...
[Peter]: Although it was tight, I was impressed by the community’s ability to complete the testing on schedule (perhaps we were helped by some features being pushed to the next release).
[Chelsea]: It seemed that the prioritization of discovered bugs was helpful for the triaging process.
[Sarina] A lot of people really stepped up this release. The two that were most visible to me were Maria Grimaldi and Chris Patti , as well as Jorge Londoño and Maksim Sokolskiy .
[Sarina] largely, new product processes worked, and involved more people than previously in the release
[Maria] Thanks to the product working group's involvement, we had a framework to prioritize issues and classify release blockers. This helped us focus even more on which issues to triage first.
[Maria] Even though we found the security issue when almost finishing the testing process, we solved it in a timely manner with community help.
[Maria] Maintainers participated more in solving issues than in previous releases.
[Adolfo] Main features were developed with Redwood in mind, and thus reduced the incidence of bugs or omissions that would’ve otherwise needed to be caught and fixed during testing. Shout-out to Raccon Gang and Opencraft for doing so with the Studio MFE and Tagging, respectively.
III. We should keep ✅
[Jenna]: Propose that we keep Product ownership of the “master” test spreadsheet, and continue to flesh out regression tests
[Jenna]: Propose that we keep the monthly meeting to track features slated for Sumac (can start again in August
[Chelsea]: Keep adding cucumber formatted tests to the test list? (Given I… When I… Then I…)
[Maria]: Keep the “Priority” test sheet column up to date and matching the product narrative so triagers can map it to test failure reports.
[Maria]: Propose that the product working group continue helping bug triages prioritize issues on all BTR boards. I say “all” since not all boards use maps to test failures, which bug triagers can classify by looking at the testing sheet, so we also need help for the rest of them.
[Maria]: Keep the release blockers framework proposed by Jenna, but document it somewhere in the release testing process docs.
[Adolfo] Developing new features and testing them with Tutor nightly (including via PR sandboxes) before the PRs land.
[Adolfo] We should continue to enlist the help of more, caring, maintainers that are not only aware of but engaged with the release process. This should not only help reduce the incidence of regressions and other bugs, but add to the pool of resources available to fix issues found during testing.
IV. What didn’t we do so well, and what could we have done to do better ❌
...