Repository navigation
unprefix the -webkit-user-select CSS property #1000
Description
Description
The user-select CSS property still needs a -webkit- prefix in Safari.
A note about the current state of CSS vendor prefix’s
- Chrome, Edge and other Chromium-based browsers had required a
-webkit-prefix forprint-color-adjust. This is no longer the case as of version 136. - As part of Interop 2025, Safari removed the need for the
-webkit-prefix for thetext-decorationCSS shorthand property.
That leaves only a handful of properties requiring a prefix:
user-selectstretch: Implemented as-webkit-fill-availablein WebKit and-moz-availablein Firefox.text-size-adjust: Needs-webkit-prefix in WebKit and Firefox. (Firefox also supports the-moz-prefix.)box-decoration-break: Needs-webkit-prefix in WebKit.line-clamp: Needs a-webkit-prefix and some additional non-standard properties (display: -webkit-boxand-webkit-box-orient: vertical) to be set in Chromium, WebKit and Firefox.
Autoprefixer is downloaded from NPM 33 million times per week. Needing to complicate build tooling with Autoprefixer for such a tiny handful of properties is an unfortunate situation.
Of those properties that do require a prefix, some are pretty niche, but that can’t be said for user-select. While line-clamp is another very useful and popular CSS feature that requires a prefix, Autoprefixer doesn’t prefix line-clamp, so it is primarily user-select that necessitates continued use of Autoprefixer). While ideally no property would require a prefix, for the reason just stated user-select in particular should be prioritised.
Specification
https://drafts.csswg.org/css-ui-4/#content-selection
web-feature
https://web-platform-dx.github.io/web-features-explorer/features/user-select/
Test Links
https://wpt.fyi/results/css/css-ui?label=master&label=experimental&aligned&q=user-select
Activity
github-actions commentedon Sep 4, 2025
This comment was automatically generated based on the information you provided. Please don't edit it.
Below is additional information about the web feature (from the web-features project) which is referenced in your proposal.
If this doesn't accurately correspond to your proposal, please update your initial comment to include
web-features: <feature-id>.To find feature IDs, use the web-features explorer.
Feature user-select
user-selectCSS property controls which elements can be selected by the user.foolip commentedon Sep 5, 2025
@o-t-w thank for submitting this proposal! I took the liberty of editing your description to link https://web-platform-dx.github.io/web-features-explorer/features/user-select/, in order to help the bot (above) pick out the right feature for this.
[-]unprefix the -webkit-user-select CSS property[/-][+]unprefix the `-webkit-user-select` CSS property[/+]dholbert commentedon Dec 9, 2025
One suggested refinement to the Test Links here:
Right now the tests exercise the
containvalue foruser-select, but no browser supports that keyword (and it's not mentioned above), so I suspect we're not intending for this interop focus-area to be a forcing factor on all browsers implementing it.Specifically these tests have subtests that check for
user-select:contain(which fail in Chromium/Gecko/WebKit):https://wpt.fyi/results/css/css-ui/parsing/user-select-computed.html
https://wpt.fyi/results/css/css-ui/parsing/user-select-valid.html
https://wpt.fyi/results/css/css-ui/user-select-inheritance.html
So: if this proposal is adopted, I suggest that we either:
(A) fork these^ tests so that
user-select: containis tested in a separate tests (which will not be part of the interop set of tests). Maybeuser-select-computed-contain.htmlwhich is a copy ofuser-select-computed.htmljust to test that one value, for example. (Slightly clumsy, but we do it sometimes for the benefit of the interop project.)...or:
(B) Just remove those 3 tests entirely; they're basic smoke-tests for properly handling the property in CSS, but the remaining tests likely cover that pretty well, too.)
Maybe for now it's best to just plan for (B) since it's simplest, and we can do (A) if & when the proposal is adopted, if we want to add a bit more test coverage.
frivoal commentedon Jan 29, 2026
-webkit-user-selectis required (for compat) by the spec:If we have evidence that it is not actually needed for compat, we should drop it from the spec AND the tests. Otherwise, it would probably be a good idea to test usage of prefixed and unprefixed separately, so splitting/rewriting the tests would makes sense, but I don't think it's appropriate to just remove the prefixed version from all tests generally.
wpt-interop commentedon Feb 12, 2026
Thank you for proposing unprefix the
-webkit-user-selectCSS property for Interop 2026.We are pleased to let you know that it has been accepted as part of the Web compat focus area.
You’ll be able to track progress on this topic throughout the year via the Interop 2026 dashboard.
Posted on behalf of the Interop team.