2026-01-05 [06:42:49.0359] I vaguely recall some recent-ish conversation _somewhere_ about defining or codifying the behaviour of "best fit" locale matching, but now I can't find any record of it. Am I misremembering things? Context here is that I happened to notice that no browser seems to fall back from `hak` to `zh`, even though the latter is a macrolanguage for the former, and CLDR encodes that relationship as a [languageMatch](https://github.com/unicode-org/cldr/blob/da8921cb81198fdab8ee927eb2cdaf6d38c1d15d/common/supplemental/languageInfo.xml#L369), and I'd like to figure out what the right place might be for fixing that. [09:20:05.0164] not that I recall, but I suspect we could defer to a lot of https://unicode.org/reports/tr35/#LanguageMatching while still allowing implementation-defined weights/thresholds/etc [14:27:02.0030] > <@eemeli:mozilla.org> I vaguely recall some recent-ish conversation _somewhere_ about defining or codifying the behaviour of "best fit" locale matching, but now I can't find any record of it. Am I misremembering things? > > Context here is that I happened to notice that no browser seems to fall back from `hak` to `zh`, even though the latter is a macrolanguage for the former, and CLDR encodes that relationship as a [languageMatch](https://github.com/unicode-org/cldr/blob/da8921cb81198fdab8ee927eb2cdaf6d38c1d15d/common/supplemental/languageInfo.xml#L369), and I'd like to figure out what the right place might be for fixing that. We recently discussed https://github.com/tc39/ecma402/issues/947 which is somewhat related, but I don't recall discussing codification of BestFit 2026-01-14 [12:54:35.0407] @sffc:mozilla.org @dminor:mozilla.org i think we have a webcompat issue in CLDR 48: https://issues.chromium.org/issues/475851786 [12:55:30.0858] haven't had a chance to test Firefox yet, do you what CLDR version Firefox is shipping with @dminor:mozilla.org ? [12:55:44.0809] * haven't had a chance to test Firefox yet, do you know what CLDR version Firefox is shipping with @dminor:mozilla.org ? [12:58:05.0384] We just landed an update to ICU78 in Firefox 147 (https://bugzilla.mozilla.org/show_bug.cgi?id=2000225), that's probably just going to release now. [12:58:18.0576] That should be using CLDR 48. [12:59:25.0493] interesting, would be interested to hear if you run into any compat issues. we found this in pre-release testing [12:59:52.0857] our preference is for CLDR to roll back this change [13:46:36.0198] Looks like Firefox is affected as well, I've raised this with webcompat and I'll file a bug. [14:18:53.0342] For reference, here's the Firefox bug: https://bugzilla.mozilla.org/show_bug.cgi?id=2010411 [15:39:37.0330] It's too late for CLDR 48 to "roll back" the change; they even already did a patch release 48.1 with minor fixes. It's possible a change could still be considered for 49.0 but the data freeze is in one week. [15:44:05.0430] But, what about this change does shu feel makes it more likely to break the web than other changes? Is it because we have evidence that it breaks the web, or is it based on a heuristic like being in English? [15:46:05.0458] I see, there's a link to https://order.jetspizza.com/tx026 in the chromium issue. 2026-01-15 [16:00:00.0136] i'm pretty sure the culprit is an old version of the date-fns library (not that old, from 2024): https://github.com/date-fns/tz/commit/213903702d7c5fcd4f01479ba7370fe917195a65 this library is fairly popular: https://www.npmjs.com/package/date-fns [16:00:16.0668] so i'm more confident that this is likely to manifest in more places than that one pizza website [16:03:53.0015] interesting. It suggests that the impact might be narrow to en-GB, at least based on the evidence we have so far. [16:06:53.0077] i agree, based on the evidence so far [16:10:21.0535] the parsing code switched from using a fixed offset to assuming the substring "GMT", which is fine for now but I know CLDR has considered proposals to change it to "UTC" instead. I don't think those proposals are currently active. [16:16:21.0396] I opened https://unicode-org.atlassian.net/browse/CLDR-19205 [08:18:15.0640] 👋 please lmk who will be presenting the update at plenary next week and timebox required [09:09:40.0526] adding to agenda after the numercs meeting -- thought it was already there [09:09:50.0738] * adding to agenda after the numerics meeting -- thought it was already there 2026-01-21 [14:45:40.0657] https://github.com/unicode-org/cldr/pull/5309 is landed. shu dminor 2026-01-27 [09:56:07.0712] when is the next 402 mtg? [11:33:38.0806] Need to schedule it, I'll go ahead and do all the ones through July with the assumption of a 10 day TG1 agenda deadline