Re: Validator doesn't know that pes is Persian

Hi Henri,

Henri Sivonen <hsivonen@hsivonen.fi>, 2025-05-23 14:10 +0900:
> Archived-At: <https://www.w3.org/mid/e4293d99-229e-41e5-afc6-e5862dbf539a@app.fastmail.com>
> 
> Do the three-letter codes really improve interop? CLDR seems to use fa-AF for Dari.

Personally, I don’t know — and I don’t have any strong opinion about it
either way. I just made the change in response to Richard having raised the
issue — and because in general it seems that for cases like this we’ve been
following a policy of deferring to people in the community known to have
more specific domain knowledge around internationalization particulars.

And I guess it’s also worth mentioning that what this change affects is
just the checker’s language-guessing feature. And the behavior of that
feature isn’t based on any specification — so it’s not possible to consult
any actual normative conformance requirements for the expected behavior.

  –Mike

> On Mon, May 12, 2025, at 10:47 AM, mike@w3.org wrote:
> > Hi Richard,
> >
> > r12a <ishida@w3.org>, 2025-05-10 14:01 +0900:
> >> Archived-At: <https://www.w3.org/mid/85c113a0-220e-c469-ea36-e046521f95b0@w3.org>
> >> 
> >> Showing results for
> >> https://w3c.github.io/i18n-drafts/questions/qa-lang-why.pes.html
> >> 
> >>    *Warning*: This document appears to be written in Persian but the
> >>    |html| start tag has |lang="pes"|. Consider using |lang="fa"| (or
> >>    variant) instead.
> >> 
> >> 'fa' is a macrolanguage. The document is written in Iranian Persian (rather
> >> than, say, Dari), and the more precise langtag is 'pes'.
> >
> > Thanks — fixed now:
> >
> > https://validator.w3.org/nu/?doc=https://w3c.github.io/i18n-drafts/questions/qa-lang-why.pes.html
> >
> >   –Mike

-- 
There is too much to say, and it goes in all directions.