notes: research/papers/kerione-a-did-for-everything

This data as json

path title type status classification words body
research/papers/kerione-a-did-for-everything A DID for Everything summary v0.1 — 2026-10-14: drafted by Claude from the PDF; unread by Mark; marked partner on Mark's word 2026-10-21, paraphrase only partner 563 # A DID for Everything **Abstract.** A Rebooting the Web of Trust VII (RWOT VII) white paper by Shaun Conway, Andrew Hughes, Moses Ma, Jack Poole, Martin Riedel, Samuel M. Smith and Carsten Stöcker (version 1.0, January 2019), subtitled "Attribution, Verification and Provenance for Entities and Data Items". It extends decentralized identifiers (DIDs) from people to any entity, including machines, locations, events and data items, and proposes chaining signed decentralized autonomic data (DAD) items to prove the provenance of a data flow. Written for architects of decentralized identity, Internet of Things (IoT) and data-provenance systems. ## Summary The paper describes the DID as a globally unique, portable identifier that depends on no central authority, and argues that it can identify everything: people, physical objects, IoT devices, locations, events, agents and pure data objects. It builds on the earlier RWOT paper on Decentralized Autonomic Data. "Why this matters" sets out the problem: identities sit in many centralized or federated systems, each its own root of trust, so a new actor in a value chain cannot independently verify a person's credentials or an object's or data item's provenance, which adds hidden trust costs. It then considers why this matters for people, businesses, objects and machines, and the world. The "comprehensive solution" section recalls three concepts from the DAD paper (DAD items, derived DIDs or dDIDs, and self-contained verification of data governance and integrity) and adds a fourth: data flows can be provenanced through a self-referential chain of signed DAD items, each including the identifier and signature of the previous one. The authors distinguish custody of data from ownership, since data can be copied, and tie data-flow provenance to a zero-trust, or "diffuse-trust", perimeterless security model for decentralized applications. Technical subsections define DID and dDID syntax, a minimal DAD, change detection, dDIDs generated on the fly or unique to the data, public derivation, and the chaining of DADs for originating, transformed and combined data. A medical imaging workflow and decentralized fog computing infrastructure are worked as use cases. Future directions include verifiable credentials for signers, location claims, pairwise DIDs with zero-knowledge proofs, and attribution graphs. The conclusion imagines global adoption while acknowledging that no single technology solves problems such as propaganda. ## Key terms decentralized identifier (DID), derived DID (dDID), decentralized autonomic data (DAD), data-flow provenance, chain of custody, digital twin, verifiable claims, zero-trust (diffuse-trust) security, fog computing ## Claims and their evidence - DIDs can be extended from people to any entity, including pure data objects. [informative] - A chain of DAD items, each carrying the prior item's identifier and signature, establishes the provenance and end-to-end integrity of a data flow. [informative] - Custody of data is not equivalent to ownership; transferring data ownership is transferring rights to use it. [informative] - An estimated 75 billion IoT devices will be connected in 2025. [unverified] (a cited projection) - Wide adoption of DIDs and DADs by phone makers and social networks would end phishing and fake news. [unverified] (presented as a vision) ## Related Decentralized Autonomic Data (DAD) and the three R's of Key Management (RWOT VI); Decentralized Identity as a Meta-platform: How Cooperation Beats Aggregation (RWOT IX). ## Provenance Drafted by Claude on 2026-10-14 from the PDF (24 pages): the abstract, "why this matters", the solution and chaining sections, future direction and conclusion, with the section headings for the rest. Not yet read by Mark.

Links from other tables