Pixel art, but all colors in the palette are out-of-gamut and are <del>clipped</del> gamut-mapped to the best of your display's ability.

The older I get, the more ideas I come up with and collect, and the fewer spoons I have available at any given time. Surely, these 5 game ideas from 15 years ago will one day materialize into something :blobcatreading:

A Small Beach A photo from a walk on 15 June, 2024. 📷 OM-D E-M5 & 45 mm ƒ1.8, shot at ƒ7.1.

Once again, a project to which I wanted to maybe contribute some CSS/design, but which uses tailwind, so I'm out before starting anything. Good part: checking the source code before actually trying to do anything just to check if the code is contributable.

A bench. A photo from a walk on 15 June, 2024. 📷 OM-D E-M5 & 45 mm ƒ1.8, shot at ƒ8.

A pine. A photo from a walk on 15 June, 2024. 📷 OM-D E-M5 & 45 mm ƒ1.8, shot at ƒ8.

Hills, Goats, Ocean. A photo from a walk on 15 June, 2024. 📷 OM-D E-M5 & 45 mm ƒ1.8, shot at ƒ8.

Random landscape from around A Coruña. A photo from a walk on 15 June, 2024. 📷 OM-D E-M5 & 45 mm ƒ1.8, shot at ƒ5.6.

That feeling when publishing 45 photos, where you write descriptions and alts for all of them, alongside editing and optimizing for publishing, haha, oof. :blobcatfearful: Anyways! I just published photos I took a bit more than a year ago in A Coruña, during our #CssWg F2F outing […]

gradient.style has finally - come out of beta - offers multiple background editing - is open source 🎉 https://nerdy.dev/gradient.style-is-out-of-beta

I'll declare a holiday when WebKit rewrites their fragmentation layout engine. I wanted the block-level multicol overflow since 100 years ago, alongside many other things that improved fragmentation can bring, like the `box-decoration-break: clone` for columns etc etc etc :blobcatsadreach:

Once `reading-flow` is in all browsers, it would be interesting to look deeper into the `order` property and what we could do with it without the current drawbacks. For example, with it and `sibling-index()` we could achieve the “insert at an arbitrary index” use case: first, assign the `order […]

You give me custom functions in CSS — I am starting to think about callback arguments (including for regular properties), arrow functions, IIFE, polymorphism, recursion, and more. Unlikely we will get any of this lol, but I'll never stop dreaming big.

Would you rather write one longhand-sized shorthand or 100 shorthand-sized longhands?

Was thinking about masonry, and ended up wanting CSS Regions once again, as it goes.

A thought: it should be possible to use CSS mixins inside CSS functions, as you could want a mixin to return multiple custom properties that you could then use for some calculation inside a function.

That feeling when

Recently followed @outofprintarchive, and now my favorite thing is to look at some of those wacky print layouts of old game magazines and think through how you could replicate them with #css. I wish designers took more inspiration from print today — this is one of the things we lost as more and […]

There is currently a hot topic in WHATWG issues about how we can introduce a better, more flexible way to associate things. I am not yet sure about what could be the best solution there, but I thought about one thing that we could want to do alongside regardless of the chosen method (and maybe […]

Everyone needs more fragmentainers in their lives.

Playing with floats, found two bugs.

Opened a new CSSWG issue with a proposal for `container-var()`: https://github.com/w3c/csswg-drafts/issues/12604 — an ability to get values of custom properties from named containers. Mentioned in the issue: - article by Zell Liew - Issue about `inherit()` function by @leaverou - Issue about […]

It should be possible to combine multiple #css layouts on the same element. Define multiple grids on an element, and then reconcile their definitions. Define a grid area, and then say that this area is another grid, or a flex container, or a multicol content. And so on, and so on. All without […]

Do want: Web Preferences API, AND browsers with it telling me if some page is using any of the corresponding media queries. Like, if I choose to prefer reduced motion, then I want to know that page contained some motion that I'd want to re-enable to opt-in. Same with reduced transparency […]

Haha, the state of CSS results did put comments that contained “has” or “where” words into `:has()` and `:where()` buckets.

A great dive into how we can calculate the number of columns in simple #css grids with math by @anatudor: https://frontendmasters.com/blog/count-auto-fill-columns/ Very nice explanations and demos, as usual. One thing I learned was that `@property` respects its wrapping `@supports`, which is […]

The main thing that #css is missing and that external tooling like CSS Modules provides is a proper way to namespace things: class names, keyframes, custom properties, all other custom idents. Related issue: https://github.com/w3c/csswg-drafts/issues/11798 My latest proposal from there […]

Regular reminder that #css is already typed. We don't need any libraries for adding types to CSS or trying to add them through JS. It is just that the IDEs, tooling, etc. should pick up the slack and actually use that information. The types of non-registered custom properties can also be […]

Give me my CSS Regions, now! :blobcatknife:

I have a strong dislike for any JS-based API that requires you to use bitwise operators and magic number enums for configuration. Like… why? What purpose does it have aside from looking “smart”?

It is sad when a `<span>`, even with `display: contents`, changes how some styles are applied.

One of the least enjoyable lines to read in the specs is “there may be implementation-dependent limits” when you encounter a cross-browser difference for a thing you'd want to rely on.

Played a bit with `text-align: justify` and reported two bugs to Firefox.

Either I am doing something incorrectly, or it's the browsers: I would expect them to use the correct glyph for a soft hyphen (00AD) when it is present in the font file, but they seem to use the regular hyphen glyph instead regardless (002D). So, you cannot have a soft hyphen that is different […]

Thinking about typography

CSS custom highlight API sounds great, until you realize that it does not work with inputs, and that its JS API is really hard to use.

Oooh, tested Safari TP 222 on my anchor positioning and scroll-driven animations' demos, and almost everything works! Some very minor nits, but otherwise things seem pretty close (more so for scroll-driven animations, anchor positioning might be a bit further).

Years until anchor positioning will be usable in production: too many. (even as progressive enhancement)

An anchor positioning Chrome bug that prevents a bunch of its production usage: https://issues.chromium.org/issues/425894800 (mostly happens with anchor-center alignment, but I saw it happen without it as well)

Thinking about anchor positioning.

imagine if custom cascade layers had an ability to put stuff on top of the unlayered styles from the beginning…

I am sad about custom cascade layers, specifically that @layer b, c; @layer a, b; results in @layer b, c, a; and not in @layer a, b, c; Basically, we cannot use partial dependencies to “prepend” layers to others. I think at-rules it is fine to make them be added in order of appearance […]

Thinking about types in CSS.

The `prefers-reduced-transparency` is a nice media feature in CSS, time to start using it more.

Apple: now all your checkboxes are consistently round across all our products! And check this out — they're made out of glass!

Photos from the pre- #cssday event at the Amsterdam University of Applied Sciences on 4 June 2025. Present in the photos: Sanne 't Hooft, @vasilis@social.vasilis.nl, @nilsbinder@ruhr.social, and @mia@front-end.social. 📷 OM-3 & 75 mm ƒ1.8

Current music: https://www.youtube.com/watch?v=CRm4Y7TG05s

New article: “Passing Data into SVG: Linked Parameters Workaround” https://kizu.dev/svg-linked-parameters-workaround/ Twelve years ago, I started a draft for an article I never published: there were a few cross-browser issues and the idea was raw and […] [Original post on front-end.social]

Second day #cssday — as good, as the first one, and now I'm sad it ended. @css__cafe tomorrow, but for those who won't be there — see you next year, hopefully! It was great to see and talk to you!

Also, was happy to talk to everyone I talked to on the first day of #cssday. Hope to talk to more of you on the second! Tired! Already made a few pages of notes! Know which articles/posts I'll write next, but have no idea when to find time and spoons for those.