I make things for making things.
@leaverou@front-end.social on Mastodon, @leaverou elsewhere
https://lea.verou.me

@lea.verou.me
I make things for making things.
@leaverou@front-end.social on Mastodon, @leaverou elsewhere
https://lea.verou.me
Yup, CMYK isn't even the only exception. Once you consider print color spaces the sky is the limit, there are color spaces that model printing processes with 6+ colors. And color spaces for displays with more than three color emitters too. That’s why we went with an array of coords in Color.js!
Oh interesting, which PR? That must have been one of the other editors, I’d remember it if I had added it lol! Yes, so many magic numbers! A lot of it color science is derived by experimenting with humans and interpolating, which doesn’t tend to produce the cleanest math…
FWIW every browser has that particular gap 😅
Oh! I may have some passing familiarity with that spec 🙃
The naïve conversion is meant as a fallback when the color profile fails, so that content is displayable. device-cmyk() was never meant to be implemented via *just* the naïve algorithm (@svgees.us’s table of ΔEs + color checker explain why 😅)
What color profile does that use? device-cmyk() is meant to just send raw components to a CMYK device (eg a printer); without a color profile the naive sRGB conversion produces results that aren’t very useful (eg 100,0,0,0 becomes #00ffff, which is wildly outside the gamut of *any* CMYK printer)
David Sedaris on undecided voters
Related: a way to post issue replies private to maintainers, so they can collaborate on a draft before making it public. The post date would be the publication one. This has come up a lot in cases where replies need deliberation/collaboration & maintainers have to collaborate elsewhere (e.g GDocs).
@cassidoo.co GitHub wishlist, cont'd:
A way to designate multiple people as issue authors. This comes up a lot, but most recently in the TPAC breakouts repo, where multiple people can chair a breakout, but only the one that submits can edit its details: github.com/w3c/tpac2026...
What if there was a Web Components conference??
@stuffbreaker.bsky.social and I are teaming up to get a little more serious about the idea of getting one together.
We want to gather data on if there is interest in the community
Please fill out and share to anyone!
Modern Web Guidance v0.0.192 is out. With it your coding agent now knows how to: • build SMS one-time-code forms that autofill (WebOTP) • make dark mode toggles that follow the OS • honor the OS text size setting • reduce INP and lazy-load video/audio • feature-detect CSS at-rules with @supports
Link Parameters is the thing that gives you the ability to style embedded external SVGs.
Something many have wanted to do for a long time!
If you want to show your support of this, give these 2 issues a thumbs-up and comment:
See which features are missing (or not fully implemented) in specific browsers and go upvote/comment on their issues
Chrome: developer-signals.dev?browser=chro...
Edge: developer-signals.dev?browser=edge...
Firefox: developer-signals.dev?browser=fire...
Safari: developer-signals.dev?browser=safa...
Correction: @janeori.propjockey.io posted an even earlier proposal for the same concept, 4 years before! I’ve now updated the data to include hers first:
This allows creating declarations with variables etc that disappear entirely when invalid instead of becoming IACVT, paving the way for real block-level conditionals.
Wow, I hadn't realized the CSS revert-rule keyword (a feature I originally proposed, albeit with a different syntax) is now Baseline, as of Sep 14! 😍
Only 2y4m from initial proposal to Baseline.
1y8m if you count from the WG resolution!
Full timeline: lea.verou.me/specs/#rever...
Thank you! Didn’t you say that people are either non-monogamous or not? That’s what I was disagreeing with.
GitHub.com has fully moved off CSS-in-JS to CSS Modules, resulting in 55% less server-side render time and 25% faster component init.
https://github.blog/engineering/architecture-optimization/improving-site-performance-by-shipping-more-css/
Also, prescriptive != descriptive monogamy. One can be non-monogamous on principle and monogamous in practice (which is basically the opposite of cheating 😅). It’s about freedom and empathy, not some intrinsic need for greater partner diversity (some people may well have that, but it’s orthogonal).
Mononormativity works like any default you grew up inside: if exclusivity was presented to you as the only serious form of love, it takes real effort to even see it as a choice. That doesn't make people intrinsically monogamous or not. It makes major mindset shifts hard, which is a different claim.
Hard disagree. I believe non-monogamy is closer to the human default than we think: pair-bonding is real, but strict lifelong exclusivity is a cultural institution with a history, not a biological imperative. Look at how much work society puts into enforcing it, and how often it fails anyway.