13:59:50 <RRSAgent> RRSAgent has joined #pwg-manifest
13:59:50 <RRSAgent> logging to https://www.w3.org/2018/03/16-pwg-manifest-irc
14:00:03 <Zakim> Zakim has joined #pwg-manifest
15:07:29 <bigbluehat> bigbluehat has changed the topic to: Web App Manifest TF - project https://github.com/w3c/wpub/projects/3 | theme song 🎼
15:07:45 <bigbluehat> bigbluehat has changed the topic to: Web App Manifest TF - project https://github.com/w3c/wpub/projects/3 | theme song 🎼 https://www.youtube.com/watch?v=M5iSEdo5VNI&feature=youtu.be&t=57s
15:08:03 <bigbluehat> bigbluehat has changed the topic to: Web App Manifest TF - Project: https://github.com/w3c/wpub/projects/3 |  🎼 https://www.youtube.com/watch?v=M5iSEdo5VNI&feature=youtu.be&t=57s
16:26:34 <Zakim> Zakim has left #pwg-manifest
16:44:27 <ivan> ivan has joined #pwg-manifest
16:44:46 <Zakim> Zakim has joined #pwg-manifest
16:44:56 <ivan> rrsagent, set log public
16:45:35 <ivan> Agenda: https://lists.w3.org/Archives/Public/public-publ-wg/2018Mar/0079.html
16:59:47 <zheng_xu> zheng_xu has joined #pwg-manifest
17:00:03 <franco> franco has joined #pwg-manifest
17:00:39 <dauwhe> dauwhe has joined #pwg-manifest
17:00:58 <ivan> present+
17:01:02 <bigbluehat> present+
17:01:04 <Hadrien> Hadrien has joined #pwg-manifest
17:01:14 <rdeltour> present+
17:01:31 <tzviya> tzviya has joined #pwg-manifest
17:01:49 <zheng_xu> present+
17:01:53 <tzviya> present+
17:02:17 <Hadrien> present+
17:02:39 <franco> present+ Franco Alvarado
17:04:25 <dauwhe> scribenick: dauwhe
17:04:59 <bigbluehat> Our project space: https://github.com/w3c/wpub/projects/3
17:05:03 <dauwhe> bigbluehat: how do we come at these problems as a group, and organize our project?
17:05:22 <dauwhe> ... I've found everything with topic:manifest
17:05:35 <timCole> timCole has joined #pwg-manifest
17:05:44 <dauwhe> ... we're focusing on what might go into WAM
17:05:55 <dauwhe> ... they have an extension mechanism
17:06:14 <dauwhe> ... we need to figure out what we need to talk to WAM task force about
17:06:25 <dauwhe> ... feel free to fix any labeling issues
17:06:34 <dauwhe> ... any questions?
17:07:04 <dauwhe> ... we have an unknown column
17:07:19 <dauwhe> ... the definitely column has to go to WAM
17:07:30 <dauwhe> ... handled elsewhere means, 'this isn't us'
17:07:37 <mattg> mattg has joined #pwg-manifest
17:07:39 <ivan> q+
17:07:45 <rdeltour> q+
17:08:04 <dauwhe> ivan: there are two things
17:08:05 <tzviya> ack ivan
17:08:19 <dauwhe> ... when we say to WAM, "we need these things" or adaptations
17:08:37 <dauwhe> ... but sometimes what they have there we can use, even if the name is weird
17:08:43 <dauwhe> ... these two things are not the same
17:08:52 <dauwhe> bigbluehat: should we change column labels?
17:09:00 <dauwhe> ivan: a new column where we say we just use WAM
17:09:15 <Hadrien> https://github.com/w3c/wpub/issues/127#issuecomment-363063172
17:09:16 <dauwhe> ... Hadrien made a list of those
17:09:34 <dauwhe> bigbluehat: some of these are large issues, which may be removed from this board
17:09:50 <tzviya> q?
17:10:45 <dauwhe> rdeltour: some of these issues are buckets indeed, and we may split them into smaller issues
17:10:52 <dauwhe> ... these columns are helpful for triage
17:11:16 <dauwhe> ... we have to figure which features can be added with current WAM extension mechanism
17:11:42 <dauwhe> ... and which features might need an additional layer on top of WAM, like an API
17:11:53 <tzviya> Hadrien's proposal for using WAM https://github.com/w3c/wpub/issues/118
17:11:59 <dauwhe> ... we can refine labels and issues
17:12:24 <tzviya> and https://github.com/w3c/wpub/issues/127
17:12:27 <bigbluehat> q?
17:12:31 <bigbluehat> ack rdeltour
17:12:32 <rdeltour> ack rd
17:12:34 <dauwhe> bigbluehat: let's work through this left-hand column
17:13:20 <bigbluehat> topic: https://github.com/w3c/wpub/issues/32
17:13:23 <dauwhe> ... we should close #32 and take it off the board
17:13:31 <dauwhe> ivan: one of those horrendously long things...
17:13:53 <dauwhe> tzviya: ivan, do you think we can close issues here?
17:14:06 <dauwhe> ivan: we should propose it, but let group decide.
17:14:12 <bigbluehat> PROPOSAL: mark #32 as "propose to close" and remove from WAM TF project
17:15:22 <dauwhe> rdeltour: we could leave open and continue to comment
17:15:40 <dauwhe> bigbluehat: let's leave open and mention in other issues
17:15:46 <rdeltour> +1
17:16:09 <ivan> +1
17:16:22 <timCole> +1
17:16:30 <dauwhe> bigbluehat: new proposal is to leave open but remove from project
17:16:36 <bigbluehat> PROPOSAL: leave #32 open but remove it from the project; mention it in other issues
17:17:11 <dauwhe> ... issue #127 is next, it's a similar thing
17:17:23 <dauwhe> ... with more actionable commentary, and links to other issues
17:17:26 <bigbluehat> topic: https://github.com/w3c/wpub/issues/127
17:17:28 <rdeltour> q+
17:17:29 <bigbluehat> https://github.com/w3c/wpub/issues/127#issuecomment-362548707
17:17:46 <tzviya> ack rdeltour
17:17:55 <dauwhe> rdeltour: let's add another column for issues from which we want to extract smaller issues before closing
17:18:23 <bigbluehat> PROPOSAL: move #127 into new Extract Smaller Issues column
17:18:29 <ivan> +1
17:18:33 <timCole> +1
17:18:34 <bigbluehat> +1
17:18:39 <rdeltour> +1
17:19:19 <bigbluehat> topic: https://github.com/w3c/wpub/issues/54
17:19:40 <dauwhe> bigbluehat: this may be one of the more general things, and it might be mislabeled
17:20:00 <bigbluehat> "Obtaining language from http headers" is the title of #54
17:20:15 <dauwhe> ivan: if I do an http request to the publication, I may get a language tag
17:20:23 <Hadrien> q+
17:20:24 <dauwhe> ... the way the issue is formulated is not exact
17:20:34 <dauwhe> ... the question is what do we do with that language tag?
17:20:51 <dauwhe> ... that language tag is the language of the resource itself, and we can't change that
17:21:05 <dauwhe> ... the language we use in the manifest is something else
17:21:13 <dauwhe> ... it is the language of the terms in the manifest
17:21:36 <dauwhe> ... whatever comes in from HTTP is irrelevant for the manifest itself
17:21:52 <rdeltour> ack ha
17:21:54 <dauwhe> Hadrien: that's not entirely true
17:22:03 <dauwhe> ... what you said is right for the resources of the publication
17:22:10 <dauwhe> ... for the WAM, people do translate the manifest itself
17:22:27 <bigbluehat> q+
17:22:27 <dauwhe> ... based on accept-language header, they do negotiation, and return a translated manifest
17:22:43 <dauwhe> ... so we need to separate language of resource from language of manifest
17:22:49 <dauwhe> ivan: I understand
17:22:59 <dauwhe> ... I have impression that what they do is incorrect
17:23:18 <dauwhe> Hadrien: on the web your browser requests a language, and server returns that language
17:23:37 <dauwhe> ivan: if I make an http request, I get an html file, and from the html file I link to the manifst
17:23:48 <dauwhe> ... what I get from http is the language of the resource
17:23:55 <dauwhe> Hadrien: you can do indirection
17:24:09 <dauwhe> ... if I say accept-langage: fr I might get french manifest
17:24:14 <rdeltour> ack bigb
17:24:22 <dauwhe> bigbluehat: it's very nuanced and not related to our task with WAM
17:24:34 <dauwhe> ... the WAM inbound language stuff defines only the content of the WAM
17:24:43 <dauwhe> ... it does not dictate language of resources of the web app
17:24:49 <Hadrien> q+
17:25:05 <dauwhe> ... I'm not sure this is a WAM issue, but it may be an issue of publication-wide...
17:25:11 <dauwhe> ivan: this is a handled elsewhere item
17:25:15 <rdeltour> ack ha
17:25:22 <dauwhe> Hadrien: right now our infoset is consistent with WAM
17:25:29 <dauwhe> ... and in our WebIDL
17:25:41 <dauwhe> ... it provides default language for members of manifest
17:25:45 <dauwhe> ... but we're consistent
17:25:50 <dauwhe> ivan: the http header does that
17:25:59 <dauwhe> ... I would say that issue is subject to close
17:26:08 <rdeltour> q+
17:26:15 <dauwhe> Hadrien: there could be lang negotiation on resources ourselves
17:26:33 <dauwhe> bigbluehat: the way the web works is no one considers the WAM language a default
17:26:39 <tzviya> q?
17:26:43 <dauwhe> Hadrien: we're doing the same thing in the infoset
17:26:48 <dauwhe> ivan: it's only about the metadata
17:26:57 <bigbluehat> PROPOSAL: move #54 to "handled elsewhere" column and possibly mark as "possibly close"
17:27:11 <rdeltour> q-
17:27:12 <dauwhe> rdeltour: sounds good
17:27:23 <bigbluehat> +1
17:27:42 <bigbluehat> topic: https://github.com/w3c/wpub/issues/58
17:27:42 <dauwhe> bigbluehat: next issue is #58
17:27:52 <dauwhe> Hadrien: this is infoset related. shouldn't be here
17:28:01 <dauwhe> bigbluehat: do we have an infoset label?
17:28:03 <dauwhe> ivan: no
17:28:20 <dauwhe> Hadrien: there's a label for metadata
17:28:23 <bigbluehat> PROPOSAL: remove topic:manifest label from #58 and move to "handled elsewhere" column
17:28:42 <bigbluehat> topic: https://github.com/w3c/wpub/issues/67
17:28:45 <dauwhe> bigbluehat: #67
17:28:59 <dauwhe> ... should linking from manifest support URI templates?
17:29:05 <dauwhe> ivan: guilty on that one
17:29:19 <dauwhe> ... I would postpone to next version. Let's not go down that road now.
17:29:26 <dauwhe> tzviya: we have a defer label
17:29:45 <Hadrien> q+
17:29:53 <dauwhe> (musical interlude)
17:30:01 <rdeltour> ack ha
17:30:12 <dauwhe> Hadrien: your idea of using uri template was different from mine
17:30:29 <dauwhe> ... and easy way of declaring a bunch of resources are part of default reading order etc
17:30:35 <dauwhe> ... i think that use case is impossible
17:30:49 <dauwhe> ... but for providing user search or api, we're going to need URI templates
17:31:03 <dauwhe> ... the impact on webidl and spec is minimal
17:31:15 <dauwhe> ... we just need a flag for a URL
17:31:38 <dauwhe> ... it's fine to defer, but we need to separate this concept from using templates for declaring a list of resources
17:31:43 <dauwhe> ivan: as I said, postpone
17:31:49 <bigbluehat> PROPOSAL: remove #67 from the WAM project board and mark as "postpone"
17:32:01 <rdeltour> +1
17:32:01 <ivan> +1
17:32:04 <dauwhe> ... I don't think this is a part of a minimum viable project
17:32:06 <timCole> +1
17:32:06 <dauwhe> +1
17:32:13 <dauwhe> q?
17:32:21 <bigbluehat> topic: https://github.com/w3c/wpub/issues/59
17:32:21 <dauwhe> bigbluehat: next is #59
17:32:33 <dauwhe> ... avoiding resource declaration duplication
17:32:39 <Hadrien> q+
17:32:42 <rdeltour> q+
17:32:49 <rdeltour> ack ha
17:32:54 <dauwhe> Hadrien: I think this is easy to solve
17:33:08 <dauwhe> ... in RWPM what's in the reading order doesn't need to be in list of resources
17:33:17 <dauwhe> ... EPUB has a lot of duplication, and the id/idref thing
17:33:48 <dauwhe> ivan: in a sense, this is a meta issue, this is how we should do all our thing
17:33:52 <tzviya> q?
17:33:54 <dauwhe> ... keep it as handled elsewhere
17:34:01 <dauwhe> Hadrien: infoset and serialization
17:34:06 <bigbluehat> PROPOSAL: move #59 to "handled elsewhere"
17:34:08 <dauwhe> rdeltour: it's also infoset
17:34:08 <tzviya> ack rdeltour
17:34:20 <dauwhe> bigbluehat: move to 'handled elsewhere'
17:34:33 <dauwhe> ivan: then I will also add topic:metadata
17:34:35 <bigbluehat> topic: https://github.com/w3c/wpub/issues/63
17:34:40 <dauwhe> bigbluehat: next is #63
17:34:50 <dauwhe> rdeltour: remove from project board
17:35:30 <dauwhe> bigbluehat: this is not related to manifest
17:35:46 <bigbluehat> PROPOSAL: remove #63 from the board; change labels
17:36:02 <dauwhe> tzviya: we'll call it "bikeshed" :)
17:36:12 <bigbluehat> topic: https://github.com/w3c/wpub/issues/163
17:36:19 <dauwhe> bigbluehat: next is #163
17:36:34 <bigbluehat> q?
17:36:41 <dauwhe> ivan: this is indirectly with the WAM
17:36:54 <dauwhe> ... we want from WAM we may want to link to other metadata
17:36:55 <rdeltour> q+
17:37:02 <Hadrien> q+
17:37:06 <dauwhe> ... the privacy policy may have it's own vocab
17:37:13 <dauwhe> ... just like linking to ONIX
17:37:13 <tzviya> ack rdeltour
17:37:20 <dauwhe> rdeltour: I agree there's a general issue
17:37:34 <dauwhe> ... but there might be differences depending on the manifest
17:37:41 <dauwhe> ... some handled in WAM, some externally
17:38:02 <bigbluehat> PROPOSAL: move #163 to "possibly" column
17:38:08 <dauwhe> ivan: now we say "possibly"
17:38:12 <rdeltour> ack ha
17:38:27 <dauwhe> Hadrien: this one is mine along with linking to external metadata record and a third one... linking to alt rep
17:38:37 <dauwhe> ... I think all three should be "possibly"
17:38:37 <bigbluehat> also https://github.com/w3c/wpub/issues/162
17:38:48 <dauwhe> ... they could have a common tech solution
17:39:03 <bigbluehat> also https://github.com/w3c/wpub/issues/159
17:39:27 <bigbluehat> PROPOSAL: move to #159, #162, #163 the "possibly" column
17:39:37 <bigbluehat> q+
17:39:43 <dauwhe> Hadrien: WAM has links, they are just specialized
17:39:57 <dauwhe> ivan: they are link to specific resources, which is different than linking to metadata
17:39:58 <rdeltour> q+
17:40:11 <rdeltour> ack bi
17:40:25 <dauwhe> bigbluehat: it's not the expecation of this task force to find a home for all things in WAM
17:40:40 <dauwhe> ... we don't know if we'll go there with our entire infoset
17:40:48 <Hadrien> q+
17:40:54 <dauwhe> ... depends on what they're willing to do, and what might be better elsewhere
17:41:17 <dauwhe> rdeltour: Marcos was lukewarm about generic metadata in WAM
17:41:40 <dauwhe> ... the basic principle was that metadata in WAM should be directly beneficial to end user (agent)
17:41:49 <bigbluehat> q+
17:41:55 <bigbluehat> ack rdeltour
17:42:09 <dauwhe> ... if we think otherwise, then we need to document with use cases and build our case to present to wam editors (and TAG)
17:42:11 <bigbluehat> ack Hadrien
17:42:27 <dauwhe> Hadrien: for  alt rep and privacy, these are actionable for end user
17:43:13 <dauwhe> ... I think we need to make sure that whatever is publication specific should be at collection level, in WAM
17:43:25 <dauwhe> ... most can go through generic extension mechanism
17:43:30 <rdeltour> q+
17:43:36 <dauwhe> ... we should only worry about when we have conflicts
17:43:46 <dauwhe> ... or things that are useful for the larger community
17:44:01 <bigbluehat> ack bigbluehat
17:44:13 <dauwhe> bigbluehat: I don't think we're looking at the WAM cohesively
17:44:20 <dauwhe> ... it's just config settings for a runtime
17:44:32 <dauwhe> ... and category data for app stores
17:44:52 <dauwhe> ... the things that might make it into WAM are things that might help with creating a runtime, like a privacy policy
17:45:01 <dauwhe> ... right now they say link to it
17:45:22 <dauwhe> ... privacy settings might go in a manifest, but not a privacy policy
17:45:37 <dauwhe> ... there's a conceptual difference between what WAM is and what they expect it to be
17:45:59 <bigbluehat> ack rdeltour
17:46:24 <dauwhe> rdeltour: when Hadrien said that we can solve things with extensions, we don't have to go to the group
17:46:41 <Hadrien> q+
17:46:43 <zheng_xu> ?
17:46:47 <zheng_xu> q+
17:46:49 <dauwhe> ... I think we should go to them, we would hope that some of our extensions are implemented by browsers
17:47:06 <bigbluehat> ack Hadrien
17:47:13 <dauwhe> ... we should liase with the editors even as we use their extension mechanism
17:47:19 <dauwhe> Hadrien: we have to do that anyway
17:47:43 <dauwhe> ... there's a big difference between using an extension, and changing the manifest itself
17:48:18 <dauwhe> ... I disagree with Benjamin. If we're not using the wam for defining a collection, and we spread information around, I don't see the point of using the WAM
17:48:31 <dauwhe> ... for an extension we would declare reading order, and if we don't what's the point?
17:48:32 <bigbluehat> ack zheng_xu
17:49:03 <dauwhe> zheng_xu: I get confused about who consumes the manifest? publisher? reading system vendor?
17:49:24 <dauwhe> ... if we define affordance layer, it could be extension of WAM
17:49:40 <dauwhe> ... but a web publication is different than web app, as the creator is not a developer
17:49:54 <dauwhe> ... I want to focus more on infoset which is related to creator/publisher
17:50:03 <dauwhe> ... might not to need extension or related to WAM
17:50:05 <bigbluehat> q?
17:50:18 <bigbluehat> topic: https://github.com/w3c/wpub/issues/73
17:50:36 <dauwhe> bigbluehat: #73, modification date
17:50:49 <bigbluehat> PROPOSAL: relabel 73 as info set; remove from project
17:50:49 <dauwhe> ... relabel and remove, as it's an infoset question
17:50:52 <rdeltour> +1
17:51:07 <dauwhe> bigbluehat: next is 118
17:51:11 <bigbluehat> topic: https://github.com/w3c/wpub/issues/118
17:51:18 <dauwhe> ... break up into smaller things?
17:51:20 <dauwhe> Hadrien: yes
17:51:46 <bigbluehat> toic: https://github.com/w3c/wpub/issues/119
17:51:49 <dauwhe> bigbluehat: #119, using RWPM
17:51:50 <bigbluehat> s/toic/topic
17:52:07 <dauwhe> Hadrien: not related to WAM, but an alternate proposal
17:52:23 <dauwhe> ... even if we go with WAM this can inspire extensions
17:52:48 <bigbluehat> topic: https://github.com/w3c/wpub/issues/122
17:52:54 <dauwhe> ivan: JSON manifest as script element--we have to discuss this with WAM
17:52:59 <dauwhe> ... we shouldn't spend time on it now
17:53:10 <bigbluehat> topic: https://github.com/w3c/wpub/issues/148
17:53:10 <dauwhe> Hadrien: this is mostly useful for single resource publication
17:53:18 <dauwhe> bigbluehat: #148
17:53:30 <dauwhe> ... limiting reading order to manifest
17:53:34 <dauwhe> Hadrien: infoset-related
17:53:46 <bigbluehat> removed from the board
17:53:51 <bigbluehat> https://github.com/w3c/wpub/issues/126
17:53:57 <dauwhe> bigbluehat: 126, progression direction
17:54:21 <Hadrien> q+
17:54:37 <dauwhe> ... probably remove or handled elsewhere?
17:54:50 <dauwhe> ivan: it's something that we would put in our extension, and they shouldn't fuss
17:54:54 <rdeltour> ack ha
17:55:04 <dauwhe> Hadrien: this goes in possibly, because it's part of our extension
17:55:33 <bigbluehat> topic: https://github.com/w3c/wpub/issues/167
17:55:33 <dauwhe> bigbluehat: 167
17:55:47 <dauwhe> ... roles for creators
17:56:14 <dauwhe> Hadrien: how do we express creators in general in our serialization?
17:56:24 <dauwhe> ... this is definitely possibly
17:56:29 <zheng_xu> q?
17:56:29 <dauwhe> ... we should discuss with WAM
17:56:36 <dauwhe> ... as this is not publication specific
17:56:43 <zheng_xu> q+
17:56:51 <bigbluehat> ack zheng_xu
17:56:52 <rdeltour> acj zh
17:56:56 <rdeltour> ack zh
17:57:04 <rdeltour> s/acj zh//
17:57:07 <tzviya> see ivan's comment https://github.com/w3c/wpub/issues/167#issuecomment-373455166
17:57:08 <dauwhe> zheng_xu: there is a lot of duplication between creators and ONIX
17:57:12 <Hadrien> q+
17:57:24 <bigbluehat> q+
17:57:25 <dauwhe> ... can this information be related similar to ONIX?
17:57:35 <bigbluehat> ack Hadrien
17:57:36 <rdeltour> ack ha
17:57:46 <dauwhe> Hadrien: handling ONIX is like spending your holiday in hell
17:58:01 <dauwhe> ... and we shouldn't assume we'll get ONIX
17:58:16 <dauwhe> ... and there is never ONIX for some kinds of publications
17:58:19 <bigbluehat> ack bigbluehat
17:59:32 <dauwhe> bigbluehat: the wam's way of doing metadata is scary, as they make a new top-level key for everything
18:00:28 <dauwhe> Hadrien: you need creator/dev/company even in web apps
18:00:40 <dauwhe> ... MS has been using manifest for their store
18:00:58 <bigbluehat> PROPOSAL: move #167 to "possibly" and add refined use cases
18:01:07 <dauwhe> ivan: we're at the top of the hour
18:01:07 <rdeltour> +1
18:01:46 <dauwhe> ivan: #150 is comments to Brady from Chrome team; 90% was for infoset rather than what we're discussing now
18:02:06 <dauwhe> ... let's put in handle elsewhere column
18:02:30 <Hadrien> q+
18:02:43 <bigbluehat> ack Hadrien
18:04:42 <rdeltour> Next call: March 28th (2 weeks from today) @ 14:00 UTC
18:05:16 <ivan> rrsagent, draft minutes
18:05:16 <RRSAgent> I have made the request to generate https://www.w3.org/2018/03/16-pwg-manifest-minutes.html ivan
18:06:01 <ivan> https://github.com/w3c/scribejs
18:06:58 <ivan> zakim, bye
18:06:58 <Zakim> leaving.  As of this point the attendees have been ivan, bigbluehat, rdeltour, zheng_xu, tzviya, Hadrien, Franco, Alvarado
18:06:58 <Zakim> Zakim has left #pwg-manifest
18:08:09 <ivan> rrsagent, bye
18:08:09 <RRSAgent> I see no action items