Date: 23 May 2017
See also the Agenda and the IRC Log
Present: Dave Cramer, Rick Johnson, Avneesh Singh, Mateus Teixeira, Ivan Herman, George Kerscher, Murata Makoto, Tzviya Siegman, Ric Wright, Bill Kasdorf, Cristina Mussinelli, Matt Garrish, Daniel Bennett, Liisa McCloy-Kelley, Jens Klingelhöfer, Liam Quin, Leslie Hulse, Garth Conboy, Junko Kamata, Julian Calderazi, Bill McCoy, Berhard Heinser, David Stroup
Regrets: Mike Baker, Nick Ruffilo, Greg Davis, Rachel Comerford, Paul Belfanti, Luc Audrain, Ryan Pugatch
Guests:
Chair: Rick
Scribe(s): Dave Cramer
Wolfgang Schindler: hi everybody!
Wolfgang Schindler: present + Wolfgang
Wolfgang Schindler: good afternoon, Dave!
George Kerscher: Hello folks
Wolfgang Schindler: hi George!
Rick Johnson: we'll get started in a minute
Garth Conboy: present +Garth
Rick Johnson: let's get started
Ivan Herman: I should never be optimistic :)
… the consultation with the AC is over
… we have 52 votes in
… all 52 votes said they want this work to be done
… no objection to doing this work at all
… it is 52; there were 35 who said they were interested in participating
… 26 intend to develop projects or use the technology
… the difficulty is that we have an IPR patent policy, which means lawyers look at these charters
… so we have three objections saying we want this work to happen, but we want changes to the charter
… I have to be vague, because not all comments are public
… so the team is working on another version of the charter, to make it more palatable to lawyers
… I saw the latest draft 10 minutes ago
… there is nothing in those changes that really change what we want to do
… it's all reformulating, making it more concise, making deliverables clearer
… I hope this version will go out to those who formally objected
… and see if this will work for them
… I am cautiously nearly almost sort of optimistic
… then we have to contact all 53 voters to see if they are ok with the new draft
… they will get five business days to review
George Kerscher: Check if you are muted, because we are getting background noise.
Ivan Herman: if that works, then we are done
Rick Johnson: we had talked two weeks ago about reaching out to those interested about the F2F meeting
Ivan Herman: what the team came up with is that the June F2F is a joint meeting of the BG and the IG
… this means anyone who is in the BG can come to this meeting
… the BG has to decide if this is OK
… then we can announce it that way
… if, by then, the WG is operational, then the WG will ... find an administrative way of saying it's a WG meeting
… the only reason we have to be careful is the same IPR issues
Garth Conboy: RVSP to F-to-F and draft agenda: https://docs.google.com/document/d/1J0VlZbMFj-33tfhZe0EK543NWFlx6dyDHauqkPwMeEI/edit#heading=h.6sesbeqec1qm
Garth Conboy: Better: https://docs.google.com/document/d/1J0VlZbMFj-33tfhZe0EK543NWFlx6dyDHauqkPwMeEI/edit
Ivan Herman: we have to play by the rules
… there's no problem announcing it
… I know that Wendy plans to reach out to you
… you could tell Wendy if it's OK for this to be a BG/IG meeting
Rick Johnson: people who show up create standards, so is it appropriate to call for objections to having the f2f be a BG/IG meeting?
… does anyone have any objections to formally make the F2F a joint BG-IG meeting?
George Kerscher: if this becomes a WG meeting
… do WG members have to sign the IPR agreement in order to participate?
… or is it automatic when you participate?
Ivan Herman: the joining process--when the AC rep nominates someone to a group, it takes place then
Garth Conboy: we want to make clear this is not a general BG meeting; I expect this to be much like a WG meeting
… focused on PWP/WP/EPUB4
Rick Johnson: my expectation is that we would use the facilities available now to announce the meeting
… then update agenda to describe that it is, in fact, a WG meeting
Garth Conboy: I pasted a link to the suggested agenda
… that gives the flavor of what will happen
Bill Kasdorf: is there an issue if someone that is in the BG that's not in the WG, travels to NYC, and is there a problem?
Ivan Herman: no
Ivan Herman: formally speaking, we can think of it as a BG/IG meeting. Let's not overthink this.
Bill Kasdorf: that's what I was hoping you would say
Ivan Herman: looking at the agenda, we should be careful to avoid issues that lead to IPR problems
Rick Johnson: any objections? Any concerns about making this a joint meeting?
Liisa McCloy-Kelley: joint meeting +1
George Kerscher: just want to make sure it's a joint meeting between IG and BG that is working for plans for when the WG officially kicks off
Ivan Herman: that's a wonderful formulation
Resolution #1: the June F2F is a joint meeting between the IG and the BG
Rick Johnson: I'll tell Wendy
Rick Johnson: it looks like Brian isn't here
Bill Kasdorf: I don't have anything; we should wait for Brian
Rick Johnson: we'll postpone this item.
Rick Johnson: I'll swap the next 2 items
Rick Johnson: CG will work on part of this
… I have committments from lots of publishers to work on this
… my expectation is that within the next few weeks we can move formalize both the CG side and the IMS side
… we at VitalSource will be proposing some specifics around the outcomes side
… and we will volunteer to work with EPUBCHeck to support that
… not sure what else we can do with EPUBCheck
… we didn't have time to talk through the epub part last week
Rick Johnson: we need resources
… BillK and Liisa were champions for this
Tzviya Siegman: https://github.com/IDPF/epubcheck/wiki/WorkPlan
Rick Johnson: does anyone has something to discuss around epubcheck, and how to keep it relevant and current
Bill Kasdorf: Tzviya posted a link to the epubcheck work plan
Liisa McCloy-Kelley: the big thing I want to figure out is
… when and how do we get people to shift away from older versions of epubcheck
… from a publishers perspective, the errors are different depending what version people are using
… how do we deprecate older epubchecks?
Tzviya Siegman: I'm not sure we can deprecate older versions
… when epubcheck 4 was released, the old version is not really available for download
… epubcheck can download older versions of EPUB
Liisa McCloy-Kelley: I'm worried about retailers integrating older versions of epubcheck into their ingestion systems
Tzviya Siegman: there's a lot of misunderstanding aobut epubcheck releases
Ric Wright: i don't know how you get people to move on from an old version
… Apple is using an old version which only checks 3.0, and so some people will want to us e that one
Garth Conboy: we're guilty of running a 3.x version
… it takes more than a month to integrate a new version into our system
… this time we're going to push some changes back to epubcheck, to make integrations easier
… it's hard to take a new version, whitelist new components, etc.
… if we focus on making that easier, it could help
Rick Johnson: do we know where epubcheck lives within w3c?
Bill McCoy: the BG can create task forces
… or it could be the CG
… the code is on github
… for "live" I think you mean who does what
… some of the people doing the work might not be in the BG/CG/IG/WG
… readium foundation has offered some help
… we want to consider it as part of the publishing@w3c umbrella
… or we could have another CG, but that might be too many groups
George Kerscher: this relates to DAISY's planning for a11y checker
… if someone gets a product out in the market that integrates these things
… then if these components change there could be a problem
… we need to figure out a way to figuring out what versions are being used, and then enable people to move to the next version
Rick Johnson: I think they'll be something like epubcheck for whatever comes out of the WG
… that will need to be part of the conversation
… Bill, I need to turn back to you here. We need someone to take responsibility to drive this forward.
… whether it's alignment with DAISY, with the CG, where's the best fit?
Bill McCoy: from my POV, if the EPUB3 CG, was about to do 3.11, one of the plank for the plan for 3.11 would be tests and EPUBCheck
… but we're behind the 8 ball, as we didn't do 3.1 in epubcheck before the merger
… the unfinished business is that epubcheck is behind
… that would be a good action item for me
… but there's also a larger issue of epubcheck integration with the larger ecosystem
Garth Conboy: … epubchuck is hard, it’s real programming
Garth Conboy: … often a problem for standards WG’s
Garth Conboy: … unclear how we make that work with the volenteer nature of WG participation (and sometimes semi-technical)
Dave Cramer: epubcheck is production code, which is not a natural deliverable of a WG
Tzviya Siegman: when I talk to Romain, he says it's hard to work on 'cause it's spaghetti code
… if you look at that writeup, he's proposing an overhaul of the whole system
… it's java-based, and even many of us who code don't know java
… it's not just a wish-list, it's a matter of real time
Bill Kasdorf: another thing we're lacking is a clear identity of who's the owner
… the point person
… I am not that person
… because it was a collaborative effort between three organizations, it was never clear who was in charge
… we need someone with public visibility to be the point person
Rick Johnson: I'm going to jump the queue
… Tzviya, you're writing in IRC, are you looking to be that person?
Tzviya Siegman: I think we're confusing EPUB Test and EPUBCheck
Bill Kasdorf: you're right, I was thinking about Test
Tzviya Siegman: I'm offering to be the point person for EPUBcheck, but I can do none of the programming
Rick Johnson: I think that's a great benefit, and I appreciate you taking that on
… and helping figure out what needs to be done
Matt Garrish: back to the CG question
… one thing we failed in the past was listing all the changes between spec versions
… I think that's something we should do going forward
… it would be great to have a clear list of spec changes to work from
… the CG can help with that
Avneesh Singh: I want to clarify some things
… EPUBCheck is not ready for open collaboration
… it has been maintained by a small group
… if you want it to be an open source process
… first the software needs to be prepared for that
… needs to be modular
… the second thing is the business of running the project
… the people who have worked on 3.1 are experts on formats/standards
… but most are not hard core coders, and are not involved in open source
… only readium is like that
… maybe we need someone who's familiar with open=source development
George Kerscher: I agree with Avneesh and delighted that Tzviya will be the point
… the 3.1 support, digging into spaghetti code, and getting an update that will help us right now, is good
… but for future, companies need to assign programmers to this, and commit time
… the other approach is to put money into hiring someone
… but we need nose-down programmers where this is part of their job description
… it worked with Romain, but DAISY had to pull him off
Rick Johnson: clarifying Q
… there's only a small group of people who really understand the code?
… and those people are working on the a11y checker?
… and this a11y checker will have dependencies on EPUBCheck?
… is that accurate?
George Kerscher: we always thought epubcheck would run first
Rick Johnson: if the a11y checker is expecting 3.1, and is expecting epubcheck to run first, there's a problem here
… there seems to be a dependency on these people who are tied up
Avneesh Singh: epub a11y spec is not tied to 3.1
… we have a dependency, but its not do or die
… for this year, Romain is fully tied up with a11y checker
… we can not have epubcheck depend on one person anyway
Tzviya Siegman: I think we're getting into too many details
… I think we need a task force, and want Romain and TObias to intend
… we need publishers to help, but we also need more information on what actual tasks are required
Bill Kasdorf: see why we needed Tzviya as point person? :-)
Tzviya Siegman: this is not a typical open-source project; it's large
Avneesh Singh: Romain can help with consulting
Tzviya Siegman: I'll work with Dave and Rachel to get this going in the CG, and get a task force
… and break into modular tasks
Rick Johnson: thanks
Ric Wright: I do have experience managing open source, and it's very very difficult
… it's harder than managing an internal team
… I do know java, and was one of the original authors of epubcheck
… so I could help the task force
… and even work on epubcheck itself, but would need to be paid
Rick Johnson: thanks everyone
… two additions to the agenda
Bill Kasdorf: I did agree to co-chair the program committe for the publishing summit in November
… We need another co-chair
… one of my priorities is related to the other topic, making sure it's not just books
… I'd like to turn it over to you for a timeline, but there's a june 12 deadline for some things
… perhaps we could have a call this week before I go to SSP
Bill McCoy: this 1.5 day event co located with tpac is the big opportunity for making sure publishing @ w3c is not about just specs
Tzviya Siegman: https://www.w3.org/2017/11/TPAC/Overview.html - note the warning about dreamforce
Bill McCoy: but about building a community
… galvanizing the community, inspired by BiB, TOC, etc
… we will open registration for TPAC week on June 12
… so we need to be open for registration on June 12
… we won't have a detailed program then
… to Bill K's point, we need to validate and refine the theme
… and turn that into a program outline
… but we won't have specific sessions and speakers
… so the work of the committee is a few calls etc before june 12, then some work in june/july to finalize program
… then there's logistics, etc, which I and w3c can do
… Bill_Kasdorf, do you want to supplement that?
Bill Kasdorf: is the program committee involved in recruiting sponsors?
Bill McCoy: from my POV, the primary responsibility is the program, not finding sponsors
Bill Kasdorf: this is a short-term commitment to shaping the program, finalizing the program and title for June 12, then shaping the program by end of July
Bill McCoy: correct. There's already a start on the wiki
Rick Johnson: Question: how should volunteers contact you?
Bill Kasdorf: email me and sign up on the wiki
Bill McCoy: given your schedule, we should have a call thursday or friday
Bill Kasdorf: could you do a doodle poll to whoever's interested today?
George Kerscher: bye
Rick Johnson: we had one other agenda item, on reaching out to non-publishers, but I'll add that to the agenda for the next meeting
… this is our meeting; if there are things you want to discuss let me know.
Rick Johnson: thanks everyone! we're done.