Overview
There is a broadly recognized need for better situational awareness within the information environment. Each year, millions of articles, books, documents, and datasets are published. Amidst this flood of information, even those with significant experience and expertise in the knowledge economy are struggling to evaluate and vet claims. This document builds on the feedback of dozens of experts across myriad fields submitted to the University of Washington Applied Physics Lab’s Verified Information Exchange Environments Program, to present recommendations for a sociotechnical system, “TrustFinder”, for collaborative management of the information supply chain. TrustFinder implements controls and standards, web and document annotation affordances, argument representation frameworks, and crowdsourcing design principles in order to harness the work of global research communities. The ultimate goal of TrustFinder is to structure the information environment to such an extent that it enables users to find trusted sources of information and rapidly assess concepts and claims.
Overview source: Curated paper metadata.
Methods and contributions
Read the source for the full argument, qualifications, and evidence.
Findings and contributions
- The report recommends TrustFinder, a sociotechnical system combining annotation, argument representation and crowdsourcing so users can find trusted sources and assess claims.
- It recommends platform-level governance features be added only when requested by affected communities rather than imposed on all users.
- It recommends TrustFinder take a facilitatory rather than authoritative role in how researchers adjust and access trust signals.
- It proposes proximal, actor- and community-centric reputation calculation instead of a universal metric to limit viral spread of context-stripped trust signals.
Methods
- Design recommendations built on expert feedback to UW APL VIEE program — The report draws on feedback from dozens of experts submitted to the University of Washington Applied Physics Lab's Verified Information Exchange Environments Program.
- Object model of 10 informational classes (Claims, Claim Combinators, etc.) — Specifies workspace contents such as Content, References, Claims, Questions, Claim/Question Instances, Claim Clusters, Claim Combinators, User Assertions and Stigmergic Tags.
- Set-cardinality reputation metrics with one decay-based score — Defines five reputation types, most via set construction and cardinality, with the Network Impact Score also using a standard decay function.
- Toulmin argument framework and crowdsourcing principles as design basis — The Background section synthesizes frameworks such as Toulmin's six-component argument structure to inform the design.
Summary sources: Paper metadata and evidence · Extracted source text.
Document downloads
PDFs are hosted here; supporting sources may use pinned archives.
- 2022_TrustFinder.pdf PDF · 1.59 MiB · extracted-text source
Citation
Citation metadata follows the unified bibliography.
Catalog details and resources
Related works
Ranked by recorded methods, topics, contributors and associated software, with contributor roles and domain-only matches labelled. Metadata overlap is not a citation or a scientific equivalence.