– DRAFT –
Internationalization Working Group Teleconference

andy: hello, I'm a new member,
… worked on a11y.
… Am an Invited Expert.

Agenda Review

Action Items

<addison> https://github.com/w3c/i18n-actions/issues

<addison> #190

<gb> Action 190 add a comment to i18n issue 2039's parent to indicate interest and call for early review/engagement (on aphillips) due 2025-10-23

<addison> #189

<gb> Action 189 follow up with shane carr about ecma repo automation (on aphillips) due 2025-10-23

<addison> #187

<gb> Action 187 ping GOOG again for participation (on aphillips) due 2025-10-16

addison: I followed up with Google folks
… Got conflicting feedback
… Some people interested, others too busy.
… Hoping for a representaive soon

<addison> #185

<gb> Action 185 follow up with PLH on automating TC39 repos (on aphillips) due 2025-10-02

<addison> #135

<gb> Action 135 follow up on XR issue 1393 about locale in session (on aphillips) due 2024-10-17

<addison> #33

<gb> Action 33 Close issues marked `close?` or bring to WG for further review (on aphillips)

<addison> #7

<gb> Action 7 Remind shepherds to tend to their awaiting comment resolutions (Evergreen) (on aphillips, xfq, himorin, r12a, bert-github) due 18 Jul 2023

<addison> #4

<gb> Action 4 Work with respec and bikeshed to provide the character markup template as easy-to-use markup (on aphillips) due 27 Jul 2023

Info Share

JcK: Trying to fix an RFC, about NFC

eemeli: I filed two TPAC break out sessions
… Introducing HTML <amount> element
… And another for talking about Henri Sivonen's English standard measures as a locale
… Maybe something to talk about in this group, too.

addison: Sounds interesting.

<eemeli> w3c/tpac2025-breakouts#71

<gb> Issue 71 Defining a new Locale: Standard Measures with U.S. English (by eemeli) [session]

addison: Did we talk about this earlier?

eemeli: See ^^ for the session about the locale
… I want to go to CLDR and show that there is interest
… If we can show this is something generic, they will likely be more interested.

Review RADAR Review

<addison> https://github.com/orgs/w3c/projects/91/views/1

addison: New incoming request
… Sustainability Guidelines
… Any takers?

xfq: It looks like it is long...

eemeli: 195 pages in print view...

<r12a> https://www.w3.org/TR/web-sustainability-guidelines/#use-optimized-and-appropriate-web-typography

addison: I'll write back and ask for more time. One month with TPAC in it will not be enough.

andy3: Mayeb also ask for a lower reading level, to make it easier to read.

<jsahleen> TPAC dates?

<Bert> TPAC is 10-14 Nov

<xfq> 10–14 November, same week as UTW

andy3: Will be easier for everybody and also for us.

addison: We can ask that as an editorial comment.

r12a: How did you determine the reading level?

andy3: Just from experience and seeing the first few sentences.

addison: They do have a plain-language summary, so maybe they are already aware of the issue.
… Do we want to split the work and have some people take separate sections?

<jsahleen> I can take a look if we have more time

xfq: I can take a look and I encourage others to take a look, too

addison: I'll make xfq the shepherd, but not the only reviewer

jsahleen: I can take a part

ACTION: addison: ask for more time on web sustainability guidelines

<gb> Created action #191

andy3: I can look at user-experience aspects

Bert: Halfway through color adjustment. No issues so far.

Pending Issue Review

andy3: Question about the Sustainability Guidelines: Any specific reason i18n was asked for a review?

addison: It is a ‘horizontal review’, we review everything as matter of principle.
… So bring i18n issues here, but a11y issues in the a11y review.

<atsushi_> https://www.w3.org/guide/documentreview/#how-to-get-horizontal-review

andy3: This group also considers accessibility to global audiences?
… Is it just about being able to localize?

<xfq> See also https://www.w3.org/International/questions/qa-i18n#answer

<r12a> and https://www.w3.org/International/articlelist#other

addison: Any comments on a spec during a review are valid, but we focus on culture, language. Not really on cognitive level, assistive tech, etc. Even if the border between them can be blurry.
… Security is another aspect, typically not i18n.

andy3: Simple language can help non-native speakers.

DOM Localization

<addison> https://lists.w3.org/Archives/Public/public-i18n-core/2025OctDec/0028.html

eemeli: I sent email and submitted a TPAC breakout.

<addison> https://github.com/mozilla/explainers/blob/main/dom-localization.md

<addison> w3c/tpac2025-breakouts#67

<gb> Issue 67 DOM Localization (by eemeli) [session]

eemeli: People probably haven't had time to look at it yet.
… Explainer for what DOM localization is.
… Targeted of course at WHAT WG and HTML.
… It is about i18n and l10n, but the place it will be added is the DOM & HTML.

addison: We can see if we have time today. We will talk about TPAC on the agenda.
… This group is not chartered to work on a specification.

<xfq> https://www.w3.org/guide/incubation.html

addison: If we want to write something ourselves, we'd need to think about a change to the charter,

r12a: If this belongs in the HTML spec, that doens't preclude us working on it here.

<xfq> +1 to r12a

r12a: We've done that before, e.g., ruby (for CSS) and bdi (for HTML)

<jsahleen> +1 to r12a

r12a: Would be useful for us, too. We'd learn more about DOM localization.

eemeli: The HTML parts are one side, which need to go into HTML spec. But there is also a resource format side, maybe comparable to CSS, and separate from HTML.

addison: Maybe we can schedule a time-limited intro now?

eemeli: I can present something now.

10 minutes introduction

eemeli: I want an external document to be able to define localization, independent of HTML, similar to how CSS defines style for HTML.
… We want something easy to use, sufficiently (but not extemely) powerful.
… Related to Message Format.

<jsahleen> Many development organizations handle localization as part of the build/bundling process. How would this work in that context?

eemeli: In <head> you can add a link to a style sheet. Should also have a link to a file for a locale.

<xfq> If a page is localized in 100+ languages, do I need to add more than 100 declarations in the head element?

eemeli: So you get an HTML that can get a message with a certain identifier and localize it.

<addison> not necessarily. think of it as "resource bundle for the web"

eemeli: Placeholders, maybe with simple tetx replacment, or more complex, like number formatting.
… Localization attributes.
… Everythign should also be accessible from JavaScript, for dynamic effects and integration with JS frameworks.
… This needs a file format.
… I think it needs to be a new format.
… Text format, modifiable by humans.
… So much if the work is defining that format.
… Building on Unicode and Message Format.

<xfq> I wonder if it is possible to do more fine-grained localization, such as localization of a web component (HTML custom element).

<r12a> andy, best to join the queue by typing q+ here (rather than zoom hand up)

eemeli: Relies on understandings we gained at Mozilla for localizong Firefox, Thunderbird, etc,
… Experience with Fluent.
… Learned about pitfalls.
… Many open questions.
… Such as how to define the right resource to load, based on language.
… Or if a particular langauge diens't have a wanted message, but another one does.

addison: That means there is some flexibility in the system?

<andy3> It would be helpful to have concrete visual examples: the processes of various forms of localization - how DOM localization would change those current processes

eemeli: Concerns about which reosurces to load, over the network, possibly at high cost,
… Interfaces for fallback.

r12a: Interesting. It is more in l10n than we usually do. We did work on ITS before.

eemeli: How active is ITS?

addison: Not active at the moment.

r12a: Just be aware of it.
… You need content negotiation, storage of information, and bringing it into HTML.
… I thought bringing stuff into HTML would be easy, but others seem to thing differently. So I find myself need to use JavaScript.
… We have, e.g., a system to bring in boilerplate text into articles.
… What kinds of text are you targeting? Is it like our boilerplate text? What is the scope.

<jsahleen> interface text

eemeli: It is the sort of tetx that Unicode Message Fornat is also aimed at.
… But not opposed to other content, such as you described.
… For building pages out of resources there are tools and localization is usually not the blocking factor.
… Focus would be on shorter pieces of content, but longer is not excluded.
… Does it need new HTML elements?

<jsahleen> Security issues?

<xfq> Like the "Learn Find Ask Follow Participate About" in https://www.w3.org/International/articlelist ?

eemeli: E.g., if all translations need to have the same number of paras, that would be a restriction.

r12a: When you are not dealing with message formt-like things, it gets trickey. E.g., in the frenche translationof our articles, things are sometimes presented in a different order.
… Provenance aspects: did a translation come from a machine?

andy3: A visual of the processes involved would help to understand it, a kind of workflow.

eemeli: I'll have something prepared at TPAC.

andy3: Can you share them also for those not at TPAC?

addison: Some demo will help people understand.

addison: TPAC planning:
… I'll prepare the wiki.
… I'll not be on site, but Fuqiao will.
… I have requested enough AV setup for dial-in participation.
… But time zone will eb a problem.
… If you have things for our ftf, send them to our public list and I'll include them

Discussion about best time for Japan and West Coast.

r12a: This telcon stays constant in Britain.

AOB?