PofoliaShared via Pofolia

Smart Cities· 2026Q1

Decoupling Evidence Custody from Multi-Party Alert Coordination for Aging in Place

Yushan Li, Satoshi Fujita

Short summary

A new design decouples evidence custody from alert coordination, allowing responders to complete tasks even if the resident's device fails, as demonstrated by a Holochain prototype where 100% of four-agent lifecycles completed after the device stopped, versus 0% for a baseline.

AI-generated from the title and abstract; the full text is not read.

Key points

  • Evidence custody is separated from coordination progression, with private evidence remaining on the resident's device.
  • Coordination objects carry only selected fields, without direct references to private evidence.
  • Participants use shared rules and lifecycle functions for validation, ensuring agreement on response states.
  • A Holochain prototype successfully completed 100% of 100 four-agent lifecycles after the resident's device stopped, unlike a baseline system.

AI-generated from the title and abstract; the full text is not read.

Abstract

When several people respond to an aging-in-place alert, they need a shared record of who has seen it, who has taken it on, and how far the response has progressed. In a holder-mediated design, the resident-side device serves every acknowledgment and claim. We separate evidence custody from coordination progression. Detailed evidence remains in a resident-side store outside the coordination network, and explicit publication creates a coordination object carrying selected coordination fields and no reference to the private evidence. Without a holder to serialize updates, participants need shared rules: validating peers check each signed action against a lifecycle function using public records, and a deterministic selection rule gives participants holding the same action set the same handling state while preserving competing actions and their authorship. In a single-host Holochain prototype, validation rejected all 31 invalid writes that bypassed the application’s checks, and participants agreed on the Resolved state after the expected actions had propagated in all 100 four-agent lifecycles with concurrent claims. Against the tested holder-mediated baseline, online observation took seconds rather than milliseconds, yet this design required no holder requests. After the holder stopped, responders who had discovered the alert completed all ten tested lifecycles; the baseline completed none. Once participants have obtained the required coordination records, they can continue even when the resident-side device becomes unavailable.

The authors' abstract, as published at the source. Smart Cities, 2026 · DOI ↗

TakeawaysPremium
Ask the paperFree account

Continue with a free account

Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.

Continue free on the web

Sign in with Google or Apple; no card needed. You come back to this paper.

On your phone:

Field: Information Systems

Information SystemsComputer Science