12:59:06 <RRSAgent> RRSAgent has joined #epub-fxl
12:59:06 <RRSAgent> logging to https://www.w3.org/2021/04/27-epub-fxl-irc
12:59:08 <Zakim> RRSAgent, make logs Public
12:59:09 <Zakim> please title this meeting ("meeting: ..."), wendyreid
12:59:56 <wendyreid> meeting: Fixed Layout Accessibility TF Telco April 27, 2021
13:00:02 <wendyreid> date: 2021-04-27
13:00:06 <wendyreid> chair: wendyreid
13:00:15 <wendyreid> regrets+ laurab
13:00:20 <KenJones> KenJones has joined #epub-fxl
13:00:43 <gpellegrino> gpellegrino has joined #epub-fxl
13:00:56 <gpellegrino> present+
13:01:00 <MattChan> present+
13:01:25 <KenJones> present+
13:02:30 <duga> duga has joined #epub-fxl
13:02:37 <duga> present+
13:03:03 <MattChan> scribe+
13:03:22 <MattChan> wendyreid: this week the biggest thing was the outline for the best practices documentation
13:03:28 <MattChan> ... i plan on getting it into github
13:03:31 <MattChan> ... there'
13:03:41 <MattChan> ... there's a respec format for WG notes
13:04:02 <MattChan> ... probably better to just do it in github directly vs doing it in G Docs and then having to translate
13:04:37 <MattChan> ... then we can use the normal issues, PRs, etc.
13:05:00 <MattChan> ... putting the outline on screen share now
13:05:03 <rsosolen> rsosolen has joined #epub-fxl
13:05:13 <MattChan> ... we probably already have content for a lot of these areas
13:05:55 <MattChan> ... start with introduction (standard W3C) > purpose, challenges and limits of FXL, overview of document
13:06:52 <MattChan> ... high level topics include reading order, alt-text/extended desc, navigation (toc, section markers and landmarks), perceiving content (readability, contrast), MO, tables, a11y metadata
13:07:14 <MattChan> ... there's going to be a lot of cross-referencing between this doc and a11y 1.1/a11y techniques
13:07:28 <wendyreid> q?
13:07:42 <KenJones> q+
13:07:48 <wendyreid> ack KenJones
13:07:51 <MattChan> KenJones: i like it
13:08:08 <MattChan> ... the main one for me is tables
13:08:24 <MattChan> q+
13:08:35 <MattChan> ... would be nice to have examples of all of these things done well
13:08:37 <wendyreid> scribe+
13:08:40 <wendyreid> ak MattChan
13:08:43 <wendyreid> ack MattChan
13:09:02 <wendyreid> MattChan: Laura just confirmed with me that Anansi is going to be able to provide samples to use in this document
13:09:17 <wendyreid> ... I don't think we'll have tables, but we'll have children's content
13:09:45 <MattChan> wendyreid: i've sent email to business group to ask members if anyone would be willing to donate
13:09:57 <MattChan> ... no response yet, but we have a meeting today, we'll see
13:10:15 <gpellegrino> q+
13:10:18 <wendyreid> ack gpellegrino
13:10:26 <MattChan> ... we're only looking for samples, so a couple of pages, nothing more than what you would see for free online
13:10:39 <MattChan> gpellegrino: talked to cristina about some italian publishers we can ask
13:11:17 <MattChan> ... one publishes FXL for children, the other is a big publisher that uses FXL for print replica for books that are text driven but the authors and agents want to have same look as paper version
13:11:36 <MattChan> ... do you think this second category would be useful?
13:11:50 <MattChan> wendyreid: i think so, there are interesting things that are done with the styling of those books
13:12:02 <MattChan> gpellegrino: okay, i will check
13:12:45 <MattChan> wendyreid: i'll get the github setup this week
13:13:01 <MattChan> ... hopefully mgarrish can give us some guidance on updating the Daisy KB
13:13:19 <MattChan> ... whether he prefers we open issues, or submit PRs directly
13:13:43 <MattChan> ... other stream of work is the experiments
13:13:52 <MattChan> ... business group is thinking right now about FXL embedded in reflowable
13:14:08 <MattChan> ... they are having that conversation today, hoping to come up with some use cases
13:14:23 <MattChan> ... our other experiment was to try out flexbox to see what we can achieve
13:14:32 <MattChan> ... gpellegrino has volunteered
13:14:45 <MattChan> ... what can we commit to getting done in the next week or two?
13:15:14 <gpellegrino> q+
13:15:19 <wendyreid> ack gpellegrino
13:16:00 <MattChan> gpellegrino: is there anyone who can help us with the CSS side? using the new features?
13:17:03 <MattChan> wendyreid: yes, there are also lots of good resources out there about flexbox, etc.
13:18:37 <MattChan> ... for the outline, does anyone want to take on doing a first draft of any of those sections?
13:18:59 <MattChan> ... that we can then collaboratively sharpen up?
13:19:01 <KenJones> q+
13:19:04 <rsosolen> q+
13:19:13 <wendyreid> ack KenJones
13:19:33 <MattChan> KenJones: i'll be happy to take on some of those, don't mind which ones apart from tables
13:19:46 <MattChan> ... can you let me know which ones?
13:19:50 <wendyreid> ack rsosolen
13:20:09 <MattChan> rsosolen: i can help with a11y metadata, but pretty much anything else too
13:20:25 <MattChan> ... also willing to help with the experiments
13:20:59 <MattChan> ... i can also find some information about the tables best practices
13:21:35 <gpellegrino> q+
13:21:53 <wendyreid> ack gpellegrino
13:21:53 <MattChan> wendyreid: rsosolen can do tables and a11y metadata for FXLs
13:22:06 <MattChan> gpellegrino: question about experiments and javascript
13:22:21 <MattChan> ... a lot of FXL use javascript, but some RS will not support this
13:22:42 <MattChan> ... is it fine if we start without javascript, and add best practices for scripts in a second round?
13:22:54 <MattChan> ... may be too much to tackle everything all at once
13:23:11 <MattChan> wendyreid: yes, let's see what we can do without it for now
13:23:43 <MattChan> ... KenJones can tackle reading order, navigation best practices for now
13:23:53 <MattChan> ... in the best practices
13:24:08 <MattChan> KenJones: can you share an example for formatting?
13:24:28 <MattChan> wendyreid: look at the source HTML for the epub spec, or the audiobook spec
13:24:32 <MattChan> ... it's very basic
13:24:42 <MattChan> gpellegrino: there's also a guide for W3C respec somewhere
13:24:55 <wendyreid> https://respec.org/docs/
13:25:02 <MattChan> wendyreid: here it is
13:25:30 <MattChan> ... don't worry too much about the syntax, not to difficult to go back and fix after
13:25:51 <MattChan> ... seems like we have everything we can do for now assigned
13:26:09 <shiestyle> shiestyle has joined #epub-fxl
13:26:15 <MattChan> ... gpellegrino can you think of a couple ideas for experiments?
13:26:29 <MattChan> ... will make it easier to assign experiments to people in future
13:27:06 <MattChan> ... AOB?
13:27:22 <MattChan> KenJones: i've got questions that I started posting to the Daisy KB, e.g. for image descriptions
13:27:53 <MattChan> ... another one, where there's a very annotated scientific image, how do we describe it? e.g. medical text with diagram of the heart
13:28:09 <gpellegrino> q+
13:28:34 <wendyreid> ack gpellegrino
13:28:34 <MattChan> wendyreid: i believe Daisy has guidance in the rest of the KB, e.g. in the image maps section?
13:29:03 <MattChan> gpellegrino: you can check diagram standard for how to describe image, and on epubtest.org there is a test for how to insert long desc in reflow epub
13:29:21 <MattChan> ... those two resources together should give you an idea of best practice for your use-case
13:29:24 <KenJones> q+
13:29:26 <wendyreid> ack KenJones
13:29:26 <gpellegrino> http://diagramcenter.org/table-of-contents-2.html
13:29:51 <gpellegrino> http://epubtest.org/test-books
13:30:14 <MattChan> KenJones: might a very visual document be rendered in a more textual way? i.e. converted from FXL to a reflow text
13:30:25 <MattChan> ... are we allowed to recommend that as a secondary method of consuming the content?
13:31:07 <duga> q+
13:31:07 <MattChan> wendyreid: depends how you do it. e.g. A book with mixed modalities, with a complicated map or something, and then the accompanying section just describes the same thing in text
13:31:13 <wendyreid> ack duga
13:31:20 <MattChan> ... we've talked about alternative renditions, but that's not supported
13:31:52 <MattChan> duga: are you looking at a way of algorithmically converting FXL to reflow? (e.g. RS turns FXL cookbook into a reflow)
13:32:07 <MattChan> KenJones: yes, so, like a fallback for a small screen or a screenreader
13:32:14 <MattChan> ... a secondary way of showing the same file
13:32:45 <MattChan> duga: right, so changing the styling so that it works in reflow, maybe merging across page breaks in the FXL
13:33:19 <gpellegrino> media queries :)
13:33:20 <MattChan> wendyreid: so like, interchangable style sheets, one for FXL and one for reflow
13:33:37 <MattChan> KenJones: yes, and discarding the positioning style
13:34:03 <MattChan> duga: I think its an interesting idea that is better than multiple renditions, it makes the content more accessible to all
13:34:09 <MattChan> ... but this might not be in scope
13:34:27 <MattChan> wendyreid: gpellegrino mentioned media queries, but RS also support these in weird ways
13:34:53 <MattChan> ... you could do it with CSS, but what you can't do is toggling back and forth between FXL and reflow
13:35:06 <MattChan> ... e.g. Kobo uses different renders for FXL and reflow content
13:35:21 <MattChan> ... we'd have to tell RS when and how to switch
13:35:38 <MattChan> duga: you could just re-open the content in the alternate renderer
13:36:01 <MattChan> ... still more simple than mixed modalities
13:36:38 <MattChan> wendyreid: so what would be the trigger? how would an epub tell RS that feature is available in the book?
13:36:52 <MattChan> KenJones: maybe if both accessMode textual and accessMode visual?
13:37:15 <MattChan> duga: basically we're just marking a part of css as "for flowing mode"
13:37:51 <gpellegrino> q+
13:37:54 <MattChan> ... there are also layout considerations where we want to combine FXL pages into a single flowing page
13:38:05 <MattChan> ... not that simple
13:38:07 <wendyreid> ack gpellegrino
13:38:19 <MattChan> ... may want to recharter if we tackle this, because it would require spec changes
13:38:45 <MattChan> gpellegrino: when you produce FXL you have to give viewport info, so when you switch to reflow mode, what happens with the viewport?
13:39:23 <MattChan> wendyreid: yes, so the epub would need to tell RS that it can switch from FXL to reflow, and when it switches to flowing mode, what RS is allowed to ignore/change from the FXL
13:39:31 <MattChan> ... so yes, this would require spec change
13:39:41 <MattChan> ... but agree that this is worth exploring
13:40:03 <MattChan> ... maybe we can suggest this to CG to do the initial experimentation
13:40:13 <MattChan> ... especially as spec is heading towards CR by the end of this year
13:40:30 <KenJones> q+
13:40:31 <MattChan> ... i'll setup meeting with Mathias and Zheng
13:40:34 <wendyreid> ack KenJones
13:43:22 <MattChan> duga: we should also make sure that people from WG are going to the CG meetings where this is discussed
13:44:04 <MattChan> wendyreid: okay, so we all have our list of action items
13:44:12 <MattChan> ... and we'll report in on progress next week
13:47:28 <wendyreid> zakim, end meeting
13:47:28 <Zakim> As of this point the attendees have been gpellegrino, MattChan, KenJones, duga
13:47:31 <Zakim> RRSAgent, please draft minutes
13:47:31 <RRSAgent> I have made the request to generate https://www.w3.org/2021/04/27-epub-fxl-minutes.html Zakim
13:47:34 <Zakim> I am happy to have been of service, wendyreid; please remember to excuse RRSAgent.  Goodbye
13:47:38 <Zakim> Zakim has left #epub-fxl
13:47:44 <wendyreid> rrsagent, make logs public