#1101: env(preferred-text-scale)

Visit on Github.

Opened May 22, 2025

こんにちは TAG-さん!

I'm requesting a TAG review of env(preferred-text-scale)

Operating systems and web browsers provide global accessibility settings for the user to increase their system text scale. Authors would like to respect the user’s text scale settings but have trouble doing so today. We propose to provide authors with access to the user’s system font scale factor by adding a new CSS environment variable -- env(preferred-text-scale).

Further details:

You should also know that... Past discussions have included an accompanying pem CSS unit that would provide better ergonomics than the environment variable in certain situations. We intend to pursue that separately, as questions about its exact behavior have been recently raised (https://github.com/w3c/csswg-drafts/issues/11636#issuecomment-2840207330). These questions do not pertain to env(preferred-text-scale).

Discussions

Discussed Jun 2, 2025 (See Github)

Jeffrey: Lots of discussion on the blink-dev thread, about how to get mobile and desktop to behave the same, and other details, which concluded that this is the best we can get. TAG could definitely double-check that.

[Review async]

Discussed Jun 9, 2025 (See Github)

Christian: it looks fine from my pov... some concerns jeffrey raised are also valid... ask these questions back to the proposers?

christian to pose Jeffrey's comment and then we see if we can close next week.

Discussed Jun 16, 2025 (See Github)

Xiaocheng: Christian already closed this out. Happy with the outcome.

Discussed Jun 16, 2025 (See Github)

Christian: Jeffrey and I have looked here, we would make as satisfied with concerns regarding accessibility.

torgo: Don't suggest to keep this open but punt to APA?

Christian: it is on the agenda for tomorrow, if everyone is happen then you could just post a comment

torgo: If we want to do it this week, Matthew has already given regrets for this week so we should just make a call

Jeffrey: I think this is ready to post and ready to close. The feature is an improvement even from an accessibility standard, the concern is that there's maybe a better way to do all od this that needs to be designed. But it's not this feature, there's no tweaks to this feature to accomplish what they need to actually do. This is fine. Given the big fix won't happen soon this is a good short term fix

torgo: I think we should close this in the interest of moving ahead

Hong Kong meetup

Lola: are there updates on budget?

Jeffrey: There's no update on funding, questino came up briefly during AC meeting uesterday. Still need to write email to Seth. W3C budgeting is halfway done

torgo: W3C has Past Submitter Status, W3C feeds into the more traditional international standards ecosystem like ISO standards. Also feeds into EU stuff

Comment by @christianliebel Jun 18, 2025 (See Github)

Thanks for your proposal. The TAG believes that it significantly improves text scaling on the web. However, we think that the accessibility requests for non-linear scaling have not been prioritized adequately. We'd appreciate it if the CSSWG could work with the AGWG and/or the APAWG to co-design a scaling system that'll do a better job of avoiding too-large headings.

The factor preferred-text-scale, although bucketed, may introduce a fingerprinting bit and identify visually impaired users. However, this information could also be derived from other sources.

Because we consider the improvement in legibility important, we conclude this review as satisfied with concerns. This reflects the consensus of the TAG.