[I shouldn't have a meaningful voice in this, but my 0.02 ...] Normally I would be strongly against disabling tests ... On woensdag 18 augustus 2021 21:26:04 CEST Nicholas D Steeves wrote: > I'm also concerned that the situation is fundamentally wrong: If tests > need to be updated with every new release, something is wrong with the > tests, or the infrastructure... Tests and infrastructure are supposed > to be the controls that enable meaningful results when testing for > correct functionality in the face of changes. But I 100% agree with this. Tests are meant to guard against regressions, so updating tests should be a rarity. If they need to be updated for every release, then they're utterly meaningless. It's easy to introduce new bugs in the tests or you change the test to get the results you desire. In the former case you should (probably) have tests for the tests and in the latter case you're just fooling yourself and others and creating a false sense of security. Which is BAD.
Attachment:
signature.asc
Description: This is a digitally signed message part.