Library of Congress

Digital Preservation

The Library of Congress > Digital Preservation > Feature Series > Digital Preservation Pioneers > John Spencer

Back to Digital Preservation Pioneers

John SpencerThe recorded music industry faces a staggering digital media storage problem, one that will ultimately result in loss of revenue and assets. And this problem has nothing to do with file sharing and copyright; it is caused by the unstructured way in which digital files are managed from artist to label to archive.

An artist used to deliver analog multitrack audiotapes and end-product masters to the label, accompanied by detailed studio logs that listed information about the recording sessions and song tracks. That process has changed. Today, artists deliver born-digital content on storage media such as CDs, DVDs and hard drives. The source tracks may be incomplete or scattered across different media. Studio logs, if present, might consist of handwritten metadata, which the label will have to manually reenter into its databases.

The obvious short-term goal is to meet shipping deadlines, both physical and electronic. However, once the projects are moved to the vault, there is no guarantee that they can be easily recovered in the future or that the label actually will have all of the individual tracks needed to remix or repurpose the recording.

Over a decade ago, John Spencer, who has been in the music recording business for more than 25 years, began to see the scope of the problem clearly. He has been working since then to help find a solution for this problem. He explains, “With the convergence of technology used in the recorded music industry, professional recording equipment manufacturers going out of business and vaults filling up with disparate carriers, something has to be done.”

John is the president of BMS/Chace – an archival data management and migration service – and an active member of several national audio engineering and recording committees. BMS/Chace is a partner in the National Digital Information Infrastructure and Preservation Program (NDIIPP) as part of the Preserving Creative America initiative. John does not side with the artist or the label; he just wants to get the best processes in place to enable all parties to benefit from their creative work. Referring to the music industry’s current digital-preservation efforts, he says, “I would say we provide a layer of triage to new projects; we’re helping to stop the bleeding. It is time to put some basic tools out there, to organize metadata and files in a structured fashion.”

An early step came in the form of a comprehensive report, published in 2003, titled Recommendation for Delivery of Recorded Music Projects (external link) (PDF). The report was written by the Delivery Specifications Committee, which was formed by members of the Producers & Engineers Wing of the Recording Academy, of which John is a member. The report defined a recommended file format and recommended storage carriers, in support of long-term preservation and sustainability. The report continues to be updated every six months.

The Delivery Specifications Committee, however, cannot predict what will last; it can only track current trends and make its best recommendations. “We keep our fingers on the pulse [of the enterprise IT industry] to determine trends that shape storage best practices,” said John. For example, since the first report was published, a number of proprietary formats have been removed. Music created on obsolete formats must now be migrated to open source files or risk being lost.

The recommendations advise against proprietary hardware and software whenever possible because it is impossible to know how long a company will support that format or even still be in business. John advises, “For future use and practicality, digital preservation files must be vendor-agnostic.”

Another of the committee’s recommendations is to save recorded music in the Broadcast Wave Format (BWF), a well-documented, software-independent nonproprietary WAV file format. This means that, for the foreseeable future, the committee anticipates that BWF files will be playable on most computers and will not require special proprietary software.

One of the most rampant IT problems in the all-digital recording industry is the lack of comprehensive metadata and specifications for how that metadata is coupled to the content. BMS/Chace is researching ways to standardize metadata generation – creative, technical and business-related metadata – to create and maintain key information about a recording throughout its life cycle.

The ideal metadata tool must be unobtrusive and easy to use. The metadata, collected by content creators and users, would be saved in a standardized XML file defined by the BMS/Chace NDIIPP project. Record labels could then use that metadata to streamline the collection and management of their recordings.

John said that the same digital-preservation challenges exist for both small and large music recording companies. “There are distinct parallels between small and large labels, and even small institutional archives. No one currently has an easy way to collect standardized metadata.” But John is relentlessly upbeat about overcoming the challenge, which he says is to “find common ground where creative people can continue to do what they do best, and ultimately hand off these precious assets to the technical folks to preserve for the future.”