Internationalization Working Group Teleconference -- 23 Mar 2017

<scribe> ScribeNick: addison

Agenda

Action Items

action-596?

<trackbot> action-596 -- Addison Phillips to Adjust the agenda to reflect use of uk time for meeting and update links to dateandtime -- due 2017-03-16 -- OPEN

<trackbot> http://www.w3.org/International/track/actions/596

to be completed for this coming week

action-600?

<trackbot> action-600 -- Addison Phillips to Prepare and submit encoding refresh -- due 2017-03-16 -- OPEN

<trackbot> http://www.w3.org/International/track/actions/600

addison to ping anne for quiet time

<scribe> ACTION: richard: move the encoding tracking table to github so addison can update it [recorded in http://www.w3.org/2017/03/23-i18n-minutes.html#action01]

<trackbot> Created ACTION-604 - Move the encoding tracking table to github so addison can update it [on Richard Ishida - due 2017-03-30].

https://github.com/aphillips/encoding

action-601?

<trackbot> action-601 -- Richard Ishida to Check comment status for uievents specs -- due 2017-03-23 -- OPEN

<trackbot> http://www.w3.org/International/track/actions/601

close action-601

<trackbot> Closed action-601.

richard: did that, close one, some still need response

action-602?

<trackbot> action-602 -- Richard Ishida to Make shacl comments in felix's email into issues and/or commenting on sandro's comments and labeling them -- due 2017-03-23 -- OPEN

<trackbot> http://www.w3.org/International/track/actions/602

richard: done

close action-602

<trackbot> Closed action-602.

addison: they went for CR today so check comment status

action-603?

<trackbot> action-603 -- Richard Ishida to Respond to payment request api comment to make sure that direction only applies to one string at a time or to override for a specific string in a collection -- due 2017-03-23 -- OPEN

<trackbot> http://www.w3.org/International/track/actions/603

close action-603

<trackbot> Closed action-603.

Info Share

fantasai: looking for info on text decoration issues

richard: point to list of things you need?

<fantasai> https://lists.w3.org/Archives/Public/www-international/2016OctDec/0145.html

fantasai: some need i18n input

<fantasai> https://drafts.csswg.org/css-text-decor-3/issues-cr-2013

<fantasai> says '=i18n=' needs input from i18n

fantasai: anything in the above link
... CSS taking writing modes to PR
... in month or so

Radar

https://github.com/w3c/i18n-activity/projects/1

<scribe> ACTION: richard: checks status of svg2 comments [recorded in http://www.w3.org/2017/03/23-i18n-minutes.html#action02]

<trackbot> Created ACTION-605 - Checks status of svg2 comments [on Richard Ishida - due 2017-03-30].

http://w3c.github.io/i18n-activity/projects/

<r12a> https://github.com/w3c/i18n-drafts/projects/1

richard: updated the article pipeline
... last friday gave a presentation in tokyo
... bridging gap between jlreq and implementation
... seems to have gone well, some initiatives in place to work on closing gaps
... talk not yet available on web in english, but is available in chinese

<r12a> http://www.typeisbeautiful.com/2017/03/11613/

richard: attendee translated to chinese and put on his blog at above link
... what I said was
... recent member sub is useful to a point
... needs maintain
... needs more detail and prioritization
... so have a roadmap, that'll help css
... do it at a lower level than recent sub
... which was "vertical works pretty well"
... jp community should organize so can pull together and and show pain points
... interest in doing a course
... on bidi
... very early heads up
... did a bunch of requirements to specdev
... vertical text and ruby in more detail
... bit about text decoration
... largely from ttml

writing modes

richard: summary of chinese experts discuss
... and on 3 issues from ttml

<r12a> For subtitling, no-one has seen vertical-lr, and very little vertical-rl in chinese too.

<r12a> Straight or wavy lines alongside the text are mostly used for things such as book-title or name highlighting, rather than for emphasis like in Latin-script text underlines. Lines alongside vertical text should appear on the opposite side of items such as emphasis marks, ruby, etc (avoids interference), but most felt that lines (straight or wavy) always appear to the left of the vertical line.

<r12a> Emphasis marks (eg. dots) by default appear on the left side, but could be right.

<r12a> Bopomofo ruby always appears on the right side, and pinyin ruby or hanzi ruby normally appear on the right too.

richard: summary of discussion above

vertical-lr is mainly exceptional

richard: mostly used for book titles & name highlighting
... jp: right (before) side
... in left-to-right lines
... one person disagreed or wasn't sure
... difference between jp and zh
... mongolian on right hand side in vertical and below in horizontal

fantasai: have info on jp and ko in spec now but need mongolian

richard: thinking of sending issues

<r12a> https://github.com/w3c/type-samples/issues/55

richard: here's an example from type samples

<r12a> https://github.com/w3c/type-samples/issues/56

<fantasai> ohhhhhh, you have a database of type samples????

<fantasai> I totally have stuff for you

richard: here's an example so-called underlining

<fantasai> <3 <3 <3 <3

<r12a> https://w3c.github.io/type-samples/

richard: please to add samples
... ttml doesn't have a sideways-rl/lr for writing mode
... expected text-orientation would do that
... explained how css changed
... really useful for horizontal scripts, like start and end, line breaks etc
... should consider it
... instead of rotating the glyphs, rotate the box

fantasai: better ergonomics for the author: don't have to set two properties
... reduced the number of combinations
... that weren't important

richard: in ttml it's not handled properly

<fantasai> correctness wasn't an issue in original Writing Modes; it *was* in XSL:FO, though

<fantasai> TTML must've been copying XSL:FO model

richard: in tblr used default positions
... and in mongolian would be wrong
... don't think chinese would be right to position ruby et al
... wasn't able to convince them
... mongolians may feel unhappy
... and mongolian is prime user of tblr
... they are putting in a note
... don't want to change default
... could introduce over/under
... maybe in ttml-next
... change default for tblr specifically
... want to know about sideways-*
... at risk might not get impls
... could be deferred to level 4

fantasai: chrome planning to work on it
... in next year or two??
... jp wants writing modes asap
... defer sideways into a level 4
... but chrome would not defer

richard: arabic layout folks want for arabic

fantasai: as soon as implemented go to rec

richard: preceded css, just a timing of introduction

fantasai: will publish forked off level 3 without that section
... some process question about can we go direct to CR
... but looking to get sideways-* asap to cr

richard: no plans to change it
... way forward for ttml

fantasai: will publish simultaneous

Accept WHATWG change to restrictions on bidi

richard: whatwg stripped out checking for unpaired unicode bidi controls
... because things not checking correctly
... to complicated to understand
... actioned to ask aharon and mati
... "it's complicated" was reply
... travis is waiting on us
... so suspect say "okay"

addison: what's the side effect?

fantasai: only changes conformance checkers really
... concerns about user-generated data are solved by isolation

(no objections)

<scribe> ACTION: richard: tell travis to go ahead removing bidi controls text from html5 and whatwg [recorded in http://www.w3.org/2017/03/23-i18n-minutes.html#action03]

<trackbot> Created ACTION-606 - Tell travis to go ahead removing bidi controls text from html5 and whatwg [on Richard Ishida - due 2017-03-30].

<fantasai> r12a: http://fantasai.inkedblade.net/style/scans/