23:30:34 RRSAgent has joined #epub 23:30:34 logging to https://www.w3.org/2021/02/04-epub-irc 23:30:44 Meeting: EPUB 3 Working Group Telecon 23:30:50 Date: 2021-02-04 23:31:45 dauwhe has changed the topic to: https://lists.w3.org/Archives/Public/public-epub-wg/2021Feb/0000.html 23:57:12 MattChan has joined #epub 23:58:48 shiestyle has joined #epub 23:59:21 sorry, but read only today. 23:59:30 toshiakikoike has joined #epub 23:59:45 present+ 00:00:08 MasakazuKitahara has joined #epub 00:00:16 present+ 00:00:21 present+ 00:00:24 present+ 00:00:53 BenSchroeter has joined #epub 00:01:34 Teenya has joined #epub 00:02:31 marisa has joined #epub 00:03:33 present+ 00:05:11 duga has joined #epub 00:05:18 present+ 00:05:38 wendyreid has joined #epub 00:05:39 present+ 00:05:45 present+ 00:05:54 scribe+ 00:06:05 dauwhe: Welcome everyone, today we'll be talking about internationalization 00:06:17 ... Ivan went through the i18n questionnaire 00:06:38 ... filed a number of issues he encountered while performing the review 00:06:50 ... some of the members of the WG replied 00:06:59 ... we should go through the issues and make some decisions 00:07:02 https://github.com/w3c/epub-specs/issues/1499 00:07:07 ... first one is #1499 00:07:14 https://github.com/w3c/epub-specs/issues/1491 00:07:15 present+ 00:07:16 issue, 1499 00:07:24 https://github.com/w3c/epub-specs/pull/1498 00:07:27 dauwhe: Also going to talk about 1491 because they're related 00:07:34 ... there is also a PR addressing this 00:08:02 ... most of the i18n issues we face are in HTML, SVG, CSS, where we rely on those specs 00:08:12 ... our area of influence is the package doc 00:08:15 ... the dir element 00:08:21 ... it doesn't have a default value 00:08:34 ... some guidance from i18n says we should have a default of "auto" 00:09:02 ... lets the system decide what to do, which depends on them, and can be overwritten by rtl or ltr 00:09:11 ... we have a PR open to address this 00:09:17 ... Matt has added this 00:09:28 ... is this a good idea and will it break anything 00:09:42 ... ltr = left to right 00:09:50 ... rtl = right to left 00:10:01 q? 00:10:15 duga: I think there is some history we want to keep with HTMLs LTR by default 00:10:27 ... for our case it makes sense to use auto because that's how things work now 00:10:33 ... the default is auto 00:10:57 ... HTML doesn't have a default, but the root is LTR, but it also has more complex inheritance 00:11:05 dauwhe: I don't want to worry about thr HTML situation 00:11:11 ... there's complexities 00:11:28 ... it seems to me that adding auto is risk free as a default 00:11:33 q? 00:11:37 ... I can't imagine writing a test where something would happen 00:12:08 duga: The scarier thing is the inherent directionality of the text winning over the dir element 00:12:14 dauwhe: I wonder that too 00:12:27 ... went down a rabbit hole 00:12:38 ... stared into the abyss... 00:13:02 duga: If there's an inherent directionality, it wins over dir, dir might mean nothing 00:13:16 dauwhe: base paragraph direction over the bits of text within 00:13:28 ... atomic elements of the package file that have no instructions 00:13:37 ... instinct to have auto as a nice to have 00:13:42 ... doesn't change existing behaviour 00:13:53 ... the only person who needs to worry is the tester 00:14:04 ... I would propose we resolve to add the auto value 00:14:17 ... does anyone object 00:15:29 present+ Garth 00:15:31 Proposed: Add the default value of "auto" to the dir element, resolved 1491 and 1499 00:15:43 Resolved: Add the default value of "auto" to the dir element, resolved 1491 and 1499 00:16:05 https://github.com/w3c/epub-specs/issues/1494 00:16:13 dauwhe: We got it! Moving on to 1494 00:16:42 ... i18n self test says *what is in the ticket* 00:16:56 ... spec specifies that directionality in unicode takes precedence 00:17:00 ... over the dir element 00:17:20 ... Matt says if there's strong directionality in the text, that takes precendence 00:17:26 ... dir doesn't control everything 00:17:36 https://github.com/w3c/epub-specs/issues/1494#issuecomment-772459326 00:17:44 ... r12a comments (see link) 00:18:29 ... what Richard says there makes sense 00:18:49 ... we need to remove that line in the spec stating that unicode directionality takes precedence 00:18:57 ... it would conflict with the guidance from Richard 00:19:14 ... address the issue by having Matt removing that bit and clarifying the text 00:19:26 ... the relationship between the string and the dir attribute 00:20:21 ... any opposition? 00:20:29 BenSchroeter: The change being that dir wins? 00:20:40 dauwhe: Removing the line that says strings can override dir 00:20:54 duga: I don't think anyone knows what it means 00:21:08 ... dir sets the base direction 00:21:21 ... but it has nothing to do with unicode for instance 00:21:27 ... the unicode API does that 00:21:31 ... we're reaching 00:21:41 ... we should just say "this is what dir does" 00:22:09 dauwhe: This is a lesson, if you get that deep into it, it's confusing 00:22:46 Proposed: Fix 1494 by removing the untrue statement about the relationship between dir and the underlying text 00:23:18 q? 00:23:29 https://github.com/w3c/epub-specs/issues/1493 00:23:36 dauwhe: We'll resolve by PR 00:24:15 ... 1493, should provide data for the interpretation of text 00:24:25 ... we are stuck with the DC terms language 00:24:30 ... which we can't control 00:24:37 ... we've used "refines" before 00:24:59 ... there's a lot of examples from Japan regarding sorting and pronunciation of names 00:25:05 ... we can't make this change to EPUB 00:25:11 ... we have these mechanism 00:25:31 ... if there's an important use case comes up that doesn't fit with what we already have 00:25:45 ... we'll address it, but we shouldn't try to create something without that information 00:25:49 duga: I agree with Matt 00:25:55 dauwhe: Me too 00:25:59 ... anyone else? 00:26:16 ... I would propose we close the issue, there's nothing for us to do here 00:26:31 Proposed: Close 1493 as we cannot fix this issue 00:26:44 Resolved: Close 1493 as we cannot fix this issue 00:27:37 https://github.com/w3c/epub-specs/issues/1490 00:28:38 dauwhe: 1490, do not assume that direction can be assumed from language information 00:29:02 ... but then we have PR 1498 00:29:22 duga: If we accept 1498, we don't care about this, according to the comments 00:29:45 dauwhe: Not to mention the i18n folks who have commented 00:29:50 ... this one is kind of moot 00:29:57 duga: Will not fix, obsolete 00:30:21 Proposed: close 1490 because mooted by PR 1498 00:30:44 https://github.com/w3c/epub-specs/issues/1489 00:31:14 dauwhe: 1489, it must be possible to indicate changes in bidirectional text for the user 00:31:22 ... we're talking mostly about DC terms here 00:31:28 ... there's no inline markup 00:31:37 ... there's unicode marker characters for this 00:31:58 ... should we add some explanation that you need to use the unicode characters to do it 00:32:10 ... I'm generally in favour of putting useful info into the spc 00:32:14 s/spc/spec/ 00:32:28 ... provides clues to us when we go back to this 00:32:45 duga: I think that's the reason we added the comment about inherent directionality 00:32:56 ... if we start talking about unicode conformance it can get confusing 00:33:09 ... we could look at HTML and see what they say 00:33:19 dauwhe: is it safe to have non-normative text? 00:33:35 duga: The comment was a brief overview of a complicated issue 00:33:43 ... we could put a generic pointer 00:34:12 dauwhe: We should probably avoid this as we don't understand it well 00:34:22 ... something obvious to us might be wrong, unlikely to be helpful 00:34:38 ... this is not our lane 00:35:27 duga: Is there a lot of usage for this? 00:35:54 dauwhe: I've so far been unable to change behaviour of a RS by changing the dir attribute 00:36:24 ... I would propose we close this, as we do meet the requirements 00:36:49 Proposed: Close 1489, we have met the i18n requirements 00:37:21 Resolved: Close 1489, we have met the i18n requirements 00:38:31 dauwhe: Moving on... 00:38:38 https://github.com/w3c/epub-specs/issues/1488 00:38:40 ... 1488 00:38:49 ... missing href-lang attributre 00:39:13 ... *see issue* 00:39:21 ... there is a PR from Matt 00:39:22 https://github.com/w3c/epub-specs/pull/1497 00:39:35 ... PR 1497 00:39:48 ... this is designed for the links in the package file 00:40:05 ... if we're linking to external metadata, we can determine the language of the file 00:40:15 ... consistent with other vocabularies 00:40:21 ... this would be new, and wouldn't break anything 00:40:25 ... any comments? 00:40:40 BenSchroeter: Trying to understand the usage of this? 00:41:09 dauwhe: Say you have a multi-lingual epub, multiple metadata records 00:41:34 ... norwegian canadian dictionary that has one record in norwegian and one in canadian english 00:41:54 ... if the UI is set in one language, it can know to present the correct metadata to the user 00:42:11 ... the RS doesn't have to determine the vocab from the resource, it can determine before download 00:42:14 ... a hint 00:42:25 BenSchroeter: Ok 00:42:36 dauwhe: I'm not aware of a RS that does this 00:42:57 duga: You can link to external metadata records? /sarcasm 00:43:12 dauwhe: We can write a test for it, but no RS will test it 00:43:28 BenSchroeter: And where will you find a norwegian-canadian dictionary 00:43:49 duga: This PR does add another untestable conformance requirement 00:43:59 dauwhe: I feel we're a little bit stuck here 00:44:09 ... it's an i18n best practice 00:44:19 ... if we say we're going to have linked metadata, we should have this 00:44:32 ... we put this in, we get tests, the concept gets marked as at-risk at CR 00:44:48 ... I think we should accept the PR, then see what testing shows us 00:45:18 Proposed: Close 1488, accept PR 1497 00:45:25 Resolved: Close 1488, accept PR 1497 00:45:31 dauwhe: Progress! 00:45:41 ... on i18n issues at least 00:46:31 ... thanks everyone 00:46:45 goodnight 00:46:45 ... see you all in the comment section 00:46:58 RRSAgent: draft minutes 00:46:58 I have made the request to generate https://www.w3.org/2021/02/04-epub-minutes.html dauwhe 00:47:04 rrsagent, make logs public 00:47:35 RRSAgent: bye 00:47:35 I see no action items