2024-01-09 [01:44:07.0719] Should spec changes or their accompanying tests land first? [02:47:18.0275] tests first from what I remember [06:07:57.0164] I think “at roughly the same time” is good enough [09:52:03.0145] for proposals, things can't land without getting stage 4, can't get stage 4 without shipping implementations, and at least Chrome is reluctant to ship without tests having landed in test262 (e.g. https://github.com/tc39/test262/pull/3966#issuecomment-1828779758 ) [09:52:11.0484] so tests have to land first for the process to work for proposals [09:52:33.0229] not just reluctant, we won't [09:52:52.0512] * not just reluctant, we won't ship any web visible features (JS or otherwise) without interopable tests (test262 or wpt) [10:03:53.0712] i'd say tests should generally land first, since there's a lag after that before implementations follow them [14:41:24.0421] sorry I meant, I'm happy with tests landing first, but if they're both ready at the same time, it's fine to be sloppy about it (we have been in the past) [14:44:50.0931] also, what's the policy on when things are ready to land? the set methods PR has a stamp from ljharb but I don't know if it requires additional reviews or what [15:08:23.0935] I don't think we have a specific policy on that. I know Jordan prefers 2 stamps [15:10:24.0574] about Set methods specifically, IIRC Jordan's last review covered everything except `*class*.js` — or have those been looked at in the meantime 2024-01-10 [16:54:33.0122] > <@pchimento:igalia.com> I don't think we have a specific policy on that. I know Jordan prefers 2 stamps More reviews seems ideal, if it doesn’t slow down velocity. Does it end up slowing things down to wait for two? [16:55:43.0893] V8 has been waiting to ship Set methods pending test262 for a while [16:55:54.0789] we would've just dumped more stuff into staging but since they were already written and pending review, we waited [16:56:31.0783] we are in a similar boat with regexp modifiers, i think. Ron has a PR open [16:58:06.0722] Do the things awaiting landing have one favorable review? [16:58:52.0782] going by the checkmarks alone, the set methods PR has one from ljharb [16:59:02.0563] regexp modifiers have none: https://github.com/tc39/test262/pull/3960 [16:59:24.0080] I don’t think it is the end of the world if a test bug lands; it can be noticed later and fixed. Actually landing tests can help bugs get noticed faster (because engines run them) [16:59:49.0131] So one review might be enough if it’s delaying things [17:00:57.0738] we could make `{v8,jsc,sm} shipping blocker` label if reviewers are amenable to using that info to prioritize [17:01:57.0112] where the label means, "the only thing remaining to ship the feature is test262 tests" [20:02:28.0476] if someone reviews the class tests (only) I’d be fine with landing it, i just haven’t looked at them [20:02:50.0025] * if someone reviews the class tests (only) I’d be fine with landing it, i just haven’t looked at them (for set methods) [08:27:34.0002] I'll try to take a look at those soon. I reviewed the class tests for `difference` and `intersection` already, I doubt the others are much different 2024-01-17 [08:08:01.0862] Richard Gibson jugglinmike: can you make it to the meeting today? [08:58:43.0564] here are the notes https://docs.google.com/document/d/1JRG6FeYmZIievBPpAdvJjDhfsEjRrdzsFPTYG86KSmM/edit 2024-01-22 [11:35:04.0677] did chakracore magically start working in the CI again, or did one of you fix it? 2024-01-29 [10:35:37.0279] cancelling the meeting during plenary week ftr