Link rot

Abstract: Purpose – This research aims to study the state of online resources cited in scholarly library and information science (LIS) journals which are ranked in ISI and available in the Emerald database in terms of accessibility and decay.Design/methodology/approach – Four LIS journals published by Emerald were selected from Thomson Reuters' JCR. The journals' issues from 2005 to 2008 were downloaded directly from the publisher web site and checked in terms of decay and availability of individual cited URLs.Findings – Original accessibility of studied online resources was 64 percent, which improved to 95 percent. The main adopted strategies that returned more results were using the Wayback Machine and Google, which revived online resources by 17 percent and 12 percent respectively.Practical implications – To increase the rate of web citations accessibility, some recommendations, including avoiding long URLs, citing documents found in digital collections availability on the web, working through systematic checkin...

Abstract: Ms. Rhodes explores URL stability, measured by the prevalence of link rot over a three-year period, among the original URLs for law- and policy-related materials published to the web and archived though the Chesapeake Project, a collaborative digital preservation initiative under way in the law library community. The results demonstrate a significant increase in link rot over time in materials originally published to seemingly stable organization, government, and state web sites.

Abstract: Today's scientific data are primarily stored and accessed via centralized Web-based infrastructure. Centralization has advantages but also carries risks such as link rot and content drift, which can hinder scientific progress. It is time to ask whether traditional, centralized Web architecture aligns with scholarly priorities and values, and to collaboratively move towards new approaches that do.

Abstract: The researchers report on a study to examine the persistence of Web-based content. In 2002, a sample of 500 citations to Internet resources from articles published in library and information science journals in 1999 and 2000 were analyzed by citation characteristics and searched to determine cited content persistence, availability on the Web, and availability in the Internet Archive. Statistical analyses were conducted to identify citation characteristics associated with availability. The sample URLs were searched again between August 2005 and June 2006 to determine persistence, availability on the Web, and in the Internet Archive. As in the original study, the researchers cross-tabulated the results with URL characteristics and reviewed and analyzed journal instructions to authors on citing content on the Web. Findings included a decrease of 17.4 percent in persistence, and 8.2 percent in availability on the Web. When availability in the Internet Archives was factored in, the overall availability of Web content in the sample dropped from 89.2 percent to 80.6 percent. The statistical analysis confirmed the association between the likelihood that cited content will be found by future researchers and citation characteristics of content, domain, page type, and directory depth. The researchers also found an increase in the number of journals that provide instruction to authors on citing content on the Web.