversity of sound recordings to be discussed in this conference, I cannot make any single recommendation but rather suggest that each archive consider the variety of media it holds, do the research on a medium-by-medium basis, and design storage environments that are appropriate for each. I will say that at present, the U.S. National Archives holds its wire, acetate, and polyester magnetic media and its mechanical media at 65 deg. F @ 30% RH. The environmental control system must operate “24/7/365” and must be monitored and recorded continuously in order to have proof-of-performance of the system.

• Physical Security. It is a given that archives have the responsibility to provide various kinds of protection for their accessioned materials. Risk reduction, therefore, requires protection from theft, vandalism, and damage as well as protection from unauthorized duplication and use. In a legal sense, risk reduction might also be expanded to include making sure to observe and honor the copyright of recordings, enforcing donor and privacy restrictions, etc.

• Fire Protection. Allied to physical protection is eliminating the risk of fire through proper selection of the archive site as well as having a fire prevention program in place, an effective fire detection and warning system, a fire suppression system that minimizes collateral damage, and having a recovery plan in place to minimize loss to media, equipment, and administrative records after an emergency.

• Water Protection. Also allied to physical protection is reducing the risk of water emergencies by site selection to defend against acts of nature, eliminating all overhead piping other than possibly a water fire suppression system, and maintaining a sound building envelope, especially roofs, windows, and skylights. A water detection system should be installed and a water emergency recovery plan should be in place and include provisions for dealing not only with the media but also the equipment, the finding aids, and the administrative records. Some additional practical hints include not using basements or attics for storage, building above the calculated 100-year flood level, training the staff not to store media on the floor, using storage furniture with the bottom shelf at least four inches above the floor level, and providing floor drains.

• Light Sensitivity. Other than new media discs that use light sensitive dyes, I know of no audio media that are particularly light sensitive. On the other hand, ultraviolet radiation from sunlight and artificial lighting sources are known to be problematic to paper labels, adhesives, and inks. Reduced level lighting also has a positive effect on electric utility costs, both from the cost of the illumination itself as well as the cost of removing the heat caused by electric light sources that increase the heat load the environmental system has to remove.

• Cleanliness. Another logical risk reduction step is to store and use sound recordings and their equipment in clean locales for two very good reasons: the presence of foreign matter on the playing surface of any audiovisual medium produces a reduced quality playback, and the introduction of dirt and other physical debris into the equipment also gets on the media that pass through it. Hence, common sense good housekeeping measures should be in effect in all collection, storage, handling, and use areas.

The traditional “no smoking/eating/drinking” rules always apply. Unpacking collections should take place away from storage and use areas so their “street dirt” isn’t introduced into those areas. Hard surfaced flooring rather than carpeting is desirable to make cleaning easier. Surface dusting and floor cleaning should take place on a regular basis and should be conducted without solvents or any other substances that have not been tested and found archivally acceptable.

• Air Quality–Pollutants. Related to cleanliness is the matter of keeping gaseous pollutants out of storage and use areas. Since many audio archives are located in urban areas, their natural air supply is likely to contain sulfur and nitrogen compounds and ozone. Man-made materials used in building components, in insulation, and in furniture can introduce formaldehyde. Cellulose acetate products bring the product of their own deterioration, acetic acid, into the space. Risk reduction factors should therefore include selection of the site for the facility, careful selection of materials to be used in it, sophisticated air filtration systems, and measuring and tracking pollutants on a regular basis.

• Biological Infestation. We often observe mildew problems with sound recording collections, especially those that have been stored in less than optimum environments. We usually don’t think of such collections as being at risk from insect or rodent infestation. In reality, though, our audio recording media and/or their packaging are subject not only to the “no-see-um” spores that bring mold but also the “macro-critters” like insects and rodents that thrive on the cardboard packaging, paper labels, and adhesives.

Risk reduction techniques include using proper temperature and humidity levels for the preservation of the materials which, coincidentally, are levels that do not provide a hospitable environment for the critters. Pre-storage inspection and fumigating collections may be needed—but careful and knowledgeable fumigation so that the process to rid the collection of the problem doesn’t damage or destroy the collection itself! Good housekeeping, of course, will remove the other enticements of food or water that attract the pests. Finally, if there is a problem, use an integrated pest management system that first identifies the specific problem and seeks to solve only it. This avoids introducing broad-spectrum pesticides that may be more than are needed.

• Strategic Dispersal. Strategic dispersal sounds like a Department of Defense term, but in audio risk reduction terms, it simply means that if there are multiple copies of an item, do not store them together. The point is, if there is a destructive emergency situation, hopefully one or the other copy will survive. Dispersal might mean putting copies on opposite sides of the same room, in different rooms, in different buildings, etc.

• Primary Containers. In practice, there are as many different types of containers on the market as there are recording media and there are variations based on manufacturing cost. All too often, the primary container that accompanies the purchase of blank stock or a new pre-recorded item, is not designed for the long-term keeping of the item. Again, because of the many audio recording media we deal with, time does not permit me to go into detail about any one container. There are considerations to be given to design, construction and materials used, the chemistry and stability of that material so it is non-reactive with the medium, and whether it is equipped with media support devices. I prefer containers for all media that have a positive closing device, an adequate writing/labeling surface, and a design that resists water penetration. A one-piece item with its body and cover permanently attached is best. If the container is plastic and the collection is a large one, some local fire codes may require a flame retardant be used in the material to lessen the danger to firefighters. Toxic gases are released from some plastics when they burn. Finally, some media, especially acetate materials, may do better in a ventilated container than in a tightly closed one.

• Storage Position. Proper positioning for audio recording media has long been recognized as an important risk reduction activity. The risk is generally related to the force of gravity pulling downward on media over long periods of time. For the linear media, wire recordings seem to pose no problem whatever their orientation is in relation to the force of gravity. Tape, on the other hand, should be stored vertically in order to protect the edges of the medium from deformity and edge damage problems caused by pressure against the reel’s flanges.

Discs are traditionally stored vertically to prevent warping and to prevent stacking too much weight in a pile of discs and causing breakage in the lower ones. However, in my opinion and based on observing what I believe to be stress fractures in some shellac-type pressings, the archivist should consider the various types of discs in a collection and make reasoned positioning decisions based on understanding the nature and structure of individual types of items. Glass-base instantaneous discs are very different from vinyl pressings and some should and could be stored vertically and some horizontally.

• Winding Ribbon-Like Media. The winding of recording media is relevant only to the linear media like magnetic wire, open reel tape, and tape in cassettes and cartridges. With wire, there is a risk that during play or shuttling from spool-to-spool, it will slip over a flange or backlash resulting in what one laboratory person describes as ending up looking like a bird’s nest. For the tape media, the results of poor winding are possible snapping of acetate tape and stretching of polyester tape.

One school of thought on tape winding for storage suggests playing the tape through from end-to-end and leaving it unrewound. Assuming the machine is properly adjusted and aligned, this should result in a smooth and uniform tape pack with even tension throughout. Leaving the tape in this tails out orientation means that should there be print through of the signal, that problem will be audible as an echo with the ghost signal following the strong signal. If left heads out and print through develops, it will be a precho (this is a made-up word, a contraction for “pre-echo”) with the ghost signal heard before the strong signal which is more distracting to the human ear and human mind.

A frequently asked question has to do with the need for periodic winding of linear media like wire and tape. There seems to be as many schools of thought as there are audio archivists and their opinions include:

• Do periodic rewinding every “x” number of years because tape is designed as a flexible medium and it should be “exercised.” This approach might mean instituting a program to rewind the entire collection over, say, five years and therefore doing 20% of the collection each year so that in the five-year cycle, everything is rewound. A subset of this would be to do a visual inspection first of that 20% to see if there is any evidence that winding is necessary.

• Another school of thought suggests, “If it ain’t broken, don’t fix it,” meaning leave it alone. This thinking may be based on the fact that any running of tape on a machine potentially puts the tape at risk from handling, wear and tear, and machine malfunctions.

• Yet another thought is to leave the tape alone unless there has been some major environmental change in the storage area that might have caused the tape to expand or contract due to temperature and humidity changes which resulted in changes in the tension of the tape packs.

• Shelving. Another aspect related to physical protection has to do with the shelving used for the storage of media. Risk reduction associated with shelving include using archivally acceptable materials, shelving that is robust and proper for the size and weight of stored media, adjustability so shelf heights can be varied as needed, shelving that does not encourage climbing, and shelving that does not allow “hiding places” behind the upright structural members. Local or institutional building codes must be observed or even exceeded and even if not required, earthquake and anti-tip devices should be installed.

In new facilities or facilities to be renovated, consideration should be given to compact shelving in order to maximize the amount of material that can be stored per square foot in the expensive to construct and expensive to operate environmentally controlled and protected vault space. Don’t forget about floor loading, because compact shelving significantly increases it!

• Shock/Vibration Protection. Another obvious subset of physical protection of media, and therefore eliminating risk, is protection of the materials from shock and vibration. This includes a generous helping of staff training so they do not consider these sometimes physically heavy valuable historical records, as “stuff” and handle it as such. Potential damage includes causing breakage of the recording media itself, problems with reels and cassette shells, broken containers, and the concern that major shocks might cause disarrangement to magnetic particles.

• Magnetic Protection. A fortunate characteristic, or unfortunate depending on one’s viewpoint, of magnetic media is that information recorded on it is erasable by the introduction of new magnetic information or by subjecting it to a stronger magnetic field. This was the marketing virtue of magnetic audio media: it could be erased and reused; but it now becomes the curse of magnetic media historians and archivists, the loss of potentially valuable recordings in order to reuse tape stock. Hence, the risk is purposeful or accidental, partial or complete, erasure. Risk elimination includes staff and user training, disabling the record function on equipment if practical, disabling those media that have an anti-recording interlock, and protecting the media from strong electromagnetic fields.

• Item Identification. The risk here is item misidentification and the potential loss of the item physically in the storage area or intellectually in the description and cataloging scheme. A rule of thumb in dealing with the situation of a recording in a container that has another number on it is to look for the container that matches the number or name of the “miscontainered” item. In most cases, what has happened is that the two recording media and their cases were inadvertently switched by some previous archivist, technician, or user.

With regard to intellectual content, there are established rules and procedures for describing audio recordings that should be used. It is possible that in some kinds of collections, especially those containing unedited or untitled items, a name or number will have to be provided by the cataloger using a system that matches the system already in use with the collection or the institution. Whatever naming system is used, it is clear that individual media and individual containers should both carry that identification, and the marking system used should be archivally acceptable.

• Inventory Control. Small collections, especially those with little access activity, require little in the way of a fancy inventory control system. As collections grow, however, the problems and the risks associated with loss or misplacement of items becomes a factor that needs to be considered. Traditional archival procedures often involve assigning records series designators and numerical sequences within series and such procedures may work with some types of audio collections. Be warned, however, that complexity continues to grow as the archive accessions multiple production generations of the same item and related media that may require different storage environments. Multiple copies, accessioning related materials out of numerical or chronological order or over a span of time, different sized storage containers, and possibly different storage positioning requirements, all add to the mix.

Consequently, in my opinion, the answer is to take a page from modern warehouse practice and develop a random location storage system. Such a system is one that takes into consideration the variable factors of environmental needs, size, positioning, etc. and assigns items to the next available shelf space that meets the requirements. Computer software can be developed or may already exist that can take into account all those variables. Bar codes or other automatic identification systems (AIS) can be used to identify the items as well as identify the shelf locations so the entire process can be virtually error free because it does not depend on frequent keyboarding of information.

• Equipment. It is obvious that equipment is necessary to deal with machine-based AV media. Without equipment the archival functions of appraisal, media inspection, storage preparations, description, reference, user service, content preservation (reformatting), and customer reproduction needs cannot be served. As risk reduction measures, there are technical requirements for the equipment: it must be technology that matches the media formats, it should be high quality professional equipment, operators for the more sophisticated technologies must be available, and the equipment must be maintainable. The options are for the archive to become an operating machine museum or finding and contracting with an operating machine museum, usually a commercial, professional duplication house.

• Personnel. As with any and all human endeavors, the success of the activity depends on the selection, training, and dedication of staff members and the leadership they are given. As a function of risk reduction, they need to be trained in the legacy media they are custodians of, trained in the new media of the users, and trained and knowledgeable in the collection they are responsible for. In addition, and not necessarily associated with risk reduction, other keys to an effective audio archives staff is to insure that they are trained in customer service, something much in vogue in management circles these days, and “collaterally trained,” knowing about similar collections elsewhere in order to provide a higher level of customer service.

In summary, then, audio collections are like some people’s lives: full of risk. It is our job as collections managers to know the risks to our media, our equipment, our standards, and our software, and to take steps to reduce or avoid them. Risk-reduction may be a full time job in your shop!

This paper represents work carried out for a federal government agency and is not protected by copyright.