This is an old revision of the document!
Table of Contents
GitHub AUTH48 Experiment: RFC 9069
High-Level Summary
Set up very similarly to the experiment for RFC 9131.
- RPC created an AUTH48 repo and invited the authors as collaborators.
- Each AQ was made into an issue (12 total).
- Both editor and author created branches and submitted PRs (10 total) to close issues.
Description
This fourth experiment involved a small document (4 authors, 14 pages, not in a cluster, 12 questions at the end of the EDIT pass). One of the authors (Tim Evens) and the AD agreed to the experiment when the document entered the queue.
GitHub Setup
In this experiment, the RPC created and controlled the repo. Both pull requests and issue tracking were used.
Repository
- Initially set to private.
- Contains:
- rfc9131.xml
- README.md
- CONTRIBUTING.md
- note-well.md
Issue Tracking
- Initial number of issues: 12
- Issue labels:
- rfced - initial AQ
- question - follow-on question added to an issue
- AD approval required - document update requires AD review and approval
- editor-ready - the editor can update the document and/or close the issue based on issue discussion
AUTH48 Notes
AUTH48 started 2021 Oct 18.
Published 2022 Feb 16.
RPC Lessons Learned
Process change: Editor should copy updated XML to the RPC server to make sure xml2rfc runs correctly. Ultimately, should be able to make it easy to run xml2rfc in the repo.
Cadence is different; can't just update the doc to help the authors see the proposed updates; must wait for confirmation before making changes.
This was hybrid process: 2 authors, the doc shepherd, and the AD engaged via GitHub; others just approved at the end via email.
Participant Survey
Survey sent 23 Feb 2022 and 8 March 2022. No responses received.
