Jake ArchibaldSep 4, 2025

Interop 2026 is now open for proposals!

This is your chance to tell browser-makers which well-defined, well-tested features you wish had better support across browsers.

⬇️ Here's how ⬇️
https://github.com/web-platform-tests/interop/blob/main/proposal_guide.md

Note: this is only for features that are already fully designed & specified. If you have ideas for entirely new features, you can still propose those on GitHub projects related to the feature, e.g https://github.com/whatwg/html/ for HTML, and https://github.com/w3c/csswg-drafts/ for CSS.

Show threadPaul Robert Lloyd 🐢Sep 4, 2025
@jaffathecake This is your chance to tell browser-makers which well-defined, well-tested features you wish had better support across browsers… and for them to ignore you and do what they’d already planned to do anyway 😒
Show threadJake Archibald
@paulrobertlloyd what do you feel was ignored last year?
Sep 4, 2025 at 6:02pmWeb
Show threadPaul Robert Lloyd 🐢Sep 4, 2025
@jaffathecake Speak to @Richr about the countless typography issues that are ignored. Or the poor support for voice styles. Any spec that wasn’t written in the last 3 years is seemingly ignored, the whole thing seems to be less about interoperability but implementing the new hotness.
Show threadVasilisSep 4, 2025
@paulrobertlloyd @jaffathecake @Richr Agreed! Multicolumn could use a lot of attention, print-stylesheets are also unpredictable. And what about the ridiculous oklch implementation, which is _still_ not fixed. Implemented in a hurry, and just left there.
Show threadJake ArchibaldSep 4, 2025
@vasilis please make proposals for these things, and send me the links so I can follow-up
Show threadVasilisSep 4, 2025
@jaffathecake Alas, I have no time to make such proposals.
Show threadLukeSep 4, 2025
@vasilis @jaffathecake It's as simple as filing a GitHub issue fwiw. There's nothing more to it. If things don't get proposed they'll definitely get ignored.
Show threadVasilisSep 4, 2025
@Lukew @jaffathecake It seems like a bit more than just filing a simple issue, when I read the Interop Proposers Guide, but I’ll see what I can do.
Show threadRichard RutterSep 5, 2025
@jaffathecake @vasilis will do Jake.
Show threadRichard RutterSep 8, 2025

@jaffathecake @vasilis Here you go Jake. 5 #interOp submissions in order of priority.

https://clagnut.com/blog/2445

My requests for Interop 2026

It’s the time of year again when browser makers ask which shiny new features they should implement in preference to fixing outstanding bugs. Despite my cynicism, I’m trying again with these submissions. They’re mostly typographic but in some cases important.

Show threadJake ArchibaldSep 8, 2025
@Richr @vasilis ohhhh, thanks for making this a blog post too!
Show threadAna Tudor 🐯Sep 4, 2025

@jaffathecake `background-clip: text`. This bug is now old enough to go to school, it's been over 7 years since I reported it https://bugzilla.mozilla.org/show_bug.cgi?id=1481498

A use case for this https://css-tricks.com/taming-blend-modes-difference-and-exclusion/#aa-text-state-change-effect

The `filter()` function. Safari has supported this for over a decade now https://bugzilla.mozilla.org/show_bug.cgi?id=1191043

`line-clamp`, `initial-letter`...

Not even bringing up SVG because there's so much there that has been under "in the future, we'll be able to" for over a decade.

@paulrobertlloyd

1481498 - background-clip with multiple values fails when one value is text

NEW (nobody) in Core - Web Painting. Last updated 2024-12-18.

Show threadEric EggertSep 4, 2025
@anatudor @paulrobertlloyd But you don’t understand! This is to make sure other browsers implement Chrome’s features, not for Chrome to implement other browser’s features! *SCNR*
Show threadRichard RutterSep 6, 2025

@anatudor @jaffathecake @paulrobertlloyd 7 years!? Pah, a mere blink in time compared to `break-after:avoid` - 28 years and counting!

See: https://clagnut.com/blog/2426

Pagination widows, or, Why I’m embarrassed about my ebook

The handling of widowed headings across pages in Apple Books is of particular concern. Since 1997, CSS has had properties to handle this, and yet browsers including Safari and Firefox still don’t support them – why not?

Show threadpwheelerSep 5, 2025
@jaffathecake @paulrobertlloyd The Big One is JXL. Snubbed I think at least 2 times!
Show threadJake ArchibaldSep 5, 2025
@pwheeler @paulrobertlloyd my personal feelings about that is it isn't worth it vs existing formats, but it wouldn't upset me to see it implemented.
Show threadpwheelerSep 5, 2025
@jaffathecake Genuinely curious as to why you feel that way? Progressive decoding, lossy & lossless, animation, very high max res--the backwards compatibility alone is killer imo! It seems to be on-par(or better) with most "modern" formats & is more flexible/future-forward.
Show threadJake ArchibaldSep 5, 2025

@pwheeler depends what you mean by backwards compatibility. It isn't really backwards compatible, it has a separate feature where it can repackage a JPEG to be a little smaller, but once you do that it can't be read by a JPEG decoder.

The lossy mode doesn't seem as good as AVIF. Progressive is interesting. Lossless is ok but fairly rare to need that on a web page.

Show threadpwheelerSep 5, 2025
@jaffathecake Yea I guess that's not really what backwards compatible means lol, but the conversion is lossless and it reduces space! It's maybe not outperforming every scenario, but that it's even competitive is so exciting to me. I'm thinking of it as more of a feature-rich, portable image standard and less as a technology for the web specifically.
Show threadHazel CoraSep 6, 2025
@jaffathecake @pwheeler from the admittedly little I’ve seen, the progressive loading is exciting to me to have along with better compression than JPEG and an alpha channel (I’m unaware of other actually-supported formats than JPEG with good progressive loading, would love to know of others if they exist though!), but in practice I get by with other formats just fine. It being “pulled” from Chrome (…when it was never even in Chrome except behind a flag…) did not warrant the response it got in the slightest and it’s weird that it’s still going
Show threadpwheelerSep 6, 2025
@h @jaffathecake Oh I think it did! Chrome has the highest browser share and their provided explanation for removing support(as you said, JXL had always been behind a flag) was due to "low interest/adoption", to which interested devs objected. At some point, I think someone from the team posted an egregious "case study" that was just...objectively, very poorly done. Which ofc leads to questions about bias, competing formats, etc.