14:46:49 <RRSAgent> RRSAgent has joined #epub-locators
14:46:49 <RRSAgent> logging to https://www.w3.org/2021/04/07-epub-locators-irc
14:47:10 <ivan> rrsagent, set log public
14:48:02 <ivan> Chair: wendyreid
14:48:02 <ivan> Date: 2021-04-07
14:48:02 <ivan> Meeting: EPUB 3 Working Group Virtual Locators Task Force Telco
14:56:20 <pilarw2000> pilarw2000 has joined #epub-locators
14:58:17 <wendyreid> wendyreid has joined #epub-locators
14:58:24 <wendyreid> present+
14:59:22 <avneeshsingh> avneeshsingh has joined #epub-locators
15:01:00 <BenSchroeter> BenSchroeter has joined #epub-locators
15:01:11 <ivan> present+ george
15:01:12 <avneeshsingh> present+
15:01:19 <ivan> present+
15:01:46 <dlazin> dlazin has joined #epub-locators
15:02:07 <pilarw2000> present+
15:02:16 <ivan> present+ tzviya
15:02:21 <BenSchroeter> present+
15:03:24 <tzviya> tzviya has joined #epub-locators
15:03:26 <George> George has joined #epub-locators
15:03:26 <mgarrish> mgarrish has joined #epub-locators
15:03:29 <tzviya> present+
15:03:35 <ivan> guest+ Pilar_Wyman, Ronnie_Seagren, Sergey_Lobachev
15:03:38 <dlazin> present+
15:03:46 <George> present+
15:04:29 <ivan> present+ mateus
15:04:47 <ivan> present+ mgarrish
15:05:47 <wendyreid> https://github.com/w3c/epub-specs/issues/1542
15:06:48 <dlazin> q+
15:07:19 <wendyreid> zakim, pick a victim
15:07:19 <Zakim> Not knowing who is chairing or who scribed recently, I propose mgarrish
15:07:45 <ivan> scribe+ mgarrish
15:07:54 <wendyreid> ack dlazin
15:08:38 <mgarrish> dlazin: we should start by laying out the problems we need to solve - we know people want to refer to page numbers from a book - easily link from an index to a place in an epub not necessarily a page
15:09:01 <George> q+
15:09:22 <tzviya> q+
15:09:23 <wendyreid> ack George
15:09:24 <mgarrish> ronnie: do we need to be able cover ranges
15:10:21 <mgarrish> George: we already have in the spec when an equivalent print page document exists there is page list in the nav doc and the accessibility spec requires the page list be there - options on how to display this in the page
15:10:29 <wendyreid> ack tzviya
15:10:54 <JuanCorona> JuanCorona has joined #epub-locators
15:11:21 <tzviya> http://idpf.org/epub/idx/
15:11:44 <mgarrish> tzviya: indexes are really important but we should avoid tying page breaks to them - need to figure out what is in scope - indexes specification lists requirements for indexes
15:11:48 <dlazin> q+
15:11:51 <wendyreid> ack dlaz
15:12:00 <ivan> q+
15:12:46 <wendyreid> ack ivan
15:12:47 <mgarrish> dlazin: my impression is that the community group is where new features are supposed to come from and wg is for standardizing - we can't dictate features exist and hope they get coded
15:13:29 <avneeshsingh> q+
15:13:37 <mgarrish> ivan: it is almost true - we do have the possibility to add new features with testing, etc. but we can't realistically add anything too elaborate - question of time
15:14:28 <mateus> mateus has joined #epub-locators
15:14:56 <JuanCorona> present+
15:15:01 <mateus> present+
15:15:03 <mgarrish> ... the original question is that there are two types of locators - one is relatively fixed like the page list that an epub creator can create and there are others that are completely dynamic - annotations are an example where you want pointers to the location
15:16:03 <mgarrish> pilarw2000: every epub needs a solid internal structure - locators is the best way to call them - an index uses them but needs to be present
15:16:17 <wendyreid> ack avneeshsingh
15:17:32 <mgarrish> avneeshsingh: from a process point of view we don't need to get into the details of whether we are creating a new spec but look at the use cases first and then decide what can be done non-normatively - there may also be some things for which we need a new revision but put these for later
15:17:36 <tzviya> +1 avneeshsingh
15:17:38 <mateus> +1
15:18:56 <mgarrish> ... introductions ...
15:24:54 <George> q+
15:28:49 <ivan> q+
15:29:26 <wendyreid> ack George
15:30:30 <wendyreid> ack ivan
15:30:34 <mgarrish> George: for background the indexing specification was developed under the idpf which didn't have strict implementation requirements which is why it is not well implemented - under w3c there are more rigorous implementation requirements
15:31:07 <mateus> q+
15:31:15 <mgarrish> ivan: the elephant in the room is what is the state of epubcfi - is it a tool we can rely on or put into recommendation track?
15:31:34 <mgarrish> ... if it is widely implemented perhaps we should consider it again
15:31:34 <wendyreid> ack mateus
15:32:25 <JuanCorona> +1 mateus
15:32:33 <mgarrish> mateus: my sense is that cfi is used silently for many different uses - we use it for deep linking for example and for extracting content within a container
15:32:57 <mgarrish> ... a larger implementation of it is in vitalsource bookshelf and for redshelf reader
15:33:12 <tzviya> +1 to mateus - CFI is broadly used under the hood
15:33:21 <mgarrish> ... I don't know if anyone loves it but it is the only choice we have - there is potential for improvement
15:33:27 <George> q+
15:33:36 <tzviya> q+
15:33:43 <mgarrish> wendyreid: commercial reading systems don't use cfi
15:33:43 <wendyreid> ack George
15:34:35 <ivan> q+
15:34:39 <mgarrish> George: if there was a locator roughly equivalent to a page and was in a publication would that allow us to use xpath to find an offset onto that page using tools that everyone has available to them?
15:34:42 <wendyreid> ack tzviya
15:35:35 <mateus> q+
15:35:51 <mgarrish> tzviya: when we were working on 3.1 garth conboy did a survey of google's content and there were no books that included cfi so from an authoring perspective it is not used - it won't be easy to replace and xpath is probably not the best option for longevity
15:36:04 <wendyreid> ack ivan
15:36:04 <mateus> +1 tzviya re xpath not being ideal
15:37:11 <dlazin> q+
15:37:18 <mateus> +100 web annotations
15:37:32 <JuanCorona> +1 web annotations
15:38:26 <mgarrish> ivan: xpath is being moved away from and has the problem that it can only point to structure not text - the other community we may want to turn to is web annotations - there are some documents coming out of that group that define a general way of addressing content - but may also be more aspirational than practical
15:39:14 <mgarrish> ... web devs want dynamic hooks using text before or after but not aware of any tools that would make it practical
15:39:22 <wendyreid> ack mateus
15:41:04 <mgarrish> mateus: annotation community would be great to involve and I would like to get dynamic locators to locations that change - as soon as you introduce a new section it's hard to ensure you can create persistent urls - hypothesis has done a lot of work on this with wikipedia and content that changes
15:41:26 <mgarrish> ... you need a selector that can point to the text
15:41:37 <wendyreid> ack dlazin
15:42:04 <tzviya> https://www.w3.org/TR/annotation-model/
15:42:49 <George> Hypothesis is planning a big push in Higher Ed.
15:43:09 <mgarrish> dlazin: one of the biases I bring to this is that ebooks are not as good as they ought to be because publishers don't have the control or time or expertise to ensure quality as they do with print - ebooks just get handed off inhouse or to conversion house - need accessibility for book creators who are not always very technical
15:44:19 <mgarrish> ... there is a plugin for indesign that can add page lists - not widely adopted and a book editor may not be aware of - we need foolproof solutions - for example be able to create a simple hash and use that
15:44:53 <mgarrish> ... we'd need to deal with the problem of content and hashes changing but need to drive people in direction of tools being in control
15:46:11 <mgarrish> wendyreid: need a path forward for this work - what are the use cases so what do we need these locators to do - the indexes spec is a good place to get some from - dynamic content is an example of where we need to explore - I can create a place in github where we can start listing these
15:46:44 <mgarrish> ... we need someone to lead the task force as well - keep track of what is going on and drive the work
15:47:05 <mgarrish> ... please consider if you would like to and get back to me
15:47:08 <dlazin> q+
15:47:13 <wendyreid> ack dlazin
15:47:44 <mgarrish> dlazin: can we discuss an Australia-friendly time for this meeting?
15:48:44 <mgarrish> ivan: harder when we have people from US west coast and Asia/Australia
15:49:02 <mgarrish> dlazin: maybe we could alternate meeting times?
15:49:50 <mgarrish> ivan: early afternoon time in europe equates to early US time might work
15:50:30 <mgarrish> ... maybe 8 or 9AM for US east coast might be 8 or 9PM for Australia
15:50:40 <George> it is 1:50 a.m. in australia right now
15:51:16 <mateus> q+
15:51:22 <mgarrish> wendyreid: I can send an email to see if we can find another time
15:52:08 <wendyreid> ack mateus
15:52:22 <mgarrish> mateus: this time collides with the publishing community group
15:53:42 <mgarrish> wendyreid: try to populate the wiki for the next meeting
15:55:17 <wendyreid> ack mateus
15:55:21 <mgarrish> ronnie: do you want to identify the actor in the use case - is it something an author or indexes would do, etc.
15:55:47 <mgarrish> wendyreid: yes, typical pattern is that "actor is trying to do X"
15:57:12 <mgarrish> s/indexes would do/indexes would do/
15:57:19 <mgarrish> s/indexes would do/indexer would do/
16:00:03 <ivan> rrsagent, draft minutes
16:00:03 <RRSAgent> I have made the request to generate https://www.w3.org/2021/04/07-epub-locators-minutes.html ivan
16:01:09 <ivan> rrsagent, draft minutes
16:01:09 <RRSAgent> I have made the request to generate https://www.w3.org/2021/04/07-epub-locators-minutes.html ivan
16:01:09 <ivan> rrsagent, bye
16:01:09 <RRSAgent> I see no action items
16:01:09 <ivan> zakim, bye
16:01:09 <Zakim> leaving.  As of this point the attendees have been wendyreid, george, avneeshsingh, ivan, pilarw, tzviya, BenSchroeter, dlazin, mateus, mgarrish, JuanCorona
16:01:09 <Zakim> Zakim has left #epub-locators