13:49:22 <RRSAgent> RRSAgent has joined #epub-a11y
13:49:22 <RRSAgent> logging to https://www.w3.org/2021/03/25-epub-a11y-irc
13:49:31 <Zakim> Zakim has joined #epub-a11y
13:49:39 <avneeshsingh> zakim, this will be epub-a11y
13:49:39 <Zakim> ok, avneeshsingh
13:49:51 <avneeshsingh> present+
13:50:07 <avneeshsingh> chair: avneeshsingh
13:58:13 <mgarrish> mgarrish has joined #epub-a11y
13:59:45 <BenSchroeter> BenSchroeter has joined #epub-a11y
14:00:14 <MattChan> MattChan has joined #epub-a11y
14:01:07 <wendyreid> wendyreid has joined #epub-a11y
14:01:09 <gpellegrino> gpellegrino has joined #epub-a11y
14:01:19 <MattChan> present+
14:01:25 <wendyreid> present+
14:01:31 <gpellegrino> present+
14:02:24 <BenSchroeter> present+
14:03:53 <avneeshsingh> https://w3c.github.io/epub-specs/epub33/a11y/index.html
14:03:55 <MattChan> scribe+
14:04:17 <avneeshsingh> https://github.com/w3c/epub-specs/issues/1500
14:04:18 <MattChan> avneeshsingh: let's move to the first issue originally created by gpellegrino and elevated by mgarrish
14:04:27 <MattChan> Topic: Page list ordering
14:04:46 <MattChan> mgarrish: right now the pagelist is required to follow spine+content order
14:05:06 <MattChan> ... pagelist will therefore match digital sequencing
14:05:17 <avneeshsingh> q?
14:05:20 <MattChan> ... but should we lift the requirement, therefore allowing the pagelist to follow print sequencing
14:05:26 <Bill_Kasdorf> Bill_Kasdorf has joined #epub-a11y
14:05:30 <MattChan> ... right now there is no choice in it
14:05:44 <MattChan> ... we did require this for toc, and we reduced that requirement
14:05:50 <MattChan> ... i don't have a strong position on this
14:05:52 <avneeshsingh> q?
14:06:00 <wendyreid> q+
14:06:03 <MattChan> ... it really comes down to which one users expect, and what makes sense
14:06:07 <MattChan> avneeshsingh: thoughts?
14:06:15 <avneeshsingh> ack next
14:06:33 <gpellegrino> q+
14:06:39 <MattChan> wendyreid: i have no seen RS that uses pagelist to dictate page number order in epub
14:06:44 <MattChan> ... kobo doesn't do it
14:07:06 <MattChan> ... pagelist is more likely treated as just an additional nav aid, like toc
14:07:09 <avneeshsingh> q?
14:07:13 <CharlesL> CharlesL has joined #epub-a11y
14:07:21 <avneeshsingh> ack gp
14:07:30 <CharlesL> present+
14:07:31 <Bill_Kasdorf> q+
14:07:40 <MattChan> gpellegrino: as far as i've tested i think that apple books uses it for UI toc, where they put "real" page numbers for different toc entries
14:07:54 <MattChan> ... from my point of view, which are the use cases for end users for the pagelist?
14:07:59 <MattChan> ... what are the requirements for RS?
14:08:00 <avneeshsingh> q?
14:08:02 <mgarrish> q+
14:08:09 <MattChan> ... what are the requirements for content producers?
14:08:32 <MattChan> avneeshsingh: one use case i would like to mention is if there is a classroom, and there are students using print, and a blind student using ebook
14:08:41 <avneeshsingh> ack bill
14:08:51 <gpellegrino> q+
14:08:53 <MattChan> ... if teacher says "move to page X", then it make sense for pagelist to match print
14:09:16 <MattChan> Bill_Kasdorf: what is the relationship of the pagebreak markers to the pagelist?
14:09:22 <MattChan> ... i thought they always had to correspond?
14:09:25 <avneeshsingh> ack mg
14:10:08 <MattChan> mgarrish: They have to correspond. So, they need to be included in the pagelist, but why should they match the order of the epub content?
14:10:33 <Bill_Kasdorf> +1 to Matt
14:10:57 <MattChan> ... e.g. if an epub has reordered pages, and you're looking for a particular print correspondence page, then how do you find it (without using a UI where you can type in a page number)
14:10:59 <avneeshsingh> ack gp
14:11:26 <MattChan> gpellegrino: another use case could be for quotations, and for finding quotations in books
14:11:33 <MattChan> ... usually quotations are based on print page numbers
14:11:44 <Will> Will has joined #epub-a11y
14:11:50 <MattChan> ... also, from UI point of view, few RS are using something like an input tool to navigate to page
14:11:58 <wendyreid> q+
14:12:11 <MattChan> ... because there can be a mix of page number styles (e.g. a, i, I, arabic numbers)
14:12:13 <avneeshsingh> ack next
14:12:18 <MattChan> ... hard for a single tool to find any given page
14:12:34 <MattChan> wendyreid: any time i've seen implementation of pagelist it has been in toc style (e.g. a list of links)
14:12:58 <MattChan> ... in this case it makes more sense to follow the correspondence order (e.g. print)
14:13:00 <avneeshsingh> q?
14:13:12 <MattChan> ... and always be in linear sequence, of course
14:13:39 <MattChan> avneeshsingh: so it seems like we are all in favour of relaxing the requirement of pagelist following content order
14:13:41 <Bill_Kasdorf> q+
14:14:18 <avneeshsingh> ack bill
14:14:33 <MattChan> mgarrish: i think we can have some sort of a recommendation, at least, that pagelist follow the print
14:14:38 <avneeshsingh> q?
14:14:57 <wendyreid> q+
14:14:58 <MattChan> Bill_Kasdorf: i've always thought of pagelist as a correspondence tool to an authoritative print equivalent
14:15:08 <avneeshsingh> ack next
14:15:13 <MattChan> ... but for publications without print equivalent, what is the purpose of pagelist?
14:15:40 <Bill_Kasdorf> q+
14:15:50 <gpellegrino> CFI?
14:15:51 <MattChan> wendyreid: this is virtual page numbers. We're exploring this topic. The need for these has become more prevalent recently
14:16:13 <avneeshsingh> ack bill
14:16:17 <MattChan> Bill_Kasdorf: this is a key distinction. We should be careful with the distinction
14:16:37 <MattChan> ... only use page number for print equivalent. The other one should be called "locators"
14:16:44 <avneeshsingh> q?
14:16:46 <MattChan> avneeshsingh: we are using the same terms in wcag 2.2
14:17:00 <avneeshsingh> proposed: relax strict requirement of sequential page numbers in EPUB 3 specification
14:17:08 <MattChan> +1
14:17:09 <mgarrish> +1
14:17:12 <wendyreid> +1
14:17:15 <Bill_Kasdorf> +1
14:17:15 <CharlesL> 0
14:17:21 <Will> +1
14:17:37 <gpellegrino> +1
14:17:39 <BenSchroeter> +1
14:17:49 <MattChan> avneeshsingh: this may require main group approval later
14:18:37 <MattChan> wendyreid: we're going to have to discuss this in the main group call, maybe on the agenda next week
14:18:38 <avneeshsingh> resolved
14:18:58 <gpellegrino> q+
14:19:35 <avneeshsingh> ack gp
14:19:41 <MattChan> avneeshsingh: mgarrish will make changes and do PR
14:19:42 <avneeshsingh> https://github.com/w3c/epub-specs/issues/1470
14:19:50 <MattChan> Topic: Rechecking accessibility
14:19:56 <George> George has joined #epub-a11y
14:20:35 <gpellegrino> q+
14:20:40 <avneeshsingh> q?
14:20:43 <George> present+
14:20:46 <MattChan> mgarrish: this came up in ISO. How long does the certification of content being accessible last for? e.g. For website the certification might degrade over 5 years, after which the accessibility certification might not be trustworthy anymore
14:20:53 <CharlesL> q+
14:20:55 <MattChan> ... less of a concern for packaged up epubs that shouldn't change
14:21:02 <MattChan> ... but what about reissues of epubs?
14:21:11 <Bill_Kasdorf> q+
14:21:25 <MattChan> ... should we have guidance for that? A specific recommendation? Leave it to authors?
14:21:27 <avneeshsingh> ack gp
14:21:35 <MattChan> avneeshsingh: it will depend on publisher processes
14:22:03 <MattChan> gpellegrino: we check all the books published by our members
14:22:12 <MattChan> ... we hash each epub
14:22:27 <MattChan> ... so when we get a new file that matches an existing isbn, we match the hashes
14:22:27 <avneeshsingh> ack charles
14:22:35 <MattChan> ... if hash mismatch, we recheck a11y
14:22:43 <Bill_Kasdorf> q-
14:23:02 <MattChan> CharlesL: for GCA, once we certify a publisher, they can stamp all books coming off that workflow as certified
14:23:12 <MattChan> ... but the publisher must get recertified each year
14:23:29 <MattChan> ... and that recertification process will bring them up to the state of the art
14:23:46 <MattChan> ... if they then reissue epub after that point, then they can update the certification as well
14:23:54 <BenSchroeter> q+
14:23:55 <wendyreid> q+
14:23:57 <MattChan> ... i.e. our solution is to keep the publisher up to date
14:24:33 <MattChan> avneeshsingh: if the user gets an old ebook, say 3 years old, how can the user ensure that the latest revision is accessible?
14:24:36 <avneeshsingh> q?
14:25:11 <MattChan> CharlesL: the user would just refer to the a11y metadata in the reprint. And it would be up to the publisher to use the latest standards in the publishing pipeline
14:25:35 <MattChan> ... between recertifications, we leave this up to the publisher
14:25:39 <avneeshsingh> ack ben
14:25:54 <MattChan> BenSchroeter: if its a new edition, i would consider that a new publication
14:26:19 <MattChan> ... that would go through whatever processes the publisher uses for normal a11y check
14:26:39 <George_> George_ has joined #epub-a11y
14:26:52 <MattChan> ... in terms of GCA certified publishers
14:26:59 <George_> present+
14:27:03 <mgarrish> q+
14:27:40 <MattChan> ... i can see a case where a title goes through the certified workflow, but on republication changes are made (e.g. inserted image) that doesn't conform with a11y standards
14:27:44 <MattChan> ... this is a potential gap
14:27:55 <avneeshsingh> ack next
14:28:05 <MattChan> ... and also, we may be moving towards a world in which there are more and more frequent updates to epubs
14:28:32 <MattChan> wendyreid: any guidelines we put in place will have to apply across the board - from GCA certified to micro indie publisher
14:28:55 <MattChan> ... for a lot of repubs, the content of the book is not really changing (maybe just update to copyright page or something)
14:29:18 <MattChan> ... so we have to be careful of mandating recheck on "new hash" or every new revision
14:29:27 <avneeshsingh> q?
14:29:36 <MattChan> ... maybe qualify by saying "if adding new content" or "if substantially changing content"
14:29:51 <CharlesL> +1 Wendy
14:29:55 <MattChan> ... but have exclusion for if you are just fixing typos or updating marketing material
14:30:03 <avneeshsingh> ack mg
14:30:22 <MattChan> ... also, re. comparison between epub and website, epubs don't replicate at the same rate
14:30:33 <MattChan> mgarrish: i think it has to be an informative section
14:30:52 <MattChan> ... what wendyreid has described is similar to the qualifications that we once had on release identifier
14:30:59 <MattChan> ... this is probably as far as we can go
14:31:08 <MattChan> ... don't want to get mired in questions of what it means to recheck
14:31:25 <gpellegrino> q+
14:31:33 <CharlesL> q+
14:31:36 <avneeshsingh> q?
14:31:45 <MattChan> ... maybe just say that standards change, and that when you do something to your epub you should just make sure that you are on top of the latest standards
14:31:46 <avneeshsingh> ack gp
14:31:54 <MattChan> gpellegrino: one of our ideas was to add date of certification
14:32:17 <MattChan> ... e.g. in accessibility summary
14:32:34 <MattChan> CharlesL: the idea of a certification date is something we could definitely add
14:32:55 <MattChan> ... that way publishers could change the certification date, or not, when they make other changes to the epub
14:33:04 <avneeshsingh> ack charl
14:33:20 <MattChan> ... i'm wondering about if the publisher modifies something 3 years from now, and URLs for conformsTo have changed
14:33:43 <avneeshsingh> q?
14:33:43 <MattChan> ... do they change the URLs to new URLs, or leave them as what they were at the point of initial certification?
14:34:05 <MattChan> mgarrish: hard question to answer, because we're not negating the old standards
14:34:22 <MattChan> ... maybe the date based system is a good way to go
14:34:39 <avneeshsingh> q?
14:34:39 <MattChan> ... hard to make normative statements because of all the possibilities
14:35:03 <MattChan> ... maybe we could rely on legislation to set the rules, and just offer some guidance
14:35:29 <George_> q+
14:35:30 <George_> q+
14:35:34 <MattChan> avneeshsingh: we can start by providing non-normative text like mgarrish suggested
14:35:46 <MattChan> ... let this mature for some number of years, and then revisit it
14:35:58 <avneeshsingh> ack george
14:36:33 <MattChan> George_: when a publisher is doing update or new edition, if they have switched to a11y 1.1, wouldn't they just start using the correct metadata URLs at that point?
14:36:35 <avneeshsingh> q?
14:36:49 <MattChan> ... publishers are not all going to go from a11y 1.0 to 1.1 all at once
14:36:57 <avneeshsingh> proposed: add non-normative text in ePUB Accessibility suggesting rechecking the publications for major changes that could effect accessibility
14:37:05 <wendyreid> +1
14:37:06 <CharlesL> +1
14:37:09 <gpellegrino> +1
14:37:10 <MattChan> +1
14:37:11 <Bill_Kasdorf> +1
14:37:13 <Will> +
14:37:14 <mgarrish> +1
14:37:16 <BenSchroeter> +1
14:37:17 <Will> +1
14:37:21 <avneeshsingh> resolved
14:37:26 <MattChan> avneeshsingh: we'll work on the specific guidance language over on the issue tracker
14:37:27 <George_> +1
14:37:41 <avneeshsingh> https://github.com/w3c/epub-specs/issues/1302
14:37:47 <MattChan> Topic: Audio hazards
14:38:37 <MattChan> avneeshsingh: in the hazards there are parameters for visual hazards in WCAG (e.g. flashing no more than 3 per second)
14:38:44 <MattChan> ... but no equivalent for audio hazards
14:38:52 <mgarrish> q+
14:38:52 <avneeshsingh> q?
14:38:59 <MattChan> ... how are publishers to classify which audio will causes issues for users?
14:39:00 <George_> q+
14:39:06 <avneeshsingh> ack mg
14:39:30 <MattChan> mgarrish: i looked into this years back with Madeleine Rothberg
14:39:50 <MattChan> ... in the github tracker, one potential standard was indicated
14:40:11 <MattChan> ... but it is harder to classify these sorts of audio hazards than with visual hazards
14:40:28 <Bill_Kasdorf> q+
14:40:41 <MattChan> ... maybe we just play it safe and recommend that if there is any sound at all then there might be a hazard
14:40:49 <CharlesL> q+
14:40:54 <avneeshsingh> ack george
14:41:04 <MattChan> ... if the absence of a clear standard, what do we say?
14:41:22 <MattChan> George_: sudden increases in volume were the problem, could potentially damage hearing
14:41:35 <BenSchroeter> q+
14:41:49 <MattChan> ... that's the sort of thing that should be flagged as an audio hazard
14:41:55 <avneeshsingh> ack bill
14:42:01 <MattChan> ... that sudden increase in sound that could damage hearing
14:42:23 <avneeshsingh> ack chrles
14:42:24 <MattChan> Bill_Kasdorf: "any audio" seems broad
14:42:34 <MattChan> ... it would catch even the books with only narration
14:42:48 <wendyreid> q+
14:42:54 <MattChan> CharlesL: increase in decibel level is probably a good way to put it
14:43:04 <MattChan> ... and then also something about tones (like in those hearing tests)
14:43:06 <mgarrish> q
14:43:10 <MattChan> ... probably some text around those is all we need
14:43:12 <avneeshsingh> ack ben
14:43:27 <MattChan> BenSchroeter: I think its kind of meaningless unless its quantifiable
14:43:47 <avneeshsingh> q?
14:43:59 <mgarrish> +1 to ben
14:44:01 <MattChan> ... anything else (e.g. warning there could be explosions) is not likely to be enforceable
14:44:13 <avneeshsingh> ack charles
14:44:28 <MattChan> CharlesL: i think some of that could be machine checked (e.g. with AI)
14:44:41 <mgarrish> q+
14:44:52 <Bill_Kasdorf> q+
14:45:06 <gpellegrino> q+
14:45:12 <gpellegrino> q-
14:45:14 <avneeshsingh> ack next
14:45:17 <MattChan> avneeshsingh: we can rely on research done on wcag on this
14:45:25 <MattChan> s/done on/done by
14:45:38 <avneeshsingh> q?
14:45:46 <gpellegrino> +1 to Avneesh, I would forward it to WCAG
14:46:02 <MattChan> wendyreid: we should probably ask for quantifiable information that we can provide to content producers
14:46:32 <MattChan> ... there is a body of research on the topic of hazardous sounds
14:46:59 <MattChan> ... and with the increasing popularity of audiobooks, i've seen increases in the creative use of sound
14:47:14 <avneeshsingh> ack mg
14:47:26 <MattChan> avneeshsingh: we need proven research to drive the parameters
14:47:40 <MattChan> mgarrish: wcag may not have an easy answer on this for us
14:48:12 <MattChan> ... and what does it mean to users? If we say that it has a sound hazard, there are a variety of ways that users could be affected (i.e. seizures? damage hearing?)
14:48:16 <MattChan> ... what does it mean to users?
14:48:18 <avneeshsingh> q?
14:48:41 <avneeshsingh> ack bill
14:48:45 <MattChan> ... more work to be done in this area
14:49:19 <MattChan> Bill_Kasdorf: I bet there's an authoritative place to define issues like this (e.g. audiology)
14:49:19 <avneeshsingh> q?
14:49:30 <MattChan> ... we should rely on those sources rather than try to strike out on our own
14:49:40 <George_> q+
14:49:40 <MattChan> avneeshsingh: for now we cannot provide the necessary parameters
14:49:48 <MattChan> ... if wcag comes back with answer, then great
14:50:11 <MattChan> ... but with the knowledge we have now, we should maintain the status quo
14:50:32 <avneeshsingh> q?
14:50:35 <MattChan> ... my preference would be if the WAI work on this on the level of the broader W3C, instead of making this an epub a11y thing
14:51:04 <MattChan> George_: i think that is the right approach, but i would expect that audiobook publishers and publishers in general would look to us first and not go digging around in wcag
14:51:27 <MattChan> ... if we put informative information, even though we can't point to the research, i think that would benefit people
14:51:38 <MattChan> avneeshsingh: is something like that already in the spec?
14:51:45 <MattChan> ... examples, patterns etc?
14:52:00 <avneeshsingh> q?
14:52:09 <avneeshsingh> ack george
14:52:16 <MattChan> mgarrish: there are very general descriptions of hazardous audio, but we don't have references to anything specific where people can go to get more info
14:52:48 <MattChan> ... e.g. there was reference in the issue to a standard to define what "loud" meant
14:52:48 <avneeshsingh> q?
14:53:08 <MattChan> avneeshsingh: George_ you can go through what we have currently in the spec
14:53:31 <MattChan> ... the agenda also lists a number of issues that mgarrish has closed since last meeting
14:53:32 <avneeshsingh> q?
14:53:49 <MattChan> ... we can comment on those right now, or you can put your comments in the issue tracker
14:54:10 <avneeshsingh> q?
14:54:17 <CharlesL> https://audiology-web.s3.amazonaws.com/migrated/niohlprevention.pdf_53996fb4c1ca13.61907521.pdf
14:54:23 <MattChan> Bill_Kasdorf provides example of CDC standard of sound that may impact hearing
14:54:41 <MattChan> avneeshsingh: next call about EUAA
14:54:43 <Bill_Kasdorf> https://www.cdc.gov/nceh/hearing_loss/what_noises_cause_hearing_loss.html#:~:text=Sound%20is%20measured%20in%20decibels,immediate%20harm%20to%20your%20ears.
14:55:11 <MattChan> avneeshsingh: i've send an updated calendar invite around accounting for DST
14:55:40 <avneeshsingh> rrsagent, make logs public
14:55:50 <avneeshsingh> rrsagent, draft minutes
14:55:50 <RRSAgent> I have made the request to generate https://www.w3.org/2021/03/25-epub-a11y-minutes.html avneeshsingh
15:04:09 <CharlesL> CharlesL has left #epub-a11y
16:56:35 <Zakim> Zakim has left #epub-a11y
18:45:00 <gpellegrino> gpellegrino has joined #epub-a11y
18:45:39 <gpellegr_> gpellegr_ has joined #epub-a11y