position-area - should be marked partial for Chrome?

[jakearchibald]

Related web-platform-dx/web-features#3382

The partial support in position-area subfeatures in Chrome feel severe enough to warrant 'partial support' for the feature as a whole. Right now, it gets a green tick on the baseline banner on MDN.

Previous discussion #27573

[jakearchibald]

As per https://github.com/mdn/browser-compat-data/tree/main/docs/data-guidelines#partial_implementation-general-usage-guidelines

The browser's support does not implement mandatory specified behavior.

True.

The browser's support is inconsistent with at least one other browser.

True, Safari.

The browser's support causes confusing feature detection results.

True. It isn't a missing feature that can be detected with eg @supports.

The browser's support has a demonstrable negative impact on web developers.

Yes. It isn't immediately clear that the feature doesn't work as advertised.

[caugner]

We have discussed this in the BCD meeting on 2025-09-30, and took the following conclusion in the BCD meeting on 2025-10-14 (today):

1. We keep the support data for position-area in BCD as is.
2. @ddbeck will raise an issue on the web-features side to ensure Chrome's limitation is reflected in the overall status.

The main reason for (1) keeping Chrome as supporting the position-area property in BCD is that Chrome supports at least one property value fully. We aim to avoid duplicating editorial decisions from web-features that imply partial or no support for a curated set of subfeatures results in overall partial support for the parent feature.

Once the web-feature data (2) reflects the Chrome limitation, we will take a look at how the Baseline banner presents itself. Daniel will also talk to Ruth et al. about how web-features are presented on MDN.

[ddbeck]

On web-features, I'm going to continue the discussion via this issue: web-platform-dx/web-features#3382. I've also put this on the agenda for my next meeting with Ruth, to talk about presentation on MDN. I don't think there's anything left in-scope for BCD here, so I suggest we close this (for now, at least).