Date: 2017-06-06
See also the Agenda and the IRC Log
Present: Wolfgang Schindler, Avneesh Singh, Ivan Herman, Matt Garrish, Dave Cramer, Tzviya Siegman, Bill McCoy, Luc Audrain, Brian O'Leary, Julian Calderazi, Junichi Yoishii, Paul Belfanti, Garth Conboy, Bill Kasdorf, Mateus Teixeira, Dan Sanicola, Benjamin Young, Leslie Hulse, Murata Makoto, Takeshi Kanai, Junko Kamata, Ric Wright, Michaela Philipzen
Regrets: Rick Johnson, Karen Myers, Rachel Comerford, George Kerscher, Cristina Mussinelli
Guests:
Chair: Paul Belfanti
Scribe(s): Garth Conboy, Ivan Herman
Murata Makoto: I cannot join WebEx.
Tzviya Siegman: https://www.w3.org/2017/05/23-pbg-minutes.html
: Getting there (on the charter)
… 2nd round of “is this okay” ends tonight. Some objections withdrawn. W3M to meet tomorrow. Likely in good-ish shape.
… admin steps todo; call for particpation likely to go out next week.
… June meeting — under auspices of IG & BG — with WG coming in late.
… June meeting: https://www.w3.org/dpub/IG/wiki/June_2017_F2F
Bill McCoy: Wendy will drive charter discussion at W3M. June meeting to be joint between IG & BG… but still kick-off of WG.
… Broader communication coming — WG and Publishing@W3C summit
… (next week)
Brian O'Leary: to summarize document & proposal
… Current situation with epubtest — BISG, Daisy, IDPF (in the pre-W3C days)
… Focused on test with EPUB 3
… BISG endorsed EPUB 3
… Help Publishing Community to know what works where
… 1. Defend open standard
… 2. Influence device and RS to get with the standard
… 3. Foster better testing and volenteers
… Must have plan for what happens with test results after said is generated
… EPUB 3 support grid originally published in mid-2012
… resources have become scarce; grid is stale
BillK: mindful or two uses of grid: showing extent of RS support (fallen into disuse) & resourse for RS developers to self-test and being their systems into compliance
… Is it still useful to publish results? — kinda negative spin
Tzviya Siegman: if it’s a tool for developers, we need to worry less about messaging; needs to be more automated.
Dave Cramer: +1 to Tzviya; not a showcase of EPUB 3 features; we need tests; epubcheck has a test suite too; can that be integrated?
… How much of Spec is really tested?
… Do RS vendors have internal tests that could be donated
Avneesh Singh: initial focus was to show (and push toward) new standard in 2012
… same could be true of EPUB 4 coming in the future
… main focus should be on increasing testing (running of tests)
Bill McCoy: test results are outdated (not latest reading systems) but that is different than saying the test suite is outdated (it is not updated to EPUB 3.1 but I believe it is not out of date for EPUB 3.0/3.0.1)
Avneesh Singh: Grouping of features seems to have worked well for A11Y tests
Tzviya Siegman: the tests are also completely manual
Takeshi Kanai: Maybe works for US publishers, but less so for, e.g., Japan
Murata Makoto: q
Takeshi Kanai: not enough testing of vertical writing, e.g.
BillK: previous comments were on mainstream tests, not A11Y — the latter working well and needing to be published.
… don’t want any changes to undermine A11Y testing
… Test suite is not out of date, it’s the tested RS’s that are
Murata Makoto: There is a Japanese test suite, should this be integrated? But may be too focused on locale-relevant issues.
Paul Belfanti: need to solid foundation before adding too many locale specifics.
Ivan Herman: Need to find a way to work with W3C web testing (HTML, CSS, et al)
Dave Cramer: http://testthewebforward.org
Tzviya Siegman: here is what the DPUB-ARIA test suite looks like https://w3c.github.io/test-results/dpub-aria/all.html
Brian O'Leary: BISG could support more limited testing/tests (automation would help), but why are we doing that?
… BISG focus on US could be an issue, and now eBooks in 170+ countries.
Murata Makoto: Japanese testsuite (denosho lab) http://densholab.jp/lab/RScheck/list.html
Luc Audrain: Test suite used regularly at Hachette to test RS’s — both features and regressions
Murata Makoto: Test result of reading systems widely used in Japan http://densholab.jp/lab/RScheck/result.html
Luc Audrain: +1 to W3C testing
Tzviya Siegman: testing will be requied for going forward in the W3C
Murata Makoto: I am not sure if the Japanese test is automated.
Murata Makoto: I think that human testers to examine visual output.
Ric Wright: Very costly to build automated test suites (invalidated by underlying Web changes), tend to be atomic tests, but real world harder to test
Dave Cramer: Should have conversation with Jeffery ??? we Web testing; Dave could contact
Murata Makoto: Follow BiB/i links in http://densholab.jp/lab/RScheck/list.html for rendering in browsers.
Tzviya Siegman: and shane mccarron
Dave Cramer: s/Jeffrey ???/Geoffrey Sneddon/
Brian O'Leary: agreement that current manual approach won’t really work; need to understand the “why” to know which way to move; pushing the info shouldn’t be end of process.
Paul Belfanti: Next steps?
Brian O'Leary: Need sub-group to come up specific proposal (Dave, Brian, Tzviya, …)
Luc Audrain: In contact with EU folks
… re European A11Y
… will bring back results to this group
… Need to be more clear on what’s going on with ISO process
… To be aware of A11Y for EPUB 1.0 or 1.1 standardizations
Murata Makoto: I have already posted a sequence of emails about this. They are linked from the CG(?) web page.
Luc Audrain: Need to make sure the EU does something non-aligned
Murata Makoto: What else is needed?
Murata Makoto: EPUB 3.0.1 revision to revise to TS to get 3.0.1
… Drafts coming
… 3.0.1 revision is on track
… Need to path to A11Y 1.0 or 1.1 standardization
… Should be new work item proposal
… The resolution will be publicly visible when submitted
Luc Audrain: we should synchronize between the ISO work and what the EU is expecting
Murata Makoto: I am not sure
Luc Audrain: I will try to collect information and share those
Avneesh Singh: makoto you are going forward on the 19th; is there a decision form this group?
Murata Makoto: I need a formal endorsement
… I want to be sure that people are not against
Avneesh Singh: if this group cannot decide before the 19th, it is still all right
Murata Makoto: I have already informed a number of people; some countries are already aware
Avneesh Singh: iso was a high priority for us; are we discussing again whether we want to do it, or only the modalities?
Murata Makoto: the normal process is the most obvious option
… the faster one is not an option any more
… we have to submit a new work item
Avneesh Singh: is this group agreeing to go to iso?
Luc Audrain: +1
Avneesh Singh: if so, the only problem is how to execute it
Luc Audrain: I support it
Tzviya Siegman: billm has mentioned there are other work, I want to be sure they are coordinated
Murata Makoto: yes, dr. Cho will come to Japan
Tzviya Siegman: I am unaware of the iso process; this is all in coordination with the w3c process
Murata Makoto: I believe
… I am not sure of w3c process
… but this will not be a w3c official document, only a report of the bg or the cg
Luc Audrain: what is the expected timing for a new work item
Murata Makoto: i prepare a new work item, submit it to member bodies, then a ballot with 5 bodies approval
… then 5 experts who work to
… then the epub cg prepare a draft technical specification
… that is about 3 months
… after the completion of accessibility 1.1 will 7-8 months will be needed
… if we use 1.0 we can start right now, but then to revise it later
Luc Audrain: about a year as a whole
Avneesh Singh: we can not resolve today, but we could a dedicated call before th 19th
… we only need the blessing from this group
… a small group will work on the details
Paul Belfanti: a formal cfc will be issued