meta: `author-ready` label name is misleading

[jasnell]

The author-ready label does not actually mean that the PR author is ready for the PR to proceed. A different name should be selected or the label should be redefined.

[Trott]

We can probably come up with a more intuitive name. If I understand how it's primarily used (and I think it was @BridgeAR and/or @addaleax who primarily championed it?), maybe ready to land? That is also misleading because it's ready to land except for the waiting period, but I'm not sure how to include that succinctly. Any suggestions?

[jasnell]

Except that it has been applied inconsistently even in "ready to land" terms. If there is to be a "ready to land" label, it needs to be applied consistently... and yes, that would be a significantly better label if that is the purpose.

"author ready" should indicate, the author considers this PR complete and ready to move forward. That would actually make sense.

[addaleax]

@jasnell I'm missing context here ... we originally had it at ready but switched to author ready because it meant that the PR author had nothing left to do (at least not in the role of the author). Can you give an example where that doesn't match our usage?

[Trott]

Can you give an example where that doesn't match our usage?

@addaleax COLLABORATOR_GUIDE says not to apply author ready until there is at least one Collaborator approval for the PR. @jasnell opened this issue after I flagged his usage of author ready here and here.

I don't think "waiting for a Collaborator approval" equates to "something left for the author to do" but that's how we use the label and how it's documented. Maybe the label should be land-able after 48/72 hours but that's probably a bit long? (...although if #22275 lands, we can at least drop the 48/.)

Anyway, I'm actually fine leaving it as author ready but I am sympathetic to @jasnell's view here (as I understand it anyway) that the name does not intuitively convey what the actual use/meaning is. I'm just having a hard time coming up with something better that is sufficiently succinct.

[bengl]

Perhaps land-able may suffice. The wait period is not consistently enforced across all PRs (because a procedure for fast-track exists), so doesn't need to be in the tag.

On the other hand, the tag right now has the description: PRs that have at least one LG, no pending reviews and a CI started. It makes no mention of the CI state at the end of its run, so perhaps landable would actually be factually incorrent. 🤷‍♂️

[BridgeAR]

I would like to keep this label as is. However, if there is confusion about landable and author-ready: it is trivial to add another label that we are able to use on top of the already existing one. The point is just: as soon as we add the landable label, it is actually against the rules that we gave us because the PR should just be landed by the collaborator instead.

[jasnell]

No further action on this.