media-right layout invoked

Foundational principles

The COMET model proposes an evolution in metadata quality grounded in three fundamental shifts that align with the open science movement.
placeholder image
three layout invoked

1. Collective stewardship

The COMET model shifts the responsibility of metadata quality from individual depositors to a distributed community partnership. Crucially, this does not undermine source authority—it complements it. The model recognises that source authority and metadata quality are distinct, and that research outputs are the product of many actors whose collective expertise should be empowered to produce a complete and accurate scholarly record.

2. Collective benefit

The COMET model addresses the fragmented benefits that result when organisations tackle metadata problems in isolation. Independent fixes by universities, funders, and service providers solve local problems but result in duplicated, conflicting metadata that never reaches the authoritative source. COMET provides a collective framework for community-contributed improvements, structured as open assertions with clear provenance, machine-actionable and scalable. Enrichments flow back to authoritative sources.

3. Trust through transparency

The COMET model establishes trust through open processes. Traditional PID workflows root trust in who registered the metadata. COMET extends this by rooting trust in process with evidence—comprehensive provenance documentation and publicly shared evaluative frameworks provide a more rigorous and robust foundation for a system the research community depends on.

diagram stage layout
stack layout invoked

Key elements

We believe these shifts can be achieved by prioritising four key elements in the work we do.
1. Unite quality metadata sources and enrichment methods

COMET confronts one of the most persistent inefficiencies in scholarly communications: valuable, high-quality curation exists in isolated systems that never reaches the wider community. COMET brings these sources together, unlocking their collective value.

Proven community enrichment methods also exist that we can build on. Where gaps remain, COMET acts as an innovative facilitator, developing new capabilities with community partners. The goal is to connect solutions across institutional and geographical boundaries to maximise the value of community investments.

2. Target diverse content types and stakeholders

Inclusive design produces solutions that reflect genuine community needs. To ensure solutions are broadly applicable rather than narrowly useful, COMET addresses multiple content types across projects to develop flexible approaches that benefit the widest possible range of research outputs and the communities that produce them.

Truly representing diverse stakeholders means fostering partnerships and collaborations in many geographies. COMET has already engaged over 200 organisations spanning research institutions, libraries, publishers, funders, infrastructure providers, and advocacy and policy groups.

3. Provide multiple round-tripping pathways

COMET creates both direct and platform-mediated mechanisms for community improvements to reach authoritative sources—for example through DOI registration agencies such as DataCite and submission and publishing platforms such as PKP’s Open Journal Systems.

4. Treat metadata fields as features

Inspired by software development that makes systematic progress possible, the COMET model treats each metadata field as a discrete, improvable feature, enabling focused problem-solving and measurable progress. Each project we undertake as a community tackles one metadata field at a time.

COMET is prioritising the following use cases:

with-table sub layout invoked
Use caseMetadata fields
Name and describe research outputsTitle; Work/Resource Type; Language
Identify and connect people associated with research outputsAuthor/Creator; Contributor
Identify and connect organisations associated with research outputsAffiliation; Funder/Funding
Identify and connect related research outputsReferences; Related Identifiers
Identify access policies associated with research outputsLicense/Rights