Re: Tracking spec triage and maintenance

We don’t use labels to triage issues but instead put issues into milestones. Could the script look at that to determine if issues have been triaged?
________________________________
From: Jeffrey Yasskin <jyasskin@google.com>
Sent: Monday, November 6, 2023 2:17:50 PM
To: spec-prod <spec-prod@w3.org>
Subject: Tracking spec triage and maintenance

Hi spec editors,

I put together a system that can scan our specification repositories on Github and report how they're doing at triaging issues and fixing the most important ones. See https://jyasskin.github.io/spec-maintenance/, with source at https://github.com/jyasskin/spec-maintenance.

To make this work well, there need to be some labels that people apply to describe issues' priorities. The script currently suggests<https://jyasskin.github.io/spec-maintenance/triage-labels/> "Priority: Urgent", "Priority: Important", and "Priority: Eventually", but we could use something shorter like "p0", "p1", and "p2". We could also have an slo.json in each repository to override the default label names. What do you want for your own specs?

I also picked default SLOs<https://github.com/jyasskin/spec-maintenance/blob/main/README.md#triage> of 1 week to triage issues, 2 weeks to close urgent issues, and 3 months to close important issues. Do those values sound reasonable, based on the metrics I gathered<https://github.com/jyasskin/spec-maintenance/blob/main/README.md#triage>? Does this need to be customizable per-repo?

How else can I make this the most useful to everyone?

Thanks,
Jeffrey