[minutes] Internationalization telecon 2017-08-31


text extract follows:


            Internationalization Working Group Teleconference

31 Aug 2017



    See also: [3]IRC log

       [3] http://www.w3.org/2017/08/31-i18n-irc


           Barry_Leiba, Addison, r12a, Fuqiao, David, andrewc,
           Bert, JcK


           Addison Phillips



      * [4]Topics
          1. [5]Agenda
          2. [6]Action Items
          3. [7]Info Share
          4. [8]RADAR and Active Work Review
          5. [9]Developments related to the gap-analysis matrix
          6. [10]WhatWG issue 2945 (hinting at translation for
             outgoing links)
          7. [11]AOB?
      * [12]Summary of Action Items
      * [13]Summary of Resolutions

    previous minutes:

      [14] https://www.w3.org/2017/08/24-i18n-minutes.html

    <scribe> ScribeNick: addison

    <xfq> [Fù qiáo]

    <barryleiba> Addison


Action Items


      [15] https://www.w3.org/International/track/actions/open


    <trackbot> action-618 -- Richard Ishida to Create a short
    document about making direction default recommendation to be
    'auto' and send to www-international@ for review -- due
    2017-08-25 -- OPEN


      [16] http://www.w3.org/International/track/actions/618


    <trackbot> action-645 -- Addison Phillips to Remind working
    group to read whatwg html issue 2945 for discussion next
    telecon -- due 2017-08-31 -- OPEN


      [17] http://www.w3.org/International/track/actions/645

    close action-645

    <trackbot> Closed action-645.


    <trackbot> action-646 -- Richard Ishida to Publish
    strings-and-bidi article -- due 2017-08-31 -- OPEN


      [18] http://www.w3.org/International/track/actions/646

    close action-646

    <trackbot> Closed action-646.

Info Share

RADAR and Active Work Review


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

    richard: worked on css align, very complex
    ... like a large math equation :-)
    ... maybe not much to worry about
    ... couple things, it's about aligning blocks inside another
    ... can do horizontally or vertically
    ... call the vertical alignment in horizontally laid text
    ... have properties like "align self" and "align content"
    ... or "jusitfy self" or "justify content"
    ... word 'justify' already confusing??
    ... and the section on baseline alignment is difficult to
    understand when differnet script baselines in use
    ... editors aware of the problem obviously, but not obvious to
    me how the baseline is chosen
    ... thinking of raising question


      [20] http://w3c.github.io/i18n-activity/reviews/

    schedule review of issues for next week


      [21] http://w3c.github.io/i18n-activity/projects/

    <scribe> ACTION: richard: send list of review items for next
    week (pendings and close? items) [recorded in

      [22] http://www.w3.org/2017/08/31-i18n-minutes.html#action01

    <trackbot> Created ACTION-647 - Send list of review items for
    next week (pendings and close? items) [on Richard Ishida - due

    richard: growing momentum to idea that mongolian handled
    differently than currently in unicode
    ... ad hoc committee
    ... not sure why not on mongolian list
    ... will suggest that
    ... key proposal is "graphetic" approach
    ... if type sound a/aa/oo they all look the same
    ... confusing when typing
    ... useful to computers in distinguishing words
    ... all sorts of complex rules, such as gender of word
    ... you have one character for every shape (not quite that
    ... type what you see

    <scribe> ACTION: richard: ask mongolian ad hoc group to use our
    list [recorded in

      [23] http://www.w3.org/2017/08/31-i18n-minutes.html#action02

    <trackbot> Created ACTION-648 - Ask mongolian ad hoc group to
    use our list [on Richard Ishida - due 2017-09-07].

Developments related to the gap-analysis matrix


      [24] http://w3c.github.io/i18n-discuss/notes/language-matrix

    richard: last week talked about above link (language matrix)
    ... graphical idea of where pain points are for world's
    ... have to be careful about what languages chosen
    ... list is a bunch in alpha order by script tag

    <r12a> [25]http://w3c.github.io/typography/gap-analysis/ethi

      [25] http://w3c.github.io/typography/gap-analysis/ethi

    richard: bit more flesh on bone for ethiopic/amharic
    ... only javanese last week
    ... ethiopic more developed, used ELReq
    ... more complete
    ... taking another step
    ... hebrew layout req and arabic layout req folks
    ... approach from this angle?
    ... send email to shervin


      [26] http://w3c.github.io/typography/gap-analysis/arab-ar-fa

    richard: click on arabic link
    ... probably add more pages for urdu, uighur. this covers ar/fa
    ... should help arabic layout folks zoom in on things that need
    ... idea is to use gap analysis to prompt to write requirements
    ... write down what does/doesn't work
    ... write tests to demonstrate
    ... and write reqs for things that don't work so know what to
    ... wanted to mention, since fuqiao is here
    ... try to reactivate chinese work
    ... maybe additional thing to consider there
    ... ask chinese experts "what's still missing"

    david: what does level represent?

    richard: level represents overall
    ... color says whether the language works or not on the web
    ... numbers are very kludgey
    ... added points for dark green, etc. and divided
    ... hokey, but gives some idea

    jck: two suggestions
    ... comment at bottom welcoming analysis of other languages

    richard: meant to do that and forgot

    jck: under 'latn' you have "serbo-croatian"
    ... and no longer politically correct

WhatWG issue 2945 (hinting at translation for outgoing links)


      [27] https://github.com/whatwg/html/issues/2945

    richard: the attribute as billed currently looks at the content
    and tries to decide what to do based on content
    ... but should reflect preference of the user
    ... that's significant
    ... that makes a "loud clang"
    ... if you're reading hindi with the page translated
    ... and the link says "translate to hindi"
    ... second page in english would hint english

    addison: don't see how page authors would use this

    (much unminuted discussion)

    David: another thing might be Referrer-Language

    richard: worth thinking about, "there's been a change in
    language here"

    andrewc: spec out how browsers handle

    david: think we conclude that bad to embed it

    richard: easier to change language preferences and see them
    ... and maybe some way to signal language change

    <xfq> +1 to david & richard

    <scribe> ACTION: addison: reply to whatwg 2945 translatehint
    issue with wg opinion [recorded in

      [28] http://www.w3.org/2017/08/31-i18n-minutes.html#action03

    <trackbot> Created ACTION-649 - Reply to whatwg 2945
    translatehint issue with wg opinion [on Addison Phillips - due


Summary of Action Items

    [NEW] ACTION: addison: reply to whatwg 2945 translatehint issue
    with wg opinion [recorded in
    [NEW] ACTION: richard: ask mongolian ad hoc group to use our
    list [recorded in
    [NEW] ACTION: richard: send list of review items for next week
    (pendings and close? items) [recorded in

      [29] http://www.w3.org/2017/08/31-i18n-minutes.html#action03
      [30] http://www.w3.org/2017/08/31-i18n-minutes.html#action02
      [31] http://www.w3.org/2017/08/31-i18n-minutes.html#action01

Summary of Resolutions

    [End of minutes]

Received on Thursday, 31 August 2017 16:50:18 UTC