What's Behind Ember Observer's Scores?
summary
Ember Observer replaces a rigid ten-check checklist with a weighted average so different criteria count more or less, rebalances scores from a 0–10 scale, and surfaces each check on an explain page so contributors see exactly what changed. New checks include partial credit for contributor counts (full at five), a lightly weighted non-default description, a check for more than one maintainer on npm, and a ≥1.0 version flag. The new system produces a shifted distribution, is extensible and testable, and is rolling out live.
AI-written summary · about
chapters (5)
- 0:00Start
- 0:05What's Behind Ember Observer's Scores?
- 1:13Can you score on accessibility?
- 4:10Smarter checks and partial credit
- 4:27New Checks
related talks
27mContributing to Ember The Inside Scoop
06mContributor Rally
24mMastering Ember from the Perspective of a N00b
05mIntroducing Elixir Observer
34mSoftware Productivity and Ember: Ember CLI
15mWhy Contributing Seems Scary
34mCommunity Data: What to Measure and Why?
16mTest as if you're in someone else's shoes