13:43:11 RRSAgent has joined #pmwg 13:43:15 logging to https://www.w3.org/2024/02/02-pmwg-irc 13:43:15 RRSAgent, make logs Public 13:43:16 Meeting: Publishing Maintenance Working Group 13:43:35 chair: wendy, shinya 13:51:36 shiestyle has joined #pmwg 13:54:46 duga has joined #pmwg 13:58:50 tzviya has joined #pmwg 14:00:19 present+ 14:00:25 Meeting: PMWG 14:00:26 present+ 14:00:26 present+ 14:00:29 present+ 14:00:40 Date: 2024-02-02 14:00:44 Chair: shiestyle 14:02:18 zakim, start the meeting 14:02:18 RRSAgent, make logs Public 14:02:20 Meeting: Publishing Maintenance Working Group 14:03:18 https://github.com/w3c/epub-specs/pull/2602 14:03:36 present+ 14:03:52 Hadrien has joined #pmwg 14:03:58 present+ Hadrien 14:03:59 present+ 14:04:04 present+ Wendy 14:04:41 gpellegrino has joined #pmwg 14:04:48 present+ 14:04:49 present+ gpellegrino 14:04:50 scribe+ 14:05:04 Shinya: We have a PR 14:05:24 ... can someone explain the PR? 14:06:37 wendy: we both change the EPUB core and EPUB RS documents, in both documents we creating a new heading for telling that renditions flow metadata apply to both FXL and reflowable EPUB 14:07:19 ... we also created diagrams to explain the behaviour we expect when rendition flow is applied 14:07:27 q? 14:07:45 ... we also change a MUST to a MAIN when we say about RSs that may ignore renditions flow 14:07:58 q+ 14:08:18 q+ 14:09:09 ivan: is it correct that this change reflexts what is happening with webtoons? 14:09:22 q+ 14:09:27 ... are we documenting existing practices? 14:09:29 ack ivan 14:10:18 q+ 14:10:31 ack gpellegrino 14:10:43 q+ 14:11:31 shiestyle: It depends: we're using rendition-flow = continuos, Apple is using it in their apps 14:12:03 gpellegrino: did we discuss it at last two TPACs? Didn't we agree to move to the CG for experimenting? 14:12:06 ack dhall 14:12:25 ack Hadrien 14:12:27 ivan: We the last WG there was not enough time, I don't remember with CD 14:12:30 CG* 14:12:54 dhall: Yes, we are using it in FXLs EPUBs 14:13:17 Hadrien: I think it really depends on what we agree it is a webtoon 14:13:28 s/with CD/with CG/ 14:13:42 ... the division is not structurale, the vertical image is split just for distribution 14:14:05 ... this helps also for optimization in loading 14:14:33 ... this content is consumed in specialized platforms outside the generic EPUB ecosystem 14:14:37 q+ 14:15:07 q+ to add to Hadrien 14:15:08 ... I think that instead of discussing about webtoons, we're misusing an existing property 14:15:24 q+ 14:15:25 ack duga 14:15:33 ... that may hurt other use cases, I think there is place for a broader discussion 14:16:19 duga: about optimization: you can scroll the images natively, instead of using images in HTML 14:17:11 shiestyle: In Japan many publishers are using EPUB as an interchange format, more then for a B2C format 14:17:55 ivan: I'm trying to remember how it started: Japan publishers where worried of having a new big market (webtoons) outside the EPUB ecosystem 14:18:10 ack iv 14:18:15 ack wendy 14:18:42 q+ 14:19:18 q+ 14:19:34 wendy: I discussed with Hadrien offline, I think his comments are valid, at the same time I think we have to respond to this emerging market 14:20:53 ... for the comment about the "logistic" (series, episodes, etc.) I think this is a metadata issue 14:21:08 q+ 14:21:21 ack tzviya 14:21:21 tzviya, you wanted to add to Hadrien 14:21:24 tzviya: I'm feeling like I felt when FXL came out 14:22:05 ... we had an implementer who implemented it in a way and we adapted the format in that way 14:22:12 ack dhall 14:22:42 dhall: content specification for a true webtoon may be more performant 14:23:36 ... I would split this PR from "webtoons", this is about how FXLs should be displayed with rendition-flow: continuos 14:23:58 q- 14:24:54 duga: we raised this issue two years ago and suggested to move it to the CG 14:25:21 ... we should have done it two years ago 14:25:48 ack du 14:25:50 ... I think also that webtoons is a trademark and we should not use it 14:25:51 ack had 14:26:35 Hadrien: to respond to David, no one asked for continuos scrolling outside of webtoons 14:26:57 q+ 14:27:05 ... continuos scrolling is difficult to get with EPUBs, and even more with EPUB FXLs 14:27:09 q+ 14:28:01 ... what I see that happen around the world is to tweak the format for doing things that are outside of the standard 14:28:03 q- 14:28:36 ... I don't think that hacking something is a good approch 14:29:29 ... I think is better just to say "this is a webtoon" 14:29:50 ... and manage that publication in a specific way 14:30:09 ... instead of tweaking a metadata that was intended for other stuff 14:30:37 shiestyle: In Japan we what to find a simple way to make comics in EPUBs 14:30:56 q? 14:31:01 ack shiestyle 14:31:06 ... we want to separate this PR with minor changes to the spec 14:31:06 q+ 14:31:50 wendy: I think it is important to speak with Amazon (Kindle) on this 14:32:13 ... to understand if this solution is good enough 14:32:23 q+ 14:32:30 q+ 14:32:59 ack ivan 14:33:05 ivan: Hadrien propose to add a new metadata to say that the publication is a "webtoon" 14:34:04 shiestyle: actually we already have special metadata for webtoons 14:34:09 ack had 14:34:23 Hadrien: in Japan there are a lot of profiles 14:34:37 ... with XML prefixes to add information 14:35:06 ... personally I don't think it is the way to move to metadata 14:35:44 ... we have metadata conformsTo metadata (used for accessibility), we may use it for telling the publication conforms to webtoon 14:35:58 ... it doesn't break anything 14:36:10 q+ 14:36:13 GautierC has joined #pmwg 14:36:34 ack gpe 14:37:44 ack duga 14:38:38 gpellegrino: for adding metadata to EPUB is not required to change the specs, you can use the meta tag 14:39:06 q+ 14:39:18 ... other comment about process, I think is better to move trough the CG, as we're doing for accessible FXL 14:39:53 q+ 14:40:10 ack tzviya 14:40:20 duga: I don't like the conformsTo approach, in any cases we need a spec to say what to do, in both cases we'll end up with RSs that will display broken webtoons 14:41:12 tzviya: I agree with Brady, if we have to create a new profile, we don't have the charter to do this, we can only maintain the spec that is already published 14:41:23 ack Hadrien 14:41:23 ack had 14:41:36 ... I think we should have everything in the main spec 14:42:38 Hadrien: to respond to Brady, I think that implementing vertical scroll is very expensive, specialized RSs are extracting images from the HTML and display them in canvas or similar 14:42:57 ... no matter what we do, it will be broken in most of reading systems 14:43:23 q+ 14:43:25 ... or we may suggest them a simpler way to ingest the EPUB and display only images 14:43:32 q+ 14:44:53 ack shie 14:44:54 ack shiestyle 14:45:00 ack ivan 14:45:12 shiestyle: I understand the performance issue, I think we should separate this discussion from the PR and wait for the Amazon comment 14:45:42 ivan: Hadrien, do you think the additional metadata is in addition to the PR? 14:45:56 Hadrien: no, it is instead of the PR 14:46:16 q+ 14:46:53 q+ 14:47:51 +1 to wendy 14:47:56 q+ 14:48:06 wendy: I understand Hadrien's point, a new metadata may lead to exotic values 14:48:23 ... where publisher expect some type of rendition 14:48:24 q+ to remind about dictionaries 14:48:27 ack ivan 14:48:55 q+ 14:49:02 q- 14:49:21 ivan: I think for the moment we'll not merge the PR 14:49:37 ... in the meantime I think we may work on test cases 14:50:28 ... and implementations, to be sure that it is feasible 14:50:34 ack dhall 14:51:06 dhall: does the PR specify a MUST or MAY conformance? 14:51:47 ack had 14:51:51 wendy: a MAY, before that rendition-flows was not available for reflowable 14:52:09 ... not you can use that value for fixed layout 14:52:10 q+ 14:53:01 q- 14:53:03 dhall: so a RS that does not support webtoons will display the publication in separated pages 14:53:36 Hadrien: In Japan there are already different profiles that create stuff on top of EPUBs 14:53:46 ... I think we should look at them 14:53:49 q+ 14:54:32 ... at the same time I think webtoons are quite different from the other type of publications, and requires some extra research 14:55:09 ack tzviya 14:55:09 tzviya, you wanted to remind about dictionaries 14:55:34 ack dhall 14:56:14 dhall: I think we are not in line with the semantic of the content... webtoons are different from web comics 14:56:33 ... so having a demo may be useful 14:58:26 wendy: I'll make a video 14:58:53 shiestyle: I'll contact Amazon for partecipating to the next meeting 14:58:59 rrsagent, draft minutes 14:59:01 I have made the request to generate https://www.w3.org/2024/02/02-pmwg-minutes.html ivan 15:03:04 duga has left #pmwg 15:03:26 gautierchomel has joined #pmwg 17:39:27 Zakim has left #pmwg 21:41:31 GautierC has joined #pmwg