Hey, back in the 2018 edition we had a project board for tracking the changes. I've been wondering as we get closer to 2021 (hard to believe I know) if we should start to do the same? I was thinking of having a org project board that collects all the issues across the
rust and the different repos. https://github.com/orgs/rust-lang/projects
@XAMPPRocky I think that's a really good idea
Alright I've made a project board. I don't have issues assigned yet as I wanted to make sure there was agreement on the structure. It's pretty straightforward hopefully.
Another thing I think we should keep in perspective is when is it too late for a change to make it in the edition. I'm not purposing we have strict policy but for example I wouldn't imagine we'd allow a edition breaking change if there's no implementation by the time we're doing the precursor stable release.
To that end has there been a thought on the edition release date? As looking up on my calendar the last rust releases of 2021 are the 9th of September, the 10th of October, and the 2nd of December.
I think it would be goo to pick a date early (though I don't know if now is too early) so that all the different projects that go into a edition can start planning around that release.
cc @Lokathor -- this is the work I was referring to
@XAMPPRocky the question of the release date is a good one. I think we should set a date that is significantly before the final releases of the year :)
cc @Ashley Mannix I saw some discussion of 2021 edition on the libs stream
"start of the year" is best, though i recognize that for this particular edition that's probably impossible.
but all future editions should be the first release of the year they name.
I don't think I agree
for one thing, it's hard to coordinate things over the holidays
spring or mid-year might be ideal
I do think we should give ourselves a bit of "leeway" -- i.e., have a rough time and then get it more precise
anyway, this is probably worth discussing over in the RFC thread, it was brought up there
Ah I was wondering how we were going to scope out the new edition. I think Libs needs to do some planning to figure out what we could do and what we'd like to do for a new edition :smile: Having a rough mid-year date should give us enough runway to try pull together
Yeah I saw the discussion on the thread, I think motivations for mid-year make sense. Though I think this year is going to be unique in terms of planning.
October 21, 2021, has a rust release it looks like, and that's sufficiently far out for the 2021 edition. Thereafter we can aim for earlier in the year.
i'd say better to aim for 1-2 releases before that (at least internally) so if people realise later they need more time, they can stretch to October
Either way, 1 release before will be the date the beta is cut.
I picked the particular date literally by just wondering "what's the last release before Halloween?"
We've created an #edition stream for coordinating a 2021 edition now :smile: There's a HackMD in there with some more notes. I'll drop a link the project board in there too
@Mara has been busy putting a roadmap together with expert levels of emoji-foo: https://hackmd.io/3eG6OZWHRbSMxoRxzwNhGQ
FYI: I applied for RustSec to be part of the CVE Numbering Authority (CNA) program https://cve.mitre.org/cve/cna.html
it looks like we're tentatively accepted, although in the tentative acceptance document they're listing us as an authority for the "Rust Project"
the only practical implication of this is that RustSec can issue CVE identifiers
(if we're accepted)
there's an onboarding meeting. I will ping email@example.com about it