romainmenke
banner
romainmenke.bsky.social
romainmenke
@romainmenke.bsky.social
Gender is a spectrum.

Fixes things, mostly with code.

https://www.romainmenke.com/
It's pretty dynamic on our end, such bots almost always trip over one of our honey pots and get banned automatically.
October 30, 2025 at 6:32 PM
same here
October 30, 2025 at 4:33 PM
Reposted by romainmenke
Here's a simple postcss-preset-env demo showing how to switch from Browserslist's `defaults` query to Baseline, and seeing how that affects the results

github.com/GoogleChrome...
baseline-demos/tooling/postcss at main · GoogleChromeLabs/baseline-demos
Contribute to GoogleChromeLabs/baseline-demos development by creating an account on GitHub.
github.com
September 16, 2025 at 6:52 PM
Thank you for taking a look!
I've submitted a bug report: bugs.webkit.org/show_bug.cgi...
299084 – Overflow hidden on body/html no longer works
bugs.webkit.org
September 18, 2025 at 12:33 PM
@nicolesullivan.bsky.social I've also submitted this through feedback assistant for Safari, but maybe you can take a look?

We were very surprised to find that Safari 26 effectively breaks all our modal dialogs in all our projects :)
September 17, 2025 at 7:28 PM
They "solved" this on apple.com by setting height to 100% on the body and the html element when the dialog is open.

This implies a jump to the top of the page each time you open and close the menu.
September 16, 2025 at 2:57 PM
Realized that even when I want to ignore the search engine "ai" answers/summaries, my brain still reads it in the few seconds it takes to scroll away. And yes when it is an unfamiliar topic all the misinformation is happily consumed :/ Added a browser extension to hide it completely.
July 24, 2025 at 6:19 PM
I think the spec is pretty nice and covers the use cases well. I am unsure about a JS API. Haven't needed it myself, so I can't speak to it.

On why I think it is important see the thread here: github.com/w3c/csswg-dr...
[css-contain-3][css-mediaqueries-5] Unified custom queries · Issue #7622 · w3c/csswg-drafts
The spec for Media Queries Level 5 currently specifies a way to define custom media properties like e.g. @custom-media --modern (color), (hover); with CSS Containment currently getting shipped befo...
github.com
July 24, 2025 at 1:48 PM
Yes, I was equally disappointed when I realized the implications of this :)
July 14, 2025 at 9:02 AM
If you have any concrete use cases, feel free to add them in the issue :)
July 14, 2025 at 6:42 AM
When using a typed argument with a default value: codepen.io/romainmenke/...
July 14, 2025 at 6:19 AM
Property registrations from outside the CSS function do not exist inside of it.

Inside a custom function all props are untyped unless they are typed arguments or a typed result.
July 14, 2025 at 6:18 AM
# c2f5fc

Sunset with just a few cloudy streaks. Everyone is always excited about the fiery streaks but I like the bits in between.
June 1, 2025 at 6:00 PM
Using layers does allow for some really neat and clean source code. We really liked the structure.

Nothing really wrong with using layers, it's more that we don't need them while also experiencing some drawbacks.

So not using cascade layers is a better fit for us.
May 19, 2025 at 1:08 PM
Needing to repeat base styles in each cascade layer seemed a bit silly.

This is just one example, but the most straightforward to explain.

Having full control/ownership over all styles we find that we don't need layers.
May 19, 2025 at 1:08 PM
We add "[hidden][hidden] { display: none; }" in our base styles.
Using the double attribute to get a double class specificity.

When layered as "reset, base, components, ..." we found that components with ".foo { display: grid }" are always visible even when they have the "hidden" attribute.
May 19, 2025 at 1:08 PM