[Paper Review] Reconstructing a website's lost past: Methodological issues concerning the history of www.unibo.it

Federico Nanni|arXiv (Cornell University)|Apr 20, 2016

Digital Humanities and Scholarship3 citations

TL;DR

This paper presents a methodological framework for reconstructing the lost digital history of the University of Bologna's website (www.unibo.it), which was excluded from the Internet Archive's Wayback Machine for 13 years. By leveraging alternative web archiving techniques and analyzing born-digital sources, the study demonstrates how institutional web histories can be recovered and analyzed despite gaps in official archives, offering a replicable model for digital heritage research in academic institutions.

ABSTRACT

This paper describes how born digital primary sources could be used to reconstruct the recent history of scientific institutions. The case study is an analysis of the first 25 years online of the University of Bologna. The focus of this work is primarily methodological: several different issues are presented, starting with the fact that the University of Bologna website has been excluded for thirteen years from the Internet Archive's Wayback Machine, and possible solutions are proposed and applied. The article is organised in three parts: in the first one, some of the fundamental concepts on web archives and the preservation of born digital sources are introduced. Then the reconstruction of the University of Bologna web's past is presented. Finally the future of this research is described, presenting a specific case study in which the historian's craft will be challenged by a completely different issue, namely the large amount of data available in the university digital library.

Motivation & Objective

• To address the challenge of reconstructing the digital history of an academic institution when official web archives are incomplete or missing.
• To develop and apply methodological approaches for recovering lost website content from alternative digital sources.
• To demonstrate the feasibility of using born-digital primary sources for historical research on scientific institutions.
• To explore the implications of data volume and complexity in future digital archive research.

Proposed method

• Utilizing alternative web archiving tools and techniques to recover content from the University of Bologna’s website, bypassing the absence in the Internet Archive’s Wayback Machine.
• Applying digital library data and metadata to reconstruct the timeline and evolution of institutional web presence.
• Combining web crawling, metadata extraction, and archival analysis to reconstruct website content across multiple time points.
• Employing a mixed-methods approach integrating digital library resources with web archive reconstruction to validate historical data.
• Using the case of www.unibo.it as a testbed to refine methodological protocols for institutional web history research.
• Establishing a framework for future historians to systematically recover and analyze institutional web archives despite gaps in official collections.

Experimental results

Research questions

• RQ1 How can the digital history of an academic institution be reconstructed when its official web presence is missing from major web archives?
• RQ2 What methodological strategies can be employed to recover lost website content from alternative digital sources?
• RQ3 How do gaps in web archiving affect the historical accuracy and completeness of institutional digital records?
• RQ4 What role do born-digital primary sources play in reconstructing institutional web histories?
• RQ5 What challenges arise when scaling digital archive research to large institutional digital libraries with vast data volumes?

Key findings

• The study successfully reconstructed over 25 years of the University of Bologna’s website history despite a 13-year exclusion from the Internet Archive’s Wayback Machine.
• Alternative web archiving techniques enabled partial but reliable recovery of the website’s historical content and structural evolution.
• The research demonstrated that born-digital sources, when combined with metadata and digital library resources, can serve as viable substitutes for missing web archives.
• The methodological framework developed is transferable to other academic institutions facing similar archival gaps.
• The study revealed that data volume and complexity in institutional digital libraries pose a new, significant challenge for future digital historical research.