[minutes] Internationalization telecon 2022-02-17


   – DRAFT –
            Internationalization Working Group Teleconference

17 February 2022

    [2]Agenda. [3]IRC log.

       [2] https://www.w3.org/events/meetings/5e4758ba-3e80-4e71-b333-5f2c4fd95abd/20220217T150000
       [3] https://www.w3.org/2022/02/17-i18n-irc


           addison, Atsushi, Bert, David, Erik, Fuqiao, JcK,


           Addison Phillips

           addison, xfq


     1. [4]Agenda
     2. [5]Action Items
     3. [6]Info Share
     4. [7]Radar
     5. [8]In-person for TPAC?
     6. [9]RTL named group support in URLPattern
     7. [10]WOT Thing Description review
     8. [11]AOB?
     9. [12]Summary of action items

Meeting minutes

    <addison> trackbot, prepare teleconference

    <addison> scribenick xfq


    addison: was it an improvement over last week's agenda?

    Bert: I'll make sure agendabot recognizes your format

    addison: maybe I shouldn't use a bullet list

   Action Items

    agendabot: does anybody have anything to add to the agenda?

    JcK: can we have plain text @@ HTML?

    addison: I only own the part between the word "Agenda" and
    "Joining Instructions"

    <addison> [13]https://www.w3.org/International/track/actions/

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

    r12a: you can have spacing between "BE PREPARED", "REMINDER",

    <JcK> That was not a request for plain text -- I'm going to
    read it that way regardless -- just for understanding of the
    implications to that format.

    addison: 1094 is dependant on the explainer
    … 1103 is also dependant on the explainer

    trackbot, close ACTION-1118

    <trackbot> Closed ACTION-1118.

    trackbot, close ACTION-1119

    <trackbot> Closed ACTION-1119.

   Info Share

    trackbot, close ACTION-1117

    <trackbot> Closed ACTION-1117.

    trackbot, open ACTION-1118

    <trackbot> Sorry, xfq, I don't understand 'trackbot, open
    ACTION-1118'. Please refer to <[14]http://www.w3.org/2005/06/
    tracker/irc> for help.

      [14] http://www.w3.org/2005/06/tracker/irc

    <addison> reopen 1118


      [15] https://github.com/whatwg/html/issues/4986#issuecomment-1024344940

    xfq: came up in my vacation
    … html folks made proposal on autocomplete
    … agenda+ in this or future meeting?

    addison: if memory serves they had some random stuff
    … address format notoriously complicated

    ACTION: addison: remind group to follow up on html#4986

    <trackbot> Created ACTION-1120 - Remind group to follow up on
    html#4986 [on Addison Phillips - due 2022-02-24].

    r12a: Do you think there is any problem, xfq?

    xfq: i think we discussed this before


    xfq: I would like to check with you all if there is a problem.
    I personally didn't find any problems.

    r12a: i'll start editing the editorial guidelines
    … I'm going to start (again) to create a new set of -2022 files
    … that applies to new articles
    … we should be able to fit it to articles we update as well

    addison: thank you


      [16] https://w3c.github.io/i18n-drafts/questions/qa-floating-times.en

    addison: a coworker working on Temporal saw one of our floating
    times articles and found it useful
    … and linked to it
    … I should really update the working with time zones note
    … because there are changes in the world since I wrote it a lot
    of years ago


      [17] https://github.com/w3c/i18n-request/projects/1

    addison: WoT TD, we have an agenda item for it
    … SPC, we have a couple of pending issues we can review today

   In-person for TPAC?

    addison: there's a questionnaire for TPAC
    … in September in Vancouver

    David: my feeling is hybrid is the best option

    r12a: for me it's a bit early to know what to do
    … the flight for me is like 12 hours long

    David: people in some countries may not even be allowed into
    Canada in direct flights

    addison: I don't know what's gonna happen
    … I will reserve some space

    Bert: if it's possible I'd like to go there in person

   RTL named group support in URLPattern


      [18] https://github.com/WICG/urlpattern/issues/43

    addison: when they filed this request there was no spec
    … in the interim year there's a spec
    … I don't think we have reviewed this
    … it's in a CG

    addison: effectively they're working with URLs
    … basically regular expressions
    … syntax for named groups
    … in a regular expression you can assign a name to an
    … as a way of querying with the content
    … rather than replicating the expression

    David: rather than "it's the third collection" you can give it
    a name

    addison: a full regular expression syntax for working with URLs
    … e.g., "give me the fragment"
    … it's possible to screw things up
    … in a text editor because of bidi effects

    r12a: logically speaking it's the same as in English

    addison: it's "code point biased"

    r12a: not because of it won't work
    … just because you won't be able to read it

    addison: usability problem
    … we should probably review this thing before too long

    ACTION: addison: add urlpattern to review (non-urgent)

    <trackbot> Created ACTION-1121 - Add urlpattern to review
    (non-urgent) [on Addison Phillips - due 2022-02-24].

    JcK: not everybody knows what TC 39 is

   WOT Thing Description review


      [19] https://github.com/w3c/i18n-activity/issues?q=is:issue+is:open+label:pending+label:s:wot-thing-description

    addison: i've done the review for WoT TD
    … have some pending issues


      [20] https://github.com/w3c/i18n-activity/issues/1465

    r12a: they say if a language can be written in a few scripts
    … what happens when there's no script subtag
    … who's having the MUSTard?
    … who would suppy that subtag?

    addison: whoever supply the WoT TD
    … @@
    … you can have a list of languages you know in a language map

    r12a: how would they know Azeri can be written in more than one

    addison: presumably if they are localizing their device into
    these languages they would know
    … but we cannot guarantee
    … my first comment here is that we prefer not to use heuristic
    … we prefer to use metadata
    … and I could make a second comment, says your heuristic is
    … they didn't pick up some of the things we did with JSON-LD
    i18n namespace and so on


      [21] https://github.com/w3c/i18n-activity/issues/1467

    addison: and you should be aware that elsewhere they say
    there's no metadata, and they cite string-meta


      [22] https://w3c.github.io/string-meta/#script_subtag

    <r12a> [23]https://www.w3.org/International/articles/

      [23] https://www.w3.org/International/articles/lang-bidi-use-cases/

    r12a: we actually do recommend heuristic, but only in some

    <r12a> [24]https://w3c.github.io/

      [24] https://w3c.github.io/string-meta/#bp_use_heuristics_1

    r12a: ^ this is what we're recommending in string-meta
    … in the top of the file you can say the whole file is LTR
    … but it may not be available

    r12a: each string should be able to be accompanied by metadata,
    but we don't say that you have to supply metadata

    addison: it's about capability


      [25] https://w3c.github.io/string-meta/#bp_inferring_from_language

    addison: ^ this somewhat prevents that

    r12a: this says "don't do these guys are doing, unless it's a
    last resort"
    … so the question is, what is the last resort?

    addison: I have a bunch of things that are not direction


      [26] https://github.com/w3c/i18n-activity/issues/1466

    addison: like something replaces Unicode normalization
    … #1466 should be easier
    … they referred to "well-formed" language tags, but they don't
    bother to define "well-formed"

    <atsushi_> +1

    addison: we encourage specs require well-formedness and
    encourage valid


      [27] https://github.com/w3c/i18n-activity/issues/1468

    addison: ^ this is about human-readable metadata
    … there's no way to ascertain languages or locales that are

    r12a: isn't that normal?

    addison: @@
    … a map can be flattened, you don't need to send fr-FR, fr-BE,
    fr-CA, fr-CH, and you can only send fr

    Bert: i wonder what http say about language priority


      [28] https://github.com/w3c/i18n-activity/issues/1469

    addison: I'll work on this before the next meeting
    … Accept-Language is just an example
    … I don't know how else you would negotiate language with a

    Bert: I think I agree with you comment


      [29] https://github.com/w3c/i18n-activity/issues/1470

    <atsushi_> +1 strongly

    addison: somewhat meta thing, but it's really common for
    developers to write English as enum value
    … write in uppercase, camel case, snake case, or one of the
    programmer cases

    xfq: +1

    JcK: it's the right thing to do


      [30] https://github.com/w3c/i18n-activity/issues/1471

    addison: it's probably the authors know better, it's just

    addison: we famously have RFC3987 to deal with @@
    … whatwg has URL

    r12a: looks good to me


    close action-1114

    <trackbot> Closed action-1114.

    close action-1113

    <trackbot> Closed action-1113.

    close action-1104

    <trackbot> Closed action-1104.

    close action-1116

    <trackbot> Closed action-1116.

Summary of action items

     1. [31]addison: remind group to follow up on html#4986
     2. [32]addison: add urlpattern to review (non-urgent)

Received on Thursday, 3 March 2022 17:18:11 UTC