Wasteback Machine: a method for quantitative measurement of the archived web

Introduction. Web archives are traditionally viewed as repositories of cultural memory, yet they have been theorised as computational sources for quantitative, longitudinal analysis of the web. This paper examines their potential for mapping the structural and environmental impacts of web pages, demonstrating broader applicability for web analytics research. Method. We introduce Wasteback Machine, an open-source, extensible framework that operationalises the analytical potential of web archives. It enables reproducible, scalable measurement of page size and composition through programmatic access, structured resource extraction and mechanisms to mitigate distortions introduced during archiving and replay. Analysis. The method is demonstrated through a case study of the United Nations Climate Change (UNFCCC) homepage, performing longitudinal analyses to capture temporal dynamics in size and compositional evolution. By situating web content within socio-technical and infrastructural contexts, the approach allows consistent comparison over time while accounting for archival limitations. Results. Findings reveal trends in page growth, complexity and cumulative digital resource use. Despite their fragmentary nature, web archives provide sufficient fidelity to reconstruct historical practices and estimate relative environmental impacts. Conclusion. Wasteback Machine demonstrates that web archives function as computational infrastructures, enabling rigorous, evidence-based investigation of web evolution and the environmental footprint of digital content.