Perf test your website/app on cell service while you wait for you dentist. It’ll either feel horrible, getting you ready for the cleaning, or feel great, getting you ready for the cleaning. The change I’m making appears to give ~5s back to visitor by not rendering a bunch of unnecessary UI. AND, there’s more work to be done!
Clear teeth, clean web.
Added a bunch of new tests, results, and thoughts. Hope you'll take a look at share the results you're seeing when testing locally!
@pablolarah @eric @piccalilli Can’t wait till all browsers support `:host(:has(…))` the way #webkit does and #mozilla partially does so that this is even more natural for #webComponent development, too.
https://wpt.fyi/results/css/css-scoping?label=master&label=experimental&aligned&q=host%20has-0
I love it when a plan comes together!!
`:host:has(...) {}` is now shipping in Chromium! 
If you're into using `@property {}` declarations with `light-dark()` values, you may want to track this issue in #firefox ...
https://bugzilla.mozilla.org/show_bug.cgi?id=1924562
Seems to get stuck in the "light" value when defined by `@property {}` 
Who ever thought that writing tests for under discussion browser features would be so relaxing...
:has-slotted, #LFG https://github.com/web-platform-tests/wpt/pull/48317
@kizu Caught an interesting bug here... seems that a parent with "transform" related styling will confused "position-anchor" resolution. 
https://issues.chromium.org/issues/364318220
If you'd love this to be fixed too, remember to add your "+1" to support Chromium Devs in understanding the priority of issues like these. 
@davatron5000 Are yall seeing:
this.attachShadow({
mode: "open",
referenceTarget: "real-input",
});
Work directly? I'm only able to achieve the relationship with:
this.shadowRoot.referenceTarget = 'real-input';
@Jamessw Issue, in case anyone following along is interested, can be found here: https://issues.chromium.org/issues/362780323
But, only when you use it right?
https://www.webpagetest.org/result/240819_BiDc7G_CY3/
More than a second of extra Blocking Time and LCP from their music "web app": https://www.webpagetest.org/result/240819_AiDcJV_E6Q/
Exciting: https://github.com/w3c/csswg-drafts/issues/10693#issuecomment-2289330553 
Love it when the WPT => WG => Resolution pipeline works exactly as it should. 
Anyone got some spec writing skills to share in moving the process through the next phase? 
WPT writing, for fun and profit: https://github.com/web-platform-tests/wpt/pull/47587
Who else is ready for `:has-slotted()`?
Non-controversial selector testing shipped! 
Now, let's get people checking out https://github.com/web-platform-tests/wpt/pull/47441 and its thoughts on `:host(:has(...)` (which great browsers like @webkit have already shipped) and `:host:has(...)` https://github.com/w3c/csswg-drafts/pull/10586 which will be really powerful when `::has-slotted(...)` starts to ship. 
A little Saturday #webPlatformTests writing session going on over here:
- https://github.com/web-platform-tests/wpt/pull/47440
- https://github.com/web-platform-tests/wpt/pull/47441
That second one's a little spicy with both unspec'd behavior based on mixed implementation and "tentative" tests for things no browser has started to support yet. 
@konnorrogers @scottjehl @knowler @Lukew @mrdisagree If you've never written a WPT, but want to get into the process, we've been doing some sharing session on that in the #webComponentsCG and would be excited to have you join.
Sadly, it's never _less_ work. 
But, hopefully it's fruitful work and we can make the world a better place for ourselves and everyone else.
Initial PR to #webPlatformTests to ensure that #elementInternals behaves the same x-browser in response to the above: https://github.com/web-platform-tests/wpt/pull/47233
How have you worked to make the web a better place today?
