12:03:01 RRSAgent has joined #i18n 12:03:06 logging to https://www.w3.org/2024/03/19-i18n-irc 12:03:17 Meeting: I18N ⇔ CSS 12:03:29 rrsagent, make log public 12:03:32 rrsagent, make minutes 12:03:33 I have made the request to generate https://www.w3.org/2024/03/19-i18n-minutes.html xfq 12:03:52 present+ florian, r12a, xfq 12:04:02 scribe: xfq 12:04:21 present+ fantasai 12:05:21 [Discuss the time slot of this meeting] 12:07:32 [Maybe figure out the details by making the time slot later for an hour or two] 12:07:40 [By email later] 12:07:48 agenda: https://www.w3.org/events/meetings/72d76e79-d8d4-4c3f-8707-fd4d3271e2ff/20240319T050000/ 12:07:48 clear agenda 12:07:48 agenda+ Agenda 12:07:48 agenda+ Action Items 12:07:48 agenda+ Info Share and Progress Reports 12:07:48 agenda+ Review on-going issues 12:07:58 r12a has joined #i18n 12:08:54 Topic: Action Items 12:08:58 https://github.com/w3c/i18n-actions/issues?q=is%3Aissue+is%3Aopen+label%3Acss 12:09:43 florian: you've been filing some requests directly and maybe we don't need this 12:09:57 ... at the same time, I don't think the CSSWG has dealth with any of them yet 12:10:03 s/dealth/dealt 12:10:25 https://github.com/w3c/alreq/issues/276 12:10:25 https://github.com/w3c/alreq/issues/276 -> Issue 276 Font fallback should allow selection of a Nastaliq font (by r12a) [gap] [doc:arfa] [p:basic] [x:webkit] [x:blink] [x:gecko] [x:alreq] [x:css-fonts] [x:arab-ks] [doc:arab_ks] [x:arab-ug] [i:fonts] [l:pes] [l:ur] [l:ks] 12:10:53 https://www.w3.org/TR/arab-ks-gap/#issue276_fonts 12:11:17 fantasai: I think we took some WG resolutions to add a generic function 12:12:01 https://github.com/w3c/alreq/issues/276 12:12:20 r12a: if you look at here ^ 12:12:34 ... I haven't heard anything from WebKit people 12:13:08 ... it's one thing to have the generic function, but the issue is how you match the fonts to that generic function 12:13:15 ... that's the next hurdle to get over 12:13:27 ... even though we have the generic syntax in the spec 12:13:43 florian: I don't know if this is something that's gonna happen in spec 12:13:57 ... it's not out of scope for you, but out of scope for the spec 12:14:33 ... I've just read the description of Nastaliq and there's one sentence I don't understand 12:14:54 "It is important not to fall back to a naskh style for languages such as Urdu and Kashmiri." 12:15:13 florian: @@1 12:15:23 fantasai: I think you should file an issue on that 12:15:38 florian: will do 12:16:26 #53 12:16:27 https://github.com/w3c/i18n-actions/issues/53 -> Action 53 come up with a set of information CSS want the i18n group to provide support for generic font families (on frivoal, fantasai) due 2023-11-01 12:17:01 r12a: I raised another request about a distinction in Khmer 12:17:23 https://github.com/w3c/csswg-drafts/issues/10036 12:17:23 https://github.com/w3c/csswg-drafts/issues/10036 -> Issue 10036 [css-fonts] Add generic font family, Khmer Mul (by r12a) [css-fonts-4] [i18n-needs-resolution] [i18n-sealreq] [css-fonts-5] 12:17:48 r12a: I'm holding off on requesting others until I see what happens with this 12:18:18 https://github.com/w3c/csswg-drafts/issues/10037 12:18:19 https://github.com/w3c/csswg-drafts/issues/10037 -> Issue 10037 [css-fonts] Add generic font family, Blackletter (by Crissov) [css-fonts-4] [css-fonts-5] 12:19:13 r12a: I'd have to go back and read it again and think about it but my initial thought is that this is unlikely to be a good candidate 12:19:43 ... because there aren't groups of people who can't read the text if it's not in blackletter 12:20:01 #35 12:20:02 https://github.com/w3c/i18n-actions/issues/35 -> Action 35 make the edits of CSS #5478 (on fantasai) due 2023-08-30 12:20:27 xfq: we discussed this issue last time 12:21:23 florian: @@2 12:21:45 ... I couldn't find a way to tweak the definition of the values with a less aggressive selector 12:21:55 ... so this provides the behaviour you want 12:22:42 r12a: my concern is that this should be the default behaviour and content author should not have to worry about it or remember that they have to 12:22:57 close #53 12:22:58 Closed -> issue #53 https://github.com/w3c/i18n-actions/issues/53 12:23:12 florian: if it's in the browser style sheet, then they don't have to do anything 12:23:30 r12a: it works in the browser, but might not work in EPUB documents or stuff like that 12:23:47 florian: an EPUB document is a browser from my point of view 12:23:55 ... it's a specialized one 12:24:05 ... but it's supposed to have the same user agent style sheet 12:24:15 ... if it doesn't, that's not spec compliant 12:24:25 r12a: they have other things like print 12:25:04 ... I don't think I've ever written a test or a WPT test that the browser has something in its default style 12:25:10 florian: it's perfectly valid 12:25:17 ... and no reason you can't 12:25:45 ... we can specify this normatively and we can test for it 12:26:44 #18 12:26:45 https://github.com/w3c/i18n-actions/issues/18 -> Action 18 Have informal explanation sessions about counter style translations with csswg members (on frivoal, fantasai) due 18 Jul 2023 12:27:18 florian: on one hand, if people are about to do this, we should try 12:27:30 ... on the other hand, @@3 12:27:46 r12a: yeah, nobody is pushing for it anymore 12:29:12 florian: is the conclusion Korean isn't unique? 12:29:54 r12a: there's a whole bunch of other non-modern script that do this 12:30:10 florian: 'auto' does the right thing 12:30:28 ... Korean is annoying because both behaviour are in modern usage 12:31:03 ... we can't just go with the auto keyword 12:31:10 ... because the auto keyword has to pick one 12:31:48 ... for ancient language not in common use that's not the case because you won't get a random author typing stuff in your comment form 12:32:29 ... I don't think that applies 12:33:21 ... Facebook is not going to have a policy about their favorite type of line breaking in Khitan Small Script being different from the default from Unicode 12:33:50 r12a: but we should not discount that possibility 12:34:40 florian: @@4 12:35:11 r12a: but we could also have people who expect to break on word boundaries and not on character boundaries 12:35:44 florian: the website owner decide the typographic style is blah blah blah blah 12:36:29 ... @@5 12:36:50 ... 2 views, one is Korean is unique in this situation and we just add auto-version-2 for Korean 12:37:25 rrsagent, make minutes 12:37:26 I have made the request to generate https://www.w3.org/2024/03/19-i18n-minutes.html xfq 12:38:19 ... @@6 12:38:33 ... we should not do the wrong thing just because it's easy 12:38:49 ... so if the right thing is to have something generic, then we should look into that 12:39:17 ... it's a question of complexity of the syntax and flexibility of the engine underlying it 12:39:35 ... and it's not inherently hard 12:39:51 ... there are way harder things in the web platform 12:40:12 r12a: presumably the best way to test that is to put a proposal together 12:41:18 #11 12:41:18 https://github.com/w3c/i18n-actions/issues/11 -> Action 11 Triage all css properties to determine which are logical, physical, or na by default (on frivoal) due 18 Jul 2023 12:41:24 Topic: Info Share and Progress Reports 12:41:46 https://github.com/w3c/html-ruby 12:41:56 florian: there's a ruby repo with the spec in it now 12:42:06 ... rb and rtc 12:42:42 ... the repo is set up 12:42:48 ... the chairs of the WG is aware 12:43:07 ... the rb feature is effectively ready to go to REC 12:43:20 ... because Firefox and Kindle supposed to support it 12:43:31 ... However, I don't know how to test Kindle 12:43:46 ... Addison has an action item 12:44:07 ... no browser other than Firefox supports tabular markup 12:44:23 ... you can't do tabular markup without rb 12:44:42 ... there is something in the HTML spec that nobody implements 12:45:00 ... multi-level ruby using the rtc model 12:46:12 r12a: I thought if you have a piece of content and then rt rt then one of them gets put above and one of them gets put below 12:46:26 florian: the spec claims that but I believe nobody implements that 12:46:29 r12a: OK 12:46:45 florian: @@ 12:46:57 ... this spec basically corresponds to what Firefox does 12:47:23 ... and it corresponds to what the CSS spec has been assuming HTML would eventually do 12:50:13 ... I also made today a presentation for JEPA 12:50:25 ... about ruby 12:50:38 ... Japan Electronic Publishing Association 12:51:37 ... in a publication context you might be able to say I achieved the book I wanted to achieve 12:51:49 ... it works, but also it doesn't 12:52:05 ... that's the point I've been trying to drive 12:52:45 ... unless you get the right patterns of ruby markup, you can't get some features work 12:53:13 ... I think the accessibility requirements from the government and the demands on textbooks have made people actually try to do things 12:53:49 r12a: if you can write down some of the things that don't work with real examples from people who are doing real publishing 12:53:55 ... that would be terrific 12:54:52 florian: I was also hearing from someone other than Murata-san about the text to speech of electronic documents containing ruby 12:55:21 r12a: talking about ruby, I'd rather stay with CJKM for the time being 12:55:58 florian: let's first work on the markup pattern 12:56:05 r12a: that's good 12:58:22 florian: we should also start thinking a little about where this goes because the WG chartered until the end of June or so 12:59:45 r12a: let's discuss with plh first 13:01:34 rrsagent, make minutes 13:01:36 I have made the request to generate https://www.w3.org/2024/03/19-i18n-minutes.html xfq