14:51:40 <RRSAgent> RRSAgent has joined #epub-locators
14:51:40 <RRSAgent> logging to https://www.w3.org/2021/06/09-epub-locators-irc
14:51:43 <Zakim> RRSAgent, make logs Public
14:51:44 <Zakim> please title this meeting ("meeting: ..."), dauwhe
14:52:02 <dauwhe> meeting: EPUB 3 Locators Task Force
14:52:04 <dauwhe> chair: dauwhe
14:52:28 <tzviya> regrets+
14:53:29 <dauwhe> dauwhe has changed the topic to: https://docs.google.com/document/d/1y5JKcwlq1rvGYJ1HoLg0GF-NRiBsBNNZopGuQ0mgHUw/edit
14:59:50 <avneeshsingh> avneeshsingh has joined #epub-locators
15:01:23 <pilarw2000> pilarw2000 has joined #epub-locators
15:01:23 <dauwhe> present+
15:01:27 <avneeshsingh> present+
15:01:31 <pilarw2000> present+
15:01:53 <dauwhe> present+ dlazin
15:02:50 <dauwhe> present+ Laurent
15:03:29 <dlazin> dlazin has joined #epub-locators
15:03:33 <dlazin> present+
15:03:34 <dauwhe> scribe+ dauwhe
15:03:59 <laurent__> laurent__ has joined #epub-locators
15:04:09 <laurent__> present+
15:04:53 <dauwhe> dlazin: we're working on what might be reasonable to solve
15:05:21 <dauwhe> ... we appear to be heading towards a defined fall-back for page numbers
15:05:35 <dauwhe> ... a way to ensure pagelist is always present, even if not defined by the author
15:05:45 <dauwhe> ... we do want to encourage a pagelist in all EPUBs
15:05:56 <dauwhe> ... but have something to fall back to that is easy to implement
15:06:22 <dauwhe> ... Laurent and Hadrien had good comments about page numbers via email
15:06:39 <dauwhe> ... we don't think CFI is a bad idea, but we think it's a solution for a different set of problems
15:06:48 <dauwhe> ... these problems are around traditional page numbers
15:07:05 <dauwhe> avneeshsingh: CFI is also a problem for future for HTML EPUB
15:07:13 <dauwhe> ... these are quite broad use cases
15:07:25 <dauwhe> ... did you decide to put some of them out of scope
15:07:41 <dauwhe> dlazin: there's a spreadsheet for the use cases
15:07:46 <dauwhe> ... the gdoc is out of date
15:07:48 <dlazin> https://docs.google.com/spreadsheets/d/1KO-HyLGUUw36F-ruAARHNiPO1aUJCCNeTv3zxGtjuHw/edit#gid=0
15:07:51 <laurent__> q+
15:08:08 <dauwhe> ... in column I is whether things are in scope
15:08:13 <dauwhe> ... those are the page number things
15:08:23 <dauwhe> pilarw2000: dan, were you there last week?
15:08:26 <dauwhe> ... I missed it
15:08:43 <dauwhe> ... I did not see in the minutes: Mary and I had been in an indexing conf
15:08:57 <dauwhe> ... we asked about terms about what to call page numbers in ebooks that don't have page numbers
15:09:03 <dauwhe> ... one suggestion was "pointer"
15:09:16 <dauwhe> ... the thing a reader would click on to get to content
15:09:34 <dauwhe> ... the thing in the index that might look like a section/page/para number that you click
15:09:45 <dauwhe> ... "pointer" is a good word for that
15:10:02 <dauwhe> ... we'd talked about translating a CFI into something readable; that would be a pointer
15:10:08 <dauwhe> dlazin: don't forget that word :)
15:10:24 <dauwhe> ... anything that we name has to be worldwide
15:10:35 <dauwhe> ... users are familiar with pages
15:10:41 <dauwhe> ... the concept is functionally a page
15:11:00 <dauwhe> ... if we were to give it some other name, that might lead to irrelevancy
15:11:15 <dauwhe> ... the email said, I'm in favor of calling these things pages
15:11:31 <dauwhe> ... and encouraging reading systems to call their auto-numbering "screens"
15:11:46 <dauwhe> ... pages don't change as you change font size, but the screens do move
15:11:54 <dauwhe> pilarw2000: would we call them screen numbers?
15:12:03 <dauwhe> dlazin: we would do page numbers, RSs would do screen
15:12:05 <dauwhe> ack lau
15:12:24 <dauwhe> laurent__: First, we cannot ignore the Kindle
15:12:29 <dauwhe> ... which has locations
15:12:42 <dauwhe> ... but it's not simple
15:12:50 <dauwhe> ... in Readium we call it position
15:12:55 <dauwhe> ... we didn't want to call it page
15:13:05 <dauwhe> ... I would never speak about screens
15:13:14 <pilarw2000> q+
15:13:16 <dauwhe> ... screens is what you see, it's what's in the viewport
15:13:22 <dauwhe> ... no two experiences are the same
15:13:33 <dauwhe> ... to find something transportable, screens is not good
15:13:43 <dauwhe> ... we could keep the name page but there is confusion with pagelist
15:14:15 <dauwhe> ... we have to choose a name that is adequate
15:14:27 <dauwhe> dlazin: we don't bother with the queues here :)
15:14:45 <dauwhe> ... to clarify, what we're trying to introduce is not dependant on viewport or screen size
15:14:56 <dauwhe> ... the problem is that reading systems already have a variable thing, and they won't drop those
15:14:57 <ivan> present+
15:15:10 <dauwhe> ... we do want to call the fixed thing pages
15:15:20 <dauwhe> ... for one thing, if there's a pagelist we'll use that
15:15:25 <dauwhe> ... and we want the same functionality
15:15:38 <dauwhe> ... and we're recogniing there's an existing variable thing
15:15:48 <dauwhe> ... I do believe screen is a bad name
15:15:54 <dauwhe> ... becaause a11y
15:16:06 <dauwhe> laurent__: RMSDK has been using position with a basic calculation
15:16:17 <dauwhe> ... size of zipped resource in bytes, divided by 1024
15:16:29 <dauwhe> ... because they are using zipped html resource
15:16:36 <dauwhe> ... it has no real semantic meaning
15:16:56 <dauwhe> ... but it is easy to calculate
15:17:18 <dauwhe> ... we are wondering about using zipped vs unzipped HTML resource
15:17:34 <dauwhe> ... if we want to keep with RMSDK and the reading toolkits
15:17:48 <dauwhe> ... this is simple and adequate--zip divided by 1024
15:18:09 <dauwhe> dlazin: brady said Google does something similar, unicode code points divided by 1000
15:18:24 <dauwhe> ... once we have a recommendation, we should ask a RS to prototype a few things
15:18:50 <dauwhe> laurent__: we have tried with uncompressed by 2??? but is about the same as compressed / 1024
15:18:58 <dauwhe> ... you won't have interop until we agree
15:19:16 <dauwhe> ... if we try we can take everything
15:19:23 <dauwhe> dlazin: I think we agree it's arbitrary
15:19:28 <dauwhe> ... and we're trying to set a standard
15:19:56 <laurent__> s/2???/2500
15:20:00 <dauwhe> ivan: Laurent, could your team and google team write down their algo.
15:20:09 <dauwhe> ... so we could compare
15:20:44 <dauwhe> ... I don't know the result of this task force
15:20:52 <dauwhe> ... to have something documented as a first step is good
15:21:13 <dauwhe> ... I don't know what other reading systems do
15:21:41 <dauwhe> laurent__: apple recalculates every time the font size or viewport changes
15:21:52 <dauwhe> ivan: it depends on screen size, font size, etc
15:21:57 <dauwhe> ... that's not what we're looking for
15:22:00 <dauwhe> laurent__: exactly
15:22:24 <dauwhe> dlazin: if you provide a pagelist, Apple will use your pagelist and the screen number; you used to be able to toggle
15:22:41 <dauwhe> ... it will have both variable and fixed
15:23:18 <dauwhe> ivan: who else can we contact? Kobo?
15:23:44 <dauwhe> dauwhe: it may even depend on which Kobo implementation
15:24:00 <dauwhe> dlazin: I can't get info from Google for a bit, as Brady and Garth are out
15:24:16 <dauwhe> laurent__: yes, we can write down our heuristic algo
15:24:22 <dauwhe> ivan: perfect
15:25:09 <dauwhe> ivan: what about bluefire
15:25:24 <dauwhe> laurent__: old bluefire is RMSDK, new bluefire is Readium
15:25:48 <dauwhe> ivan: can someone ask about Japanese reading systems on Thursday night?
15:26:07 <dauwhe> laurent__: I have to leave now for a readium call
15:26:29 <dauwhe> dlazin: the time for this is arbitrary, we're open to moving it
15:27:08 <dauwhe> ivan: the epub call is 4pm CET / 10AM ET
15:27:19 <dauwhe> ... it's simpler for me if that slot is taken by EPUB
15:27:58 <dauwhe> dlazin: we are proposing to move to Friday 10AM ET
15:28:07 <dauwhe> ivan: the first would be June 25
15:28:13 <dauwhe> dlazin: we'll propose it
15:28:27 <dauwhe> pilarw2000: are we still doing wednesday nights?
15:28:40 <dauwhe> ivan: yes, but laurent and I can't participate
15:28:56 <dauwhe> dlazin: we won't move those
15:29:11 <dauwhe> ivan: we'll try and see what others say
15:29:47 <dauwhe> dlazin: with dave and ivan here it's a good time to talk about strategy
15:29:51 <dauwhe> ... what should we be producing
15:29:58 <dauwhe> ... and how do we get adopted by reading systems?
15:30:12 <dauwhe> ... let me lead the discussion
15:30:24 <dauwhe> ... maybe it's an experimental thing where we need adoption
15:30:32 <dauwhe> ... it's an attempt not a spec
15:30:37 <dauwhe> ivan: there is an intermediate thing
15:30:44 <dauwhe> ... if I become administrative
15:30:51 <dauwhe> ... a WG can publish WG notes
15:31:01 <dauwhe> ... about any subject
15:31:18 <dauwhe> ... sometimes it could be a design that is not final
15:31:42 <dauwhe> ... in this case my option would be that we have a doc published as w3c note on locator issue in general
15:31:48 <dauwhe> ... describing the problems and solutions
15:32:00 <dauwhe> ... and talk about the role of EPUB CFI
15:32:05 <dauwhe> ... and republish CFI with a note
15:32:14 <dauwhe> ... and the same note would document this algo
15:32:29 <dauwhe> ... this would not be a w3c recommendation
15:32:41 <dauwhe> ... would raise more interest; more people would look at it than a CG doc
15:32:46 <dauwhe> pilarw2000: that's what Wendy has said
15:35:31 <dauwhe> dauwhe: this sounds like incubation, and we could sell this
15:35:41 <dauwhe> avneeshsingh: is the ultimate objective the algo?
15:35:48 <dauwhe> pilarw2000: are we?
15:36:00 <dauwhe> dlazin: I think so? It's a part of what we're providing. It's like an appendix.
15:36:16 <dauwhe> ... we are striving for an agreement that all RSs should have a fallback
15:36:33 <dauwhe> ... we need to decide about backward compatibility
15:36:44 <dauwhe> ... should we address all the existing epubs in the world
15:36:52 <dauwhe> ... I don't think we need to
15:37:02 <dauwhe> ... I think it's acceptable to be an initiative for the future
15:37:11 <dauwhe> ... it would be OK for us to handle backwards compat
15:37:21 <dauwhe> ivan: what do. you mean by backward compatibilty
15:37:32 <dauwhe> ... today, if author provides a page list, it's clear what the RS should do
15:38:05 <dauwhe> ... if an EPUB does not have a pagelist, then the RS may do something but there is no incopatibility
15:38:22 <dauwhe> dlazin: do we need to support all existing EPUBs?
15:38:38 <dauwhe> .. is this in authoring, ingestion, or reading system?
15:38:47 <dauwhe> ivan: no author will calculate the bytes in the xip
15:39:02 <dauwhe> ... from my perspective, ingestion and RS is identical
15:39:09 <dauwhe> dlazin: I think RS makes the most sense
15:39:18 <dauwhe> ... you could do it in authoring
15:39:28 <dauwhe> ... it could be every 1k character
15:39:34 <dauwhe> ... and indesign could implement it
15:39:40 <dauwhe> avneeshsingh: as far as algo is concerned
15:39:48 <dauwhe> ... why do we worry about this?
15:39:58 <dauwhe> ... if there is an algo, it could be used anywhere
15:40:09 <dauwhe> dlazin: some are only practical in RS
15:40:21 <dauwhe> avneeshsingh: there are complexities in vertical writing
15:40:31 <dauwhe> ... how would that work?
15:40:45 <dauwhe> pilarw2000: indexing is done and embedding in doc. that's part of authoring
15:40:55 <dauwhe> ... I'm not sure I'd agree it's RS-dependent
15:41:17 <dauwhe> dlazin: if you have an index in a book, and the index header is "carrot"
15:41:25 <dauwhe> ... 12, 38, 44
15:41:32 <dauwhe> ... indexer needs to put links in
15:41:39 <dauwhe> pilarw2000: it's flipped around
15:41:49 <dauwhe> ... I've embedded the index entry in the text
15:42:03 <dauwhe> ... and then some process results in the pointer
15:42:08 <dauwhe> ... the page number comes after
15:42:21 <dauwhe> ivan: this means that whatever algo we come up wiht
15:42:40 <dauwhe> ... should be oblivous to whether it is done in RS or in authoring tool
15:42:50 <dauwhe> ... so shouldn't be length of zip file
15:43:02 <dauwhe> ... could be number of unicode characters
15:43:08 <dauwhe> dlazin: there is another alternative
15:43:19 <dauwhe> ... that's hard, though, for reasons avneesh explains
15:43:30 <dauwhe> ivan: I don't think avneeshsingh is right here
15:43:40 <dauwhe> ... unicode is there; how they are displayed is a different question
15:43:45 <dauwhe> avneeshsingh: it is different
15:44:08 <dauwhe> ... may be a small number of chars between pages
15:44:36 <dauwhe> dlazin: you need page numbers for xref or index
15:44:39 <dauwhe> ... so we could say
15:44:53 <dauwhe> ... the pagelist is primary, algo is fallback
15:45:01 <dauwhe> ... you can only do indexing and xrefs with a pagelist
15:45:12 <dauwhe> ... that lets algo be deferrred to the RS
15:45:22 <dauwhe> pilarw2000: you're also talking about TOC, endnotes
15:45:29 <dauwhe> dlazin: TOC doesn't need page numbers
15:45:37 <dauwhe> ... more important for index
15:45:42 <dauwhe> ivan: I'm not sure I follow
15:45:53 <dauwhe> ... the term authoring is too broad
15:46:40 <dauwhe> pilarw2000: I'm working with the book right before it's final
15:46:49 <dauwhe> ivan: the content is already there
15:47:34 <dauwhe> dlazin: if Pilar finds a book without a pagelist, I'd expect her to request a new version with a pagelist or make one herself
15:47:41 <dauwhe> ... there could be a fallback at authoring time
15:47:58 <dauwhe> ivan: that's the algo doable for both author and RS
15:48:14 <dauwhe> dlazin: if Pilar is inserting page numbers, she can use arbitrary numbers
15:48:27 <dauwhe> ivan: if indexing is not there, and RS still does it
15:48:41 <dauwhe> ... then you solve the use cases around shared locatio
15:48:52 <dauwhe> ... if the algo covers severaal use cases that's important
15:49:01 <dauwhe> pilarw2000: I don't generate the page numbers
15:49:13 <dauwhe> ... the publisher creates the page numbers for all the use cases
15:49:23 <dauwhe> ... InDesign is great for that
15:49:27 <dauwhe> dlazin: I still don't agree
15:49:40 <dauwhe> ... there's no problem if we use the same algo but I don't think it's necesary
15:49:55 <dauwhe> ... one case is a compiled ebook we don't touch, reading system generates pagelist
15:50:07 <dauwhe> s/pagelist/positions
15:50:23 <dauwhe> ... the other case is where the page numbers are written into the files
15:50:50 <dauwhe> ivan: is it better if it's the same algo
15:51:29 <dauwhe> dlazin: the algo is an implementation of the goals. It is an appendix.
15:51:37 <dauwhe> ... we derive the algo from our goals and requirements
15:51:48 <dauwhe> ... we want it to be useable at both authoring time and RS time
15:51:59 <dauwhe> ... and needs to work for all languages
15:52:12 <dauwhe> ... but doesn't need to work identically
15:52:20 <dauwhe> avneeshsingh: should be consistent within a language
15:52:39 <dauwhe> pilarw2000: as we go from spanish to english
15:52:53 <dauwhe> avneeshsingh: if you have ten books in that language it should work consistently
15:53:14 <dauwhe> dlazin: from strategy perspective
15:53:30 <dauwhe> ... we need to remember is to reach out to adobe and other authoring systems
15:53:35 <dauwhe> ... this will be vastly more useful
15:53:48 <dauwhe> ... the best case is for every epub to have a pagelist
15:53:59 <dauwhe> ... and for that we need ID and scrivener and Word
15:54:03 <dauwhe> ivan: yeah
15:54:14 <dauwhe> ... I have a script that produces EPUBs of W3C docs
15:54:20 <dauwhe> ... I start with HTML
15:54:30 <dauwhe> ... for me to generate a pagelist doesn't work
15:54:34 <dauwhe> ... I'd need a script
15:54:42 <dauwhe> dlazin: that's not the primary case
15:55:12 <dauwhe> pilarw2000: when I'm indexing, instead of using page numbers I have distinct text for every hit under a concept
15:55:26 <dauwhe> ... every instance has a unique chunk of text
15:55:33 <dauwhe> ivan: yes
15:56:16 <dauwhe> pilarw2000: we do need to tell people that's what you should do
15:56:42 <dauwhe> dlazin: next time, we should start working on the note
15:56:46 <dauwhe> ... we have things to write down
15:56:56 <dauwhe> ivan: is that all?
15:57:04 <dauwhe> avneeshsingh: thanks for helping me catch up
15:57:09 <dauwhe> dlazin: thanks for coming!
15:57:43 <ivan> rrsagent, draft minutes
15:57:43 <RRSAgent> I have made the request to generate https://www.w3.org/2021/06/09-epub-locators-minutes.html ivan
16:04:25 <dauwhe> Zakim, end meeting
16:04:25 <Zakim> As of this point the attendees have been dauwhe, avneeshsingh, pilarw, dlazin, Laurent, laurent__, ivan
16:04:28 <Zakim> RRSAgent, please draft minutes
16:04:28 <RRSAgent> I have made the request to generate https://www.w3.org/2021/06/09-epub-locators-minutes.html Zakim