Publishing Working Group Telco — Minutes

Date: 2019-07-01

See also the Agenda and the IRC Log


Present: Tzviya Siegman, Wendy Reid, Dave Cramer, George Kerscher, Ric Wright, Deborah Kaplan, Ivan Herman, Luc Audrain, Laurent Le Meur, Franco Alvarado, Avneesh Singh, Rachel Comerford, Jun Gamou, Vladimir Levantovsky, Ben Schroeter, Garth Conboy, Benjamin Young, Bill Kasdorf, Marisa DeMeglio, Brady Duga, Gregorio Pellegrino, Jasmine Mulliken, Teenya Franklin

Regrets: Tim Cole, Joshua Pyle


Chair: Tzviya Siegman

Scribe(s): Nick Ruffilo, Wendy Reid


Tzviya Siegman:

Tzviya Siegman: Interesting agenda today - any comments on the minutes from last week?
… approved. I will mention that I put in a pull request for the summary - Ivan if you could approve it.

Resolution #1: last week minutes approved

Ivan Herman: will do.

Tzviya Siegman: Now we have the usual agenda item. I hope everyone read the proposal for the suggestions to the working group.

Tzviya Siegman:

Tzviya Siegman: Those of us involved in the discussion - the chairs of the WG and CG. We were talking about how we want to move forward with audiobooks - do we look at them from the web or epub context. The chairs are leaning towards web context - building off web publishing spec. We want your feedback as implementors and creators.
… then we’re going to talk about what’s next. This is the opportunity for us as a group on how ot move forward. We have to recharter, so this is a discussion that will affect things

Laurent Le Meur: I read the proposal and there is an existential crisis here. I believed we were closer to a result. When I see the different aspects of this note - first, I want to draw again the picture that I thought was set…
… there is web publication which is a core structure to represent publications on the web via a JSON manifest and an HTML TOC + resources. There is an audiobook profile that is a set of constraints that would be applied to this.
… so constraints on metadata - you apply specific data. And there are restraints on types of content - you can only apply audio content. We discussed that later there would be audio profiles. A text profile. An academic profile… A manga/graphic novels profiles….
… Different profiles can allow for different types of publications, and then have a packaging profile that can be applied to any other profile that would allow a publication to be exposed to the supply chain.
… the web publication - because it’s the core or template, it cannot be used as a stand alone standard - it must be used in part of a profile. This is why when we discussed at the F2F that someone talked about WP being a framework…

Bill Kasdorf: +1 to Laurent

Laurent Le Meur: I don’t like the word, it’s too large, but it’s a set of templates that can be used. So can a template be a standard? The 2nd question is the packaging method - can a file format be a w3c spec? But the definition is clean.
… The 2nd part - should we take epub and put it on the web? I’m not sure how we can make it any closer than it is now. it’s a set of HTML files + resources… XML is not loved on the web, but a set of XML files is more complex by browsers than JSON…
… and if we wanted to put epub closer to the web, we should get rid of XHTML and go to HTML

Bill Kasdorf: +1 to that too

Tzviya Siegman: I will point out a quick response. One thing that got this discussion started was - at the F2F - there was a counter commitment that they would not implement web publications. While I agree that WP is a framework…
… it’s a very vulnerable approach to specifications. We don’t have any other proposals but audiobooks at this time. It’s dangerous. We’re facing fatigue as we reach out to talk about testing and implementation - we got no response.
… we felt the working group was in need to get to a point that it would make something viable in the industry. The only real-world business case we have today is audiobooks.

Luc Audrain: As a publishing business co-chair, the co-chairs of the publishing business were not involved in the early discussion of the documents. Obviously this document will be discussed in the publishing business call…
… we have a feeling that there is some hasty decisions here. I was at the F2F and did not see a general consensus to NOT have WP as a rec track. There are some people hoping to have WP as well as Audiobook be rec track with future profiles.
… visual narratives has been proposed by the business group. I feel that web publication was not said to be adapted for scholarly publishing - which is and was at the very beginning was a force for this implementation…
… so we still need to have some explanation as to why WP wouldn’t be used for scholarly publications. We need to know the clear business usage for scholarly publishing.

Tzviya Siegman: The work on visual narratives has not yet come to the working group. As far as scholarly publishing, the web publication spec is not useful as it stands today to scholarly publishing.

Dave Cramer: I wanted to make a brief comment on profiles. Profiles is a way of organizing a spec and being able to identify a profile doesn’t necessarily identify a business need or a commitment to implement both on the part of content creators…
… or people who may digest what the content creators create. This is my major concern with web publications as specified today. The browsers aren’t participating in the discussion… I don’t foresee much of the book publishing industry using web publications anytime in the future. Even if this is an idea for the future, it doesn’t seem to be an idea for today.
… and our struggles in implementations and testing - I just see too few forces that see this as a solution to things that our industry really wants to do.

Tzviya Siegman: I would like to add: one of the key points in the proposal - we’re not ending web publications. We want to publish it as a note that could be revisited in the future. The one area we do see broad interest is audiobooks. Hopefully we’ll focus on that.

Avneesh Singh: I want to talk about the next year. We have a timeline which is approaching - June of next year until the charter we should be having something in rec track. Luc’s concerns are valid. What we heard in Boston, the reason why this direction is that we only have one profile.
… visual narratives will come, but WP will change. This is the reason why we had the discussion on having a base specification. I think it’s OK to publish as a WG note for now, then there will be major issues with huge complexities. We are not sure this framework is supportive to the profiles of the future.
… lets keep it as a note. If it should be a WP based audio or EPUB based audio. If we don’t have WP-based audio, then we’re killing a seed that could grow into a tree. Audiobooks could make it alive.

Luc Audrain: +1

Tzviya Siegman: +1 to Avneesh

Bill Kasdorf: +1

Wendy Reid: +1

Ivan Herman: +1 to Avneesh

Avneesh Singh: epub audio - daisy members say - if it’s not bringing additional value over the daisy format, why do it? Moving to WP based audio may be the same logic, but why not go for the future item and sow the seed.

Wendy Reid: I wanted to quickly respond to Luc’s concerns in the business group. We did bring it to the steering committee on Friday. I wanted to give more context to this discussion as the audiobooks decision is also - we’ve seen our group struggle…
… as slow as the standards world is, I think that publishing is slower. We work in an industry where publishing moves slow and doesn’t break things. If we put it in a note, that doesn’t say that in 5 years it isn’t resurrected…
… in attending the dpub summit, publishers are seeing a renewed interest in epub3. We were a bit ahead of the jump. Lets lean in to the thing thats going to work. There was enthusiasm for audiobooks. But keep moving forward in a way that continues fostering innovation in the industry instead of forcing them into something they aren’t ready for

Bill Kasdorf: I wanted to stress the concept of incubation instead of standardization. What I’m hearing is the working group saying “we have this framework/core to a specific state. We’re going on hiatus and either let audiobooks build their thing…”
… and let those developments inform our understanding of the core - is it spec ready or not. In parallel, lets let the community group keep defining epub3 - which is accepted by the industry.

Garth Conboy: +1 to Bill (too)

Ivan Herman: I don’t want to repeat what Avneesh and Wendy said. Let me just add some additional things to clarify what we are saying. Look at the web publication document I would say that two big areas…
… One is what we call the manifest - which is some sort of data structure to contain all the data structures that data needs. The other is how things would work if it was fully on the web - all the security issues, all the weird URLs, etc.
… the former - the manifest, the data structure: is not the real problem. The problem lies with the 2nd issue - without saying it clearly, we have already stepped back already when we’re talking about audiobooks, that they aren’t created for the web…
… So if we look at the web publication as it is today, what an implementation would mean is to have it fully on the web with all those security things with use cases, business model, etc.
… what we feel that within a year, there is no way we can properly do that with testing simply because there is no interest in that. The way we should consider it is that ‘yes we put it in a note, and maybe in a few years, people will be ready’
… on the other hand, we would take the vocabulary and put it together with the audiobook to create a standalone audiobook structure. It may not be 100% on the web, but that’s not the primary use and business case… Thats what we call an audiobook standard.
… And hopefully we’ll get business use case for this.
… when we look at the manga world, we can use the JSON-LD for the manifest, etc. We can make an editorial decision about if the audiobook stays in it’s own document or works together with a separate manifest specification. What we really put aside is this beautiful idea that doesn’t seem to have enough traction is to publish a book exactly on the Web that the browser can handle without special software…
… we feel there is absolutely no chance between now and next June. We’d have to have candidate recommendation next fall, and items coming in. A final word - the timing. Yes it is a big time, and yes it’s a hurry
… we may have to burn some steps too quickly - which includes how we brought the business group in late… The point is - without going into the details - only a week ago we didn’t have a clear idea…
… when we discussed it with the WG chairs - we have a charter, we have to finish it in a year - how are we going to do it? Last week we were still uncertain where to go…

George Kerscher: there was no mention of a profile for corporate documents, annual reports, formal documents that people want to publish. There’s no business around these types of publications but they are essential on the web. We’re suffering because there is no specification to do this, so people use PDF…
… I think that even though there is no corporation that sees this as a business case, I see it as a requirement for access to information in an information society. We could find a profile by anyone who wants to publish a document. the WHO has thousands of documents not accessible to those who need to read them.

Tzviya Siegman: that would first be brought to the business group.
… The two questions in the agenda: “Would you prefer to move forward with audiobooks in a web context or epub context?” I would like to hear from implementors.

  1. Would you prefer to move forward with audiobooks as is (in a Web Context) or audiobooks in an EPUB context? We would like to hear about implementation interest from both “listening system” vendors and content creators.
  2. What are the next steps for EPUB? What comes after EPUB 3.2? Going forward, we will ask you to identify particular gaps in the EPUB specification.

Avneesh Singh: Regarding the 2nd question. Epub 3 will gain a lot of traction after this move. One thing I should say is that we should not release the standards revision as the software release. We should wait for some adoption, gather feedback and then make the plans for the big version.
… now on to the rechartering - I see 2 drivers. Membership from publishing. How to sustain and acquire new members. The other is how to get people from the web world. In IDPF the membership fee was $5000, now we’re moving from 70-80k…

Luc Audrain: +1 to Avneesh

Avneesh Singh: why a publisher from the IDPF world would pay so much more for participating in a WG. There must be a compelling reason. We need to reach out. Many people from the publishing world were not in the WG in Lyon when we discussed this. We need to involve the epub3 CG and even beyond. Take a breath from immediate rechartering and do big outreach.
… My question for filling up gaps in web for publishing is - everyone in the W3C is good at pointing out the principles, but will they come and do the work to drive it forward and will they use such a specification. Whatever we do in WG, it must have a strong business case. That is the next step - to justify the jump in membership fee

Luc Audrain: About the future of epub - I wouldn’t say epub 3.2 is getting traction - just that the stability of epub has gained it traction. We have wanted to give the publishing and ebook industry something stable. 3.2 is an update from 2011.
… now we have something stable for the industry. It’s time to say adieu to epub 2… Now we have confidence with epubcheck. There are still business issues that need to be taken into account. There are spine and documents not in the same order - this is stability that is in question…
… a huge market needs to be listened to - and speaking to the future of epub and difficult issues, and an unendless evolution is very bad…

Tzviya Siegman: To clarify, this is not a revolution in epub. Nobody is going to leave epub sitting again - we need to figure out what is happening with it again. Is there going to be a 3.3? Will there be an epub 4?

Nick Ruffilo: I come at this from a different point of view: I liked the idea of taking a look at who we are, and there are a lot of users that would benefit from a webbier epub and all of our work. Every car manufacturer putting out manuals in PDF. We need to figure out how to get more representation from groups
… that aren’t part of traditional publishing

Geoff Jukes: +1 Nick

Nick Ruffilo: there’s more money/size in some of these companies, which can help us a lot. There’s more to publishing than just trade, scholarly, comics, we do not have full representation, what we are missing could push us in a new direction

Garth Conboy: Couple of comments - as we’ve been having discussions amongst the chairs - I’ve found myself waking up every other morning that audiobooks in the web world makes sense, and other days it seems epub…
… I think coming out of the WG proper, we’re not going to get traction on web publications. It’s likely we’ll get traction on audiobooks. I’ve come to that perspective that that is the path, so I endorse WP and audiobooks and LWP into a single spec for audiobooks…
… looking at epub - dave and team have done a good thing with epub3.2 it’s having adoption impact. I can’t quite say get rid of epub2, but we’re making good progress. I dont’ think we’re in a position to say what to do with epub… There are issues we can deal with in minor updates. We can discuss footnote popups, etc…
… on the other end, we looked at 3.1 items - HTML serialization, browser-friendly-format, etc. If we want to embark on that work - i think it’s all possible. We need business input, we need to watch what happens with epub2… We need time…
… we can leave that open to the community group when we decide what the right thing to do is.

Tzviya Siegman: I just wanted to add that we see - from our experience - that writing these things takes a very long time. We imagined for epub 3.1 and what we originally envisioned from web publications that will take a long time to work one…
… we need to engage with more of the W3C. We do not have a web-friendly format for fragment identifier… I’ve had people ask why scholarly publishing isn’t embracing - but if we can’t have fragment identifiers, we can’t get adoption. We do these things in isolation. We need to shift this towards more spec friendly way…
… that is why a future iteration needs to exist in a working group format. It’s not a decision we have to make today. We can’t just say epub 3.2 is doing well and just needs time. We know we will need years to get a new spec done.

Ivan Herman: I’m going back to another question that George and Nick raised - I’m playing devil’s advocate. We have a bunch of large companies publishing in PDF - but I could say ‘yeah, that’s correct, but why don’t they publish in epub and they can publish to the web in a reader like edge has’
… for those use cases, web publication is over the top. We would have taken an enormously huge step and I’m not sure we can justify it.

Tzviya Siegman: I would like to come out with some action items. Does the working group prefer web context or epub context. So we’d be building off the web publication spec. I’ve heard from kobo and google that the web publication spec is the way to go…

Geoff Jukes: I think web publications is the way to go. What this group is striving for is big and important. As a smaller publisher we get to do things that larger publishers don’t. Nick’s comments on how there’s an entire universe of publishing that has nothing to do with ‘publishers’
… i think shooting for the moon is the only way to get there. 10-15 years from now we’ll be accessing publications without anything special. Web publications is our way of starting on that path. It may not be the thing everyone adopts. W3C is in the position to start it. It has the brains & organizations and positions.
… A different example is the automotive industry. Love or hate Tesla - they have - through disruption - have created massive changes in an industry that hasn’t changed in so long. Publishing hasn’t changed in a very long time. And shoot for the moon. I think wpub is the way to go.

David Stroup: +1 to Web Pub and Geoff’s feedback

Tzviya Siegman: Another niche topic is synchronized media.

Wendy Reid: +1 to Geoff :)

Marisa DeMeglio: Luckily with sync media, we’re positioned to take whatever direction the group decides to take. I feel good about it. Our position is to wait and see - which is what we’ve been doing so far.

Laurent Le Meur: For audiobooks, we will implement audiobook profile by the end of the year. We’ve already implemented it using the readium manifest. It’s in the plan before the end of the year for mobile and desktop to have an audiobook integration.

Tzviya Siegman: that would be the web publication basis?

Laurent Le Meur: yes. Starting from an epub is a non-starter for us. We start from web publications - either readium or web publication.

Tzviya Siegman: Sounds to me like implementors would prefer web-publication start…

Nick Ruffilo: I wanted to address Ivan’s comment. As an entrepreneur, I just want a document on the web that works and I can sell. If it’s EPUB and anyone can read it without hoops, that would be great.
… at the end of the day, it could be EPUB as long as it works 100% of the web. When the web app manifest was released, it was a big change, allowing one app to be built and opened anywhere, we can to be going towards a universal document standard with accessibility, etc. I think that’s the vision I have.

Avneesh Singh: Indeed, a good way to read EPUB is required

Proposed resolution: Audiobooks will be based on WP (Tzviya Siegman)

Tzviya Siegman: +1

Deborah Kaplan: +1

Geoff Jukes: +1

Luc Audrain: +1

Nick Ruffilo: +1

Avneesh Singh: +1

Bill Kasdorf: +1

Ivan Herman: +1

Marisa DeMeglio: +1

Dave Cramer: +1 to web context, as Avneesh is very persuasive

Garth Conboy: +1

Ben Schroeter: +1

Wendy Reid: +1

Laurent Le Meur: +1

David Stroup: +1

Ric Wright: +1

Resolution #2: Audiobooks will be based on WP

Tzviya Siegman: we need to discuss the future of epub and WP, which is much bigger discussion - it’ll be a discussion for the business group as well
… it’ll be a discussion for the next few weeks. it would require a recharter, but we’ll have to talk about where epub fits. We have the consideration of the IDPF membership.

Ivan Herman: I suggest we refocus on audiobooks … for that we don’t have to recharter. The overall agreement is that if we take the scope - we can get by without rechartering.
… i would prefer NOT to recharter. What we cannot do in this working group is work on epub items that are not part of the charter. So lets go on with the charter as is, we’ll have to have some new publication plans
… the business group or community group needs to have a plan of what should be done with epub - if there is something to do with EPUB, should it be part of the WG? These are all open questions.
… By that time, it’ll be january or february of next year. It’s not like we have to do it immediately.
… that’s the fundamental idea

Tzviya Siegman: thank you all
… discussion to be continued. We will likely be talking about the details of audiobooks on the web.

1. Resolutions