Complex Citation Object Standardization Hackathon

Hackathon Report: Citation Object Standardization 1. Introduction The virtual “Complex Citation Object Standardization Hackathon” took place on 20 November 2025, 13 - 17 UTC / 14 - 18 CET / 8 - 11 EST. It was organized by the Technical Task Group of the Complex Citation Implementation IG based on the recommendations developed in the Complex Citation WG (Agarwal et al., 2025). The recommendation aims to change the scholarly communication landscape to support authors to cite a large number of existing data and other research objects in their manuscript in a way that enables the attribution of credit for each object as well as the provenance needed for transparency and traceability. 2. Objectives The hackathon was organized with the objective to evaluate the drafted metadata schema for a Complex Citation Object (CCO) by the larger group to provide input for the finalization of version 1 of the metadata schema. This includes aspects of the metadata schema itself and the chosen namespaces but also content evaluation. 3. Hackathon Structure In preparation of the Hackathon, a draft metadata schema was developed. This was the base of the discussion in the Breakout Groups to validate it against participants’ use cases and identify gaps, shortcomings, inaccuracies and inconsistency, which were addressed in the full group discussion. The Hackathon itself was organized in hourly parts with onboarding options every full hour. These were also dedicated times for a report back from the breakout groups and discussions on raised issues and priorities for the next breakout slot. The complexity of the use cases discussed in the Break-outs were increased in subsequent breakouts. This was reflected in the Agenda. The theme of the 3rd Break Out session was agreed to be changed to continue the discussion on theme 2 and Break Out 2. The final discussion wrapped the hackathon up, identified open questions to be addressed, and action items and priorities were set (see Next Steps section). Agenda and notes are available here: https://docs.google.com/document/d/1hAvxIoxqGHFo6UIMFzQ1b8vo4xSxbzJPgOGzMtGkB zM Original Agenda: 13:00 - 13:15: Welcome and instructions, organization of Break-out 1 (BO1) 13:15 - 13:55: Break-out 1 (40 min): Theme1 “Object Listing: simple case: many datasets” and others (BO1 Group 1 - Lou, BO1 Group 2 - Jonathan) 1 13:55 - 14:00: BREAK 14:00 - 14:20: Main Room: Onboarding of new participants and report back from BO1 groups, organization of BO2 14:20 - 14:50: Break-out 2 (30 min): Theme2 “Object intra-relationships: using triplets” (BO2 Group 1 - Lou, BO2 Group 2 - Jonathan) 14:50 - 15:00: BREAK 15:00 - 15:20: Main Room: Onboarding of new participants and report back from BO2 groups, organization of BO3 15:20 - 15:50: Break-out 3 (30 min) Theme3 “Tooling for Objects i.e. In-text Object Referencing” (BO3 Group 1 - Lou, BO3 Group 2 - Jonathan) Changed to continue discussion of Theme2 / BO2 15:50 - 16:00: BREAK 16:00 - 17:00: Main Room: Report back from BO3 groups and final discussion with Q&A, reflections, discussion of priorities/next steps 4. Key Outcomes All materials and outcomes are available in the Hackathon’s google drive folder https://drive.google.com/drive/folders/177wCys_2p_NgveRXWwXXM49Nen0ouCFq BO group1 was dedicated to a data use case, which used the schema template to discuss the usability of the approach. As a result an adjusted schema suggestion was derived and documented: https://docs.google.com/spreadsheets/d/1Xb3Cqxg-9dIPer_IhI4kXJUF8pOM8H7d/edit BO group2 was dedicated to a sample use case. The group used the spreadsheet to generate and document a practical CCO example: https://docs.google.com/spreadsheets/d/1Xb3Cqxg-9dIPer_IhI4kXJUF8pOM8H7d/edit Additionally, problems with the metadata template were highlighted in the group’s notes. The group decided to finalize their use case documentation after the hackathon. Both groups expressed problems or required clarification on the use of the proxy section of the metadata. This is one of the aspects to revisit for the updated template. Other points of confusion were the correct place for creator and other general metadata for the citation object to be mapped into DataCite and the scope of the metadata to be allowed for the linked digital objects listed. There were no other major issues with the template identified. General questions to be or under discussion with specific schema experts are: ● Decision on primary namespace 2 ● Suitable resource type: Initial external experts suggest the use of DCAT, specifically, dcat:Resource and its specialization with dcterms:type 5. Next Steps ● Suggestion (Uwe?): Get examples from papers for evaluation of the CCO metadata, maybe with the use of LLMs based on e.g. Example of paper that would have been a good candidate for using CCO. We can use it as a text. Baldry, K., Johnson, R., Strutton, P.G. et al. A biological ocean data reformatting effort. Sci Data 11, 215 (2024). https://doi.org/10.1038/s41597-024-03038-0 ● Natalie and Jonathan finalize discussion on Physical Sample example ● Shelley: Take-up the question on the need for additional roles (funder, contributor, maintainer of CCO) in Adoption Task Group ● Martina: draft the Hackathon report and distribute for review ● Greta + Shelley, Charlotte, Ian, Marc (interested with low capacity: Chantel, Martina): Describe AI use case and the specifics and add it to the materials in the Google drive folder ● Rob: work on his use case and add it to Google drive folder ● Small group (Jonathan, Lou, Alexandra, James, Martina and Rebecca): o Revise the CCO metadata draft and discuss the need for a proxy section; ensure that the example includes the use of external vocabularies; revisit the primary namespace and resource type decision (recommendation from Simon Cox/Aleksandra to use dcat:Resource with the dcterms:type property to indicate the specific type), definition of subclasses etc. o Support the ESIP January 2026 virtual meeting hackathon (led by Jonathan) with the title: Complex Citation Object Generator Hackathon (90 minutes, tentatively planned for Jan. 21st at 11 AM EST 6. Conclusion The hackathon showed that the Complex Citation approach is relevant and supported by the community. The drafted metadata schema requires some adjustments before finalization but is generally on the right track. The first version of the metadata schema is an essential for the adoption but also the communication of this proposed change of the credit system requiring a community change and engagement with a large number of collaborators. Acknowledgements: The hackathon was funded by the RDA Tiger project. 3 Appendix: Materials All materials and outcomes are available in the Hackathon’s google drive folder https://drive.google.com/drive/folders/177wCys_2p_NgveRXWwXXM49Nen0ouCFq : ● Example Template ● Template Documentation ● Introduction to Hackathon (20 November 2025).pptx ● Hackathon results: o BO1 Data use case: o BO2 Physical Sample use case: Documentation - Physical Samples.xlsx o Rebecca’s simplified schema: Ringuette1 New Draft of Example Template and Model Validation use case Ringuette2 Model Validation Documentation Ringuette2 Model Validation Documentation Further materials of the Hackathon are: ● Schema.org standard ● ORE Specifications and User Guides - Table of Contents - primarily working in JSON-LD (see below) ● Open Archives Initiative: Object Reuse and Exchange Resource Map (JSON-LD) ● schema.org validator tool ● ESIP-approved method to represent datasets on Schema.org (also mirrored for software metadata where it makes sense in codemeta v4.0, coming soon) ● Complex Citation WG Recommendations: Agarwal, D., Ayliffe, J., J. H. Buck, J., Damerow, J., Parton, G., Stall, S., Stockhause, M., and Wyborn, L. (2025). Complex Citation Working Group Recommendation. Zenodo. https://doi.org/10.15497/RDA00130 ● Use cases 4