[Ietf-languages] Fwd: Recommendation not to register variant subtags of the form 0nnn

[John Cowan <[e-mail]>]

---------- Forwarded message ---------
From: John Cowan <[e-mail]>
Date: Wed, Jul 29, 2020 at 12:20 PM
Subject: Re: [Ietf-languages] Recommendation not to register variant
subtags of the form 0nnn
To: Phillips, Addison <[e-mail]>




On Wed, Jul 29, 2020 at 11:48 AM Phillips, Addison <[e-mail]>
wrote:

That would not happen with a 4-digit variant starting with a digit: they
> would always come at the end of a language tag.
>

It's true there is no actual ambiguity: 004 is always going to be a region
and 0004 is always going to be a variant, and so "qq-004" is Cuecue as used
in Africa, whereas "qq-0004" is some variant of Cuecue, not necessarily
date-based.  But human beings could easily make a mistake, because we are
used to the idea that leading zeros have no semantic meaning in a number.


> I tend to agree that we should avoid confusables and thus should be
> careful not to register such subtags just generally. But never is a long
> time and I could see someone (attempting to) register e.g. Gregorian year
> based subtags for well-attested years pre-1000, e.g. “ar-0661” (or maybe
> “ar-0072” ;-))
>

I'd go with ar-ce661 and ar-ce72 in those cases.  When speaking or writing,
I would always use a phrase like "the year 72" even when I wouldn't bother
with "the year 1958".


>  As I said, I agree in general principle, but wouldn’t we discuss it in
> specific at the time?
>

We would; this is just nemawashi.



John Cowan          http://vrici.lojban.org/~cowan        [e-mail]
        "Not to know The Smiths is not to know K.X.U."  --K.X.U.