Digital Archives And Collections

A digital archive is not simply a pile of scanned paper.

8 frameworks, from 1965 to today · Digital Humanities · Humanities

still a working view ended · each leads to its entry below

Overview

A digital archive is not simply a pile of scanned paper. It is a set of files, descriptions, rules, and services that together let people find, trust, and reuse cultural material over time. The subfield of digital archives and collections is the part of digital humanities concerned with building and sustaining those systems: converting and describing materials, keeping them readable for decades, capturing things that were born online, and making collections usable for research and the public. What is at stake is memory itself. A library that loses its catalogue, a website that vanishes when a company folds, or a community whose records are described in someone else's categories all raise the same question: who gets to decide what survives, and on whose terms?

From Machine-Readable Catalogues to Electronic Records

The prehistory lies in library automation after the Second World War. Libraries needed to share catalogue records, and in the 1960s the MARC format (Machine-Readable Cataloguing), developed under Henriette Avram at the Library of Congress, encoded bibliographic description so computers could exchange it1. MARC made the card catalogue into data and established a principle that still governs the field: description is a structured, standardized layer separate from the object described1.

A parallel strand came from government and business. As offices replaced paper with computers in the 1970s and 1980s, archivists confronted records that had no physical carrier to shelve. Electronic records management grew out of this: the idea that records must be captured, classified, and scheduled for retention or destruction from the moment of creation, with their context intact. The University of British Columbia ran a major research project on the preservation of electronic records in the early 1990s, and national archives in several countries began issuing guidance. The hard problem was not storage but evidential trust: a database record could be altered invisibly, so archivists argued for audit trails, fixed metadata, and documented provenance.

The Digital Library Turn

In the 1990s the web turned digitization from a specialist activity into a mass program. The US Digital Library Initiative, funded from 1994, joined computer scientists and librarians to build large testbed collections; Project Gutenberg, begun by Michael Hart in 1971, had already shown that volunteers could produce a vast text corpus. National libraries followed: the Library of Congress's American Memory had begun delivering digitized collections online in the mid-1990s, and the Bibliothèque nationale de France launched Gallica in 1997. These projects established the basic workflow that persists today: select, image, describe, serve.

The description layer became the field's central technical concern. Metadata—literally data about data—had to record who made an object, when, of what, and under what rights. The Dublin Core element set, agreed at a 1995 workshop in Dublin, Ohio, offered a minimal fifteen-element vocabulary simple enough for non-specialists. Richer standards followed for particular domains, such as EAD (Encoded Archival Description) for finding aids and METS for packaging digital objects. Because institutions used different schemas, the problem of metadata interoperability arose: how to make records from one system legible to another. Solutions included crosswalks (mappings between schemas), harvesting protocols such as OAI-PMH (the Open Archives Initiative Protocol for Metadata Harvesting, released in 2001), and later linked data, which identifies entities by persistent web addresses so that a person or place can be referenced consistently across collections.

Preservation as a Discipline

Digitization exposed a problem paper did not have in the same way: digital objects decay silently. Files become unreadable as formats and software disappear, and storage media fail. Digital preservation emerged as a discipline with its own concepts. The Open Archival Information System (OAIS) reference model, published in 2002 and later standardized as ISO 14721, described the functions any trustworthy repository must perform—ingest, archival storage, data management, access, and preservation planning—and gave the field a shared vocabulary. Practical techniques included migration (converting files to current formats) and emulation (recreating the original computing environment so old software can run)2. The LOCKSS program, developed at Stanford from the late 1990s, and the Dutch-led Portico, founded in 2002, built distributed systems in which many libraries hold copies, so that no single failure destroys a journal. Preservation was reframed from saving objects to maintaining access over time.

Capturing the Web

The web created a new kind of collection problem: material that exists only online and changes or disappears constantly. The Internet Archive, founded by Brewster Kahle in 1996, began systematic crawling and by the early 2000s offered the Wayback Machine for viewing past versions of pages34. National libraries built their own programs, often constrained by legal deposit law that had never anticipated websites; the International Internet Preservation Consortium, founded in 2003, coordinated tools and standards. Web archiving raised distinctive questions: what to select from an effectively infinite space, how to capture dynamic pages and embedded media, and how to represent a crawl faithfully when the live site is gone.

Community Archives and Cultural Protocols

A significant correction to the field's institutional bias came from community archives: collections built by and for the groups they document, rather than about them. Grassroots projects—ethnic, LGBTQ, labor, and Indigenous—argued that mainstream archives had often described communities in outsiders' terms or excluded them altogether. The response was participatory: communities deciding what to keep, how to describe it, and who may see it. This connected to Indigenous data sovereignty movements, which hold that communities have the right to govern data about themselves. Protocols such as the First Archivists Circle's "Protocols for Native American Archival Materials" (2006) and later frameworks like the CARE Principles for Indigenous Data Governance (2019) set out expectations for consultation, attribution, and restrictions on culturally sensitive material. The Mukurtu platform, developed with the Warumungu community in Australia from the mid-2000s, implemented culturally specific access rules in software—showing that ethical protocols can be built into systems, not just stated in policy.

Collections as Data

As digitized corpora grew, researchers wanted to compute over them rather than read them one item at a time. Computational reuse of collections—sometimes called "collections as data"—treats a digitized archive as a dataset for text mining, image analysis, or network analysis. This required new work: releasing bulk data with clear rights statements, documenting provenance and gaps, and providing interfaces for machine access. The HathiTrust Research Center, established in 2011, gave scholars computational access to millions of digitized volumes while respecting copyright through a "non-consumptive" research model that returns analysis rather than full text. The field's practitioners increasingly argued that collections should be designed for reuse from the start, and that the biases of digitization—what got scanned, whose language dominates—must be documented as part of the data.

Where the Field Stands

Today the main approaches descend directly from these strands. Repository systems and standards such as OAIS and PREMIS (a metadata standard for preservation events) underpin institutional practice; web archiving is routine for national libraries and increasingly for research projects; community and Indigenous protocols have moved from the margins into professional ethics codes and funding requirements; and collections-as-data work has become a standard expectation for major digitization programs. The open questions are live ones. How should archives preserve and describe machine-generated and algorithmic material? Who bears the cost of perpetual preservation when institutions close? How can computational access be opened without overriding the access restrictions communities have deliberately set? And as generative systems begin to produce records of their own, the field faces the question it has always faced in new form: what counts as a document, and who decides.

References

  1. MARC standards (introduction, History), Wikipedia
  2. Digital preservation (Migration, Emulation), Wikipedia
  3. Web archiving (History and development), Wikipedia
  4. Internet Archive (introduction, History), Wikipedia

The frameworks of Digital Archives And Collections

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1965-Present

    Metadata Standards and Interoperability

    How can catalogues be exchanged between systems?

    Fly to its tower in Nousfarer

  2. 1970-Present

    Electronic Records Management

    How are digital records kept trustworthy over time?

    Fly to its tower in Nousfarer

  3. 1990-Present

    Community Archives

    Who decides what a community's records say?

    Fly to its tower in Nousfarer

  4. 1994-Present

    Digital Library

    How are collections digitized and served online?

    Fly to its tower in Nousfarer

  5. 1996-Present

    Digital Preservation

    How do we keep digital files readable for decades?

    Fly to its tower in Nousfarer

  6. 1996-Present

    Web Archiving

    How do we capture material that exists only online?

    Fly to its tower in Nousfarer

  7. 2006-Present
  8. 2011-Present

    Collections as Data

    How can collections be used as computable data?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Metadata Standards and Interoperabili…Electronic Records ManagementCommunity ArchivesDigital LibraryDigital PreservationWeb ArchivingIndigenous Data Sovereignty and Cultu…Collections as Data

Beyond the field

Where Digital Archives And Collections's ideas came from in other fields, and what they led to there: the strongest 12 of 56 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Digital Archives And Collections is a planet in Nousfarer

Its city has a tower for each of these 8 frameworks. Land, walk in, and climb one concept at a time.

Land on Digital Archives And Collections

Related fields

Fields its ideas are linked to

More in Digital Humanities

See where Digital Archives And Collections sits on the Map of Human Knowledge