This experiment has been run twice (during AUTH48 for RFC 8446 and RFC 8829). The idea was to use GitHub instead of email for AUTH48 state. Details below.
RFC 8446 | RFC 8829 | |
---|---|---|
I-D | draft-ietf-tls-tls13-28 | draft-ietf-rtcweb-jsep |
Pages submitted | 156 pages | 115 pages |
I-D approved | 2018-03-21 | 2018-03-01 (into MISSREF state) |
AUTH48 start | 2018-06-14 | 2020-07-06 |
Publication | 2018-08-10 | 2021-01-20 |
GitHub repo | https://github.com/tlswg/tls13-rfc | https://github.com/rtcweb-wg/jsep |
AUTH48 details | ||
Time in state | 8.2 weeks | 28.3 weeks |
# questions at start | 58 | 60 (yielded 78 issues in GitHub) |
Does GitHub seem to provide an easy-to-use mechanism for:
RFC 8446 | RFC 8829 | |
---|---|---|
AUTH48 time using GitHub | 8.2 weeks | 28.3 weeks |
Comparing with standard process | ||
Concurrent Avg. AUTH48 time* | 3.2 weeks | 10.1 weeks |
Avg. AUTH48 time for docs over 100 pages** | 4.4 weeks (n=7) | 3.5 weeks (n=6) |
* using standard AUTH48 process for the 3 months preceding publication.
** using standard AUTH48 process for the 18 months preceding publication.
The authors and editor agreed to use the GitHub Pull Request work flow:
The authors and the editor set up notifications so that many of these steps automatically generated emails to involved parties (assigning/commenting on issues, use of @mentions, creating/accepting/rejecting PRs). 766 messages were generated during AUTH48.