Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

Greetings @connorjclark. I'm sure you are busy, but I want to make sure this PR doesn't slowly fade into oblivion 😄

Let me know if there is anything I can do to answer any questions or help make reviewing these changes easier for you or your team.

If I understand correctly: currently we weigh the a11y audits based on each axe rule's severity (that should be pretty consistent, but it's possible severity of some rules may have changed without us noticing). And your proposal is to additionally not score an audit if the severity is minor and merely a "best practice", ie. not in wcag (weight: 1 -> 0) – but otherwise, we treat "deque best practice rules" the same as rules backed by wcag. Is that correct?

If so, I think that would be a good change.

Could you help me understand the actual delta in audit weights? I'm having trouble telling by just looking at the diff.

We can't make scoring changes w/o a breaking release, but there will be one in October.

If I understand correctly: currently we weigh the a11y audits based on each axe rule's severity (that should be pretty consistent, but it's possible severity of some rules may have changed without us noticing). And your proposal is to additionally not score an audit if the severity is minor and merely a "best practice", ie. not in wcag (weight: 1 -> 0) – but otherwise, we treat "deque best practice rules" the same as rules backed by wcag. Is that correct?

@connorjclark Yes. Your summary is correct 👍

Could you help me understand the actual delta in audit weights? I'm having trouble telling by just looking at the diff.

We can't make scoring changes w/o a breaking release, but there will be one in October.

@connorjclark Agreed, this will affect scoring changes.

I've created a branch to make the default.js changes more readable. https://github.com/GoogleChrome/lighthouse/pull/16667/files.

For easier diffing It adjusts default.js by:

• removing comments on (most) unchanged lines
• removing the reordering of line items

I've also summarized the changes in this screenshot.

Please note, in the screenshot, I have also added an additional question. The TLDR is: What do you think about the values I have chosen for Deque 'best-practices' audits?

I'm open to suggestions, adjustments, and discussions. Many thanks for taking a look at this 🙇

Greetings @connorjclark, just wanted to make sure you saw my previous comment. Thanks!

I think we should continue weighing best-practice checks that are not Minor. I want to keep deferring to Deque's expertise, but raising the bar a tiny bit to omit the minor ones feels alright to me.

Whether they are equal to the wcag one is a bit more subjective. I could see us down-shifting the rubric for best-practices (minor: 0, moderate: 1, Serious: 3, Critical: 7). I'll ask around internally.

Thanks for producing #16667. FYI, it there may be a few weeks until this is accepted, but I don't think we'll require any additional work from your end. Thanks for taking the time to make this proposal.

EDIT: @paulirish and I quickly settled on just doing what you have in your PR (minor + best-practices = 0), not the "down-shifting" thing I mentioned above. Had no strong reason to choose one over the other, so we preferred the smaller change.

@connorjclark sounds great. let me know if there is anything else you need from me and many thanks for taking the time to review 🙇

Thanks Ben!