Research Data, Software & Reproducibility Policy

Preserve the evidence. Make the method understandable.

The JR Institute intends to manage research data, software, models, workflows, and computational environments so that findings can be evaluated, reproduced where practical, preserved over time, and shared responsibly.

Developing Framework

This page presents a planned public standard. Final data-management templates, repositories, retention schedules, access controls, software-preservation systems, and review authorities should be established before active research programs expand.

Policy Purpose

Research claims are strongest when the supporting record can be examined.

Data, code, configuration, instruments, preprocessing, prompts, models, statistical decisions, and software dependencies may all affect a result.

The Institute should preserve enough information to explain how conclusions were produced, while protecting participants, security, intellectual property, and legitimate restrictions.

Core Principles

Plan early, document clearly, preserve durably, and share responsibly.

01

Data Management Planning

Define collection, storage, access, documentation, sharing, retention, and disposition before work begins.

02

Reproducible Methods

Preserve code, parameters, workflows, environments, and methodological decisions.

03

Responsible Openness

Share as openly as possible and restrict only as necessary for legitimate reasons.

04

Trusted Metadata

Describe provenance, formats, variables, transformations, versions, and limitations.

05

Secure Stewardship

Protect sensitive, confidential, regulated, proprietary, and security-relevant information.

06

Long-Term Preservation

Use durable formats, stable identifiers, redundant storage, checksums, and migration planning.

Scope

This policy is intended to apply to research data, source code, scripts, software, machine-learning models, prompts, workflows, notebooks, images, recordings, simulations, instrument outputs, survey records, laboratory records, and documentation needed to understand a result.

Data Management Plans

Projects should define how research materials will be collected, named, organized, documented, backed up, secured, shared, retained, and disposed of.

  • Identify expected data types, formats, and scale
  • Assign responsible stewards and access roles
  • Define backup and recovery procedures
  • Address consent, privacy, sponsor, and security restrictions
  • Plan for repository deposit and long-term preservation

Ownership, Custody, and Stewardship

Research materials may be subject to institutional, sponsor, collaborator, participant, contractual, or legal interests.

Project leaders should understand who owns, controls, may access, may publish, and must preserve each category of material.

Metadata, Provenance, and Documentation

Records should describe where data came from, how it was collected, what transformations occurred, which versions were used, and what limitations affect interpretation.

Variable definitions, units, coding schemes, exclusions, preprocessing, quality checks, and analytical decisions should be preserved.

A file without context may be technically preserved while remaining scientifically unusable.

Software, Models, Code, and Computational Environments

Computational research should preserve, where practical, source code, dependency information, package versions, build instructions, parameters, hardware assumptions, random seeds, prompts, model versions, and configuration files.

Containers, virtual environments, notebooks, workflow descriptions, and test cases may be used to improve reproducibility.

Open Access, Data Sharing, and Reuse

Research outputs should be shared as openly as ethical, legal, contractual, and security obligations permit.

Shared materials should include clear licenses, citation expectations, access conditions, known limitations, and any restrictions on commercial or sensitive reuse.

Sensitive, Confidential, and Restricted Data

Sensitive data may require de-identification, encryption, controlled repositories, data-use agreements, limited access, secure analysis environments, or prohibition on public release.

Restrictions may arise from participant consent, privacy, law, sponsor terms, export controls, cybersecurity, indigenous or community interests, proprietary rights, or dual-use concerns.

Retention, Integrity, and Preservation

Research records should be retained for a period appropriate to sponsor requirements, publication, intellectual property, participant protection, audit, verification, legal obligations, and future scientific value.

Preservation should include version control, checksums, access logs, backups, format review, and migration planning.

Repository Deposit and Persistent Identification

Suitable datasets, software, models, and documentation should be deposited in an approved repository with stable identifiers and exportable metadata.

Repository records should link related preprints, publications, projects, authors, funding, licenses, versions, and corrections.

Project Closeout, Personnel Departure, and Transfer

Before a project closes or personnel depart, responsible leaders should confirm custody, access, retention, documentation, repository deposit, transfer permissions, and future stewardship.

No individual should remove the only usable copy of institutional research records, software, or data without authorization.

Framework date: July 2026

Research Record Lifecycle

Plan, collect, document, analyze, share, preserve, and close.

  • Plan. Define stewardship, formats, access, security, backup, sharing, and retention.
  • Collect and document. Preserve provenance, methods, variables, transformations, and quality checks.
  • Analyze reproducibly. Record code, versions, parameters, models, prompts, and computational environments.
  • Share responsibly. Use suitable repositories, licenses, access controls, and clear metadata.
  • Preserve and close. Confirm custody, retention, integrity, identifiers, and long-term stewardship.
Data or Reproducibility Questions

Ask about data plans, repositories, software preservation, access, or reproducibility.

Inquiries may concern metadata, sensitive data, code, models, computational environments, sharing, licensing, retention, deposit, or project closeout.

Submit a Research Data Inquiry

Identify the project, data type, software or model, access concern, deadline, and available supporting information.

Contact the Institute