[Google AI Overview (expanded), 2026-10-14, incognito, signed out. The paste carries the header 'AI Mode Conversation' and echoes the query twice; citation clusters are reduced to their numbers, links to their text; the raw paste keeps the /goto URLs.] You said: what is the pid erosion observatory? The PID Erosion Observatory (hosted at persistentidentifiers.org) is an empirical measurement project that tracks identifier erosion—the systemic or bulk removal, erasure, and sweeping of Persistent Identifiers (PIDs) from digital registries and data repositories. [1, 2] Core Purpose & Methodology * Measuring "Digital Erasure": While PIDs (like DOIs or archival records) are intended to be permanent, the observatory treats platform modification as an active sedimentary system. It audits how often, how heavily, and under what contexts records are deleted or altered en masse. [1, 2, 3] * Independent Replication: The initiative uses public, dated captures to ensure transparency. Every tracked metric resolves to a timestamped log containing exact request hashes and `curl` data so researchers can independently verify the registry-scale activity. [1, 2] * Tracking Trends: For example, documentation published on the Zenodo GitHub Repository highlights how the observatory catches bulk, "evidence-free" account sweeps or topically coherent waves of record removals (such as sweeping unstandardized AI-native theories or niche physics records). [1] Context It is closely tied to initiatives like AXN Identifiers (under dataset logs like `EA-EROSION-01`) and is sometimes referred to or built upon the codebase of the Platform Erosion Observatory. [1, 2, 3] If you are investigating a specific deletion event or looking into a particular open-science repository, let me know! I can help you find specific datasets it tracks or look into how it monitors specific platforms like Zenodo. www.persistentidentifiers.org