Ephemeral to enduring: The Internet archive and its role in preserving digital media

Abstract

The Internet was first created as a communications network, but it has produced myriad effects in many sectors of technology, business, and society. Among important changes are the growth of electronic publishing and the increasing use of the Internet as a medium for disseminating diverse documents, literature, art, journalism, commentary, and miscellaneous expressions of information and thought. Digital media, and the rise of Internet distribution of it, has caused profound changes in librarianship as well. ********** In spring 2001, it was estimated that more than 93 percent of all new information is born digital--originating in digital form. (1) But just as the Internet, particularly the World Wide Web, is noted for its growing capacity and diversity of material, it is also notorious for its impermanence. The average life span of a Web site is approximately six weeks. (2) Lyman cites that 44 percent of Web sites online in 1998 were not found the following year. (3) How to preserve digital media has been a growing concern of librarians, archivists, computer scientists, and scholars of various backgrounds. Digital libraries have been formed--often as cooperative efforts between similar brick-and-mortar libraries--and libraries have developed protocols to add digital media and bibliographic data to their current collections. According to Feldman, these archiving activities--important and revolutionary as they are to the future of information service--still focus [on] reproducing print products in a digital venue. (4) In addition, there is increasing pressure to produce and distribute information in solely digital formats in many sectors of publishing. However, there are grave concerns that governmental and commercial interest in preserving digital media has been less than enthusiastic, and thus the results have been less than rigorous or impressive. * Archiving the Internet There still remains the challenge of archiving the Internet itself. There is a significant growth in Internet-only material, including multimedia (images, audio, and video), journalistic accounts, fiction, personal journals, academic and business documents (papers, presentations, and proceedings), and government documents. Scholars view the Internet as an important artifact that warrants archiving and preservation, but efforts to archive the Internet have been piecemeal. Right now, there is no international standard, no agreed-upon protocol, and no single consortium that has developed a way to archive the Internet. In the absence of a centralized effort, a number of Internet-archival projects have arisen. Most of projects are in the purview of national libraries, such as the British Library's new project to archive Internet Web sites, focusing on sites of British interest. (5) As in the case of the U.K. effort, most government and national libraries have strategized to preserve national material--both official government Web sites as well as Web pages--that is registered under each nation's domain. Among projects is one that is easily set apart from the others. The most well-known of the Internet-archiving projects and, in some ways, the most controversial it is also one of the earliest projects of its kind and it has the broadest mission. The Internet Archive (IA) has proven to be both an anomaly and a vanguard in its attempt to archive as much of the Internet as it can via its technological processes. What IA collects, how it collects, and how it gives access to its collections are important considerations in defining and evaluating the role of Internet-archival projects, particularly as a major activity of digital librarianship. IA was founded as a nonprofit organization in 1996 by Brewster Kahle, founder and CEO of Alexa Internet. IA is the noncommercial sister organization of Alexa Internet (now a wholly-owned subsidiary of Amazon.com), and the two companies share technology. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.