How the tapes were digitised

Process

The reel-to-reel tapes located in the Roberto Gerhard Archive at the Cambridge University Library were digitised with a reel-to-reel tape player.

Prof. Adkins worked with two research assistants for three months, five days a week, 9 hours a day to transfer the contents of the physical media to a digital format.

The archive is not chronologically ordered. The larger tapes were stored in a large, unorganized collection, while some collections of the smaller tapes were collected on tape spikes or grouped in bags. The organization of this collection in this way was likely undertaken by Roberto, and some might have been undertaken by Poldi. As such, there was no way to follow the second principal of archiving, the principal of original order, since it was not clear what the original order would have been.

It was determined to collect each tape brand together, under the idea that these might have been purchased around a similar time in the composer life. Nonetheless, given Gerhard’s tendency to reuse and rerecord over tapes, this was only a half-solution to an unanswerable problem. Nonetheless, tapes were given unique identifying numbers so that they could reliably be referenced in the collection.

Cataloguing

The ID of each tape followed the following format:

CUL_OR01_Gerhard_000101_L01

Wherein CUL referred to Cambridge University Library, OR01 refers to Open Reel Collection No 1 (in case further tapes were found later), and Gerhard refers to the fact that this is a recording from the Gerhard Archive. The next four digits (0001) refers to the tape number assigned to the resource. So this would range from 0001 to 0610. The following two digits refers to the tape number in that resource. If a resource was a collection of several tapes (either multiple tapes in the same box, bag, or spike) then this number would increment for each item.

The final element referred to the Left (L) or Right (R) channel of the recording, and the take of that recording. Many tapes were mono but several were stereo. As such, each tape was recorded in stereo to capture left and right audio information incase they were unique.

Some tapes were recorded at different speeds, and it was often difficult to determine exactly what the recordings were intended to sound like. In cases where the speed wasn’t clear based on notes left by Roberto or Poldi, or where it clearly changed at different parts of the tape, the tapes were recorded more than once, each time at a different speed. This ensured that digital artifacts weren’t introduced into the recordings by relying on adjusting the speed of playback through digital means, and ensured that the most accurate information was captured for the archive.

Over to you

How might you have approached this problem? What unique issues do you think you might have encountered when attempting to digitise an archive collection of this size?