Publication ethics

Integrity of results

Authors must report their results completely and accurately, including results that weaken their conclusions. Fabricating, altering or selectively omitting results is misconduct.

Responsible disclosure

Authors must report problems to maintainers before submission. Where a problem may be a security vulnerability, authors must follow the project’s security policy, and the journal will not publish exploitable details before that policy permits.

Fairness to maintainers

Articles describe software and data, not the competence or character of the people who maintain them. Maintainers are given the opportunity to respond before publication, and their responses are published alongside the article.

Originality and attribution

Submissions must be the authors’ own work and must credit prior reports of the same problem, including existing issues in the project’s tracker.

Use of AI tools

Authors, reviewers and editors must disclose any use of AI tools in their work for the journal. AI tools cannot be listed as authors, and the people named remain fully responsible for the content.

Handling concerns

Concerns about a submission or published article, including suspected misconduct, should be raised with the editors through the submissions repository. Where a concern involves an editor, it is handled by another member of the editorial team. Proven misconduct leads to rejection or retraction, with the reason stated.

Appeals

Authors may appeal an editorial decision once, by stating their reasons on the submission issue within 30 days. The appeal is considered by an editor who did not make the original decision.