I want an easy way to understand which materials have been digitized within the context of an archival collection


*Current status: *
In the context of the current global pandemic, researchers in many archives are entirely remote; in the context of other times, most researchers cannot come to our reading room in order to view records.

Simultaneously, our staff have been taking heroic measures to digitize materials and associate them with archival description via digital object records.

But if I say “oh, sweet, let’s see which materials in the Third World Women’s Alliance records have been digitized”, there’s no easy way to do this. A user must click on each archival object in a collection to discover whether a digital object is associated with it.*

Digital objects are tied to archival objects and the network of archival description, but they aren’t accessible that way in search. If I do a search and the record that was digitized doesn’t have the same title as the collection, I can’t know that they’re related. If I want to only see digitized materials, I can’t have the same discovery experience that I do for other archival records.

  • There is a hack available – because ASpace indexes the full json object, if I search for “digital_object” within a collection, the archival objects associated with digital objects in a collection will appear in a search.

*Why this is a problem: *
Digital materials are hidden in the context of normal search and browse within the finding aid. Users do not know that these materials are immediately available.

*Ideal outcomes: *

  1. From a collection view, I want to be able to click a filter called “digital materials available” or similar and get a list of archival objects. I want for my results list to give the breadcrumbs of where these materials are in the context of the finding aid. See this example from the Princeton finding aids site.

  2. I would like a clear signpost – an icon or a badge that says “digital materials available” or similar – whenever archival object records with associated digital objects are displayed. This may be in a search or a browse context, within the “collection organization” view, and in the tree view of the hierarchy (and anywhere else I may not have mentioned – if there’s a digital object, tell us!!!!). Duke’s implementation of ArcLight does a very nice job of this.

Philosophy and wider consequences:
Ideally, ArchivesSpace should create a digital archives experience that mirrors the sensemaking experience of being in the reading room. The finding aid provides context, content, and explanations of the relationships between historical records, agents, and activities.

Currently, because digital objects are not easily searched and found, many repositories maintain parallel systems of item-based digital libraries. This means duplication of metadata, duplication of systems, duplication of effort, and no single system of record for archival description. Perhaps just as crucially, we lose the power of context in these digital library systems. A file of materials called “communist propaganda,” for instance, has very different meaning if it is found in a judge’s personal records series than if found in the discovery files for a case, even if the informational contents are exactly the same. A researcher only seeing the information wouldn’t understand the role that these records held in the collector’s life and work. Without context, historical meaning is lost.

Another ticket for another time, but the “digital object” browse is really a problem. Digital objects should not include archival description. That’s what resource and archival object records are for. The requirement for a title in the digital object record creates enormous opportunities for the duplication and mis-match of information. Instead, digital object records should be the analogues to top containers, and should point to the place where the materials being described in the archival object can be found. This would improve usability of the staff site too. Having 100,000s of digital objects independently discoverable, often with meaningless titles and without context is super-confusing.


Nancy Kennedy
November 17, 2020, 5:36 PM

This would be a big help! I’d advocate for implementing this in the Staff User Interface too. Staff would then be able to search for digital_objects within not-yet-public records, for example. It would also be helpful to have this option, even if not using the PUI.

Maggie Hughes
7 days ago

Dev Pri reviewed this ticket in the March 2021 meeting. The first ideal outcome listed should be part of a larger discussion related to Digital Objects, and so we are changing the status of this ticket to Awaiting More Information. It was decided to move forward with the second ideal outcome, so this has been split off into a new ticket: . Please edit/comment/watch updates to that work via ANW-1206.




Maureen Cresci Callahan