When I teach the Society of American Archivists’ DAS Courses “Arranging and Describing Electronic Records,” participants spend a lot of time on exercises intended to help build knowledge about the concept of the archival information packet. The reason we do that is because I believe that once you understand the elements of the packet, you have a much better idea what the end product of your arrangement and description efforts should look like. Each repository’s packet will look a little bit different since (1) the tools and technologies it uses are different, and (2) DACS allows a lot of room for local practice and decision making. Over the past few weeks, I’ve been grappling with the need to revise and standardize an AIP will allow the University of Illinois ingest over 5TB of electronic files (both born digital and digital surrogates) into our library’s repository infrastructure. Here is what I came up with [updated May 6, 2014]:
Collection Level Descriptive Information: Each collection gets a collection level descriptive record in our Archon database (will be migrated to ArchivesSpace), with the e-records or digital surrogates described in linked digital object record.
Digital Object Record: Strategically, we decided that at the most basic level, there will be one digital object record describing the entirety of all digital materials for a particular collection:
Content Data, Preservation Description Information: All digital content and PDI is kept in a single folder in our staging server (10TB share managed by campus academic computing): I say staging server, because we are prepping this content for deposit to the “Medusa” digital preservation system. [updated: in practice, we have not used the manifests folder, since preservation metadata will be generated by our digital preservation system.
- Root folder is the nine-digit unique ID for the collection to which it is linked.
- Preservation folder holds holds disk image or whatever was turned over to us.
- Access folder holds copies of any files put into access system, after they have been weeded and arranged. Different internal structures are used for born digital materials and surrogates, will explain more in upcoming post.
- contents of each sub-folder in access folder described in a digital object record in Achon. More on this in a subsequent post.
- pdi folder contains (1) checksums for preservation and access copies, from Karen’s Directory Printer and (2) any other documentation regarding accession, provenance, etc.
Access: In true MPLP fashion, access is provided to that portion of the material suitable for broad distribution, via PHP ‘directory-browser’ script that I branded as our “E-records Browser”: http://archives.library.illinois.edu/e-records/index.php?dir=University%20Archives/1519030/
In addition, complementary materials not suitable for posting online, but not restricted are accessible in ‘nearline’ location. Users can either access them in the archives or request access, by contacting the archives.
In the next few posts, I’ll say a bit more about how this fits into our “MPLP” approach to processing and providing access to electronic records.