Closed
Description
Would it be useful to have a Testing WG? Things a WG might do:
- OMG fix all those flaky tests.
- Create some guidelines around tests. (For example, when should a file containing multiple tests be split into multiple files?)
- Maybe work with the Build WG on the issue of buildbot failures, although I suspect the Build WG is all over that problem. (But just in case: Is there a way to test for buildbot failures? Automate notification to relevant parties so they can be fixed promptly? Is there a way to automate retry when a buildbot fails? Maybe a Test WG can be given limited privileges on the CI infrastructure such that the tedious/mechanistic elements of dealing with buildbot failures can be delegated to them?)
- Maybe see about porting as much of the Python test harness code to Node as is sensible.
Anyone interested?