This is the last of a three part series describing the creation of the Challenger collection, a project to digitize and make available primary source materials related to the Space Shuttle Challenger and the disaster that occurred on January 28, 1986.
The collection can be accessed here.
The final step in creating the Challenger collection was to upload the material to the repository. We first needed to create the appropriate structures on the repository to hold the material. Each type of material required a slightly different structure. Once the structures were created we began to make the material available to the public.
Our repository is hierarchical. The Challenger collection is under the larger heading of “Centers” (for the Christa Corrigan McAuliffe Challenger Center for Education and Teaching Excellence. Other headings include, for example, the Henry Whittemore Library, and the Academic Affairs Office. The Challenger collections itself is broken down into two categories, called series. We created one series for the images and one series for everything else, which we titled “Documents”. The Document series was further broken down into three series: Correspondence, Ephemera, and Reports.
There were two methods to upload the items to the repository: individual or batch. Individual uploading involved fewer steps but was more time consuming. We chose this method for the correspondence, of which there were three, and the governmental reports, of which there were two. The second method, a batch upload, allowed for multiple items to be sent to the repository at once, but required more steps. The first step was to store the material where it could be downloaded to the repository (in contrast to the individual method, where the material could be uploaded directly, the batch-uploaded material needed to be hosted elsewhere. We chose Dropbox for this purpose). The next step was to download a spreadsheet specifically created for the task of a batch-upload, populate it appropriately, and re-upload the spreadsheet to the repository. If all was done correctly a confirmation email would arrive once the material was successfully sent to the repository.
In order to make sure the material is findable to the public, we had to ensure the correct metadata was associated with each item. Unfortunately the batch-upload method was not very efficient for doing this. While the batch-upload spreadsheet contained cells for metadata, with larger sets of files it became difficult to associate the correct metadata with each file. For this reason we chose to individually add the appropriate metadata to each batch-uploaded item after it had been posted to the repository. The process of adding metadata was time consuming but indispensable to a functioning repository.