Unifon script?

Doug Ewell doug at ewellic.org
Thu Oct 3 21:34:00 CEST 2013

Kent Karlsson <kent dot karlsson14 at telia dot com> wrote:

> Den 2013-10-02 20:42, skrev "Doug Ewell" <doug at ewellic dot org>:
>> Actually, you can't assume 'Latn' from "en-fonipa" either.
>> Suppress-Script identifies the *default* script for the language when
>> written normally, but the presence of a variant for a phonetic
>> alphabet completely overrides this default. You certainly could not
>> assume anything of this sort from, say, "ru-fonipa" or "ko-fonipa".
> While I'm sure that is formally correct, I see (in the registry)
> Type: variant
> Subtag: heploc
> Description: Hepburn romanization, Library of Congress method
> Added: 2009-10-01
> Deprecated: 2010-02-07
> Preferred-Value: alalc97
> Prefix: ja-Latn-hepburn
> Comments: Preferred tag is ja-Latn-alalc97
> The recommendation in that comment seems to imply that one should
> include a script subtag in the kinds of situations now under discussion
> (variant has different script than supress-script for the language or
> there is no supress-script for the language). Otherwise "ja-alalc97"
> would have been the recommended tag. Note that the entry for alalc97
> has no prefix field.

I think romanizations are a little different from phonetic alphabets
here. "ja-Latn-alalc97" really means:

1. Japanese.
2. Not only that, but Japanese in the Latin script.
3. Not only that, but using the 1997 ALA-LC romanization.

A phonetic alphabet doesn't exhibit quite the same hierarchy:

1. English (or Russian or whatever).
2. Not only that, but English in the Latin script.
3. Not only that, but in IPA (or Unifon).

IPA and Unifon are *based on* Latin, to be sure, but I don't see the
same straight-line relationship as the one that exists from Latin to a
romanization. Others may see it.

> Thus it may be better to use "ru-Latn-fonipa", "ko-Latn-fonipa" rather
> than just "ru-fonipa" or "ko-fonipa" (even though the latter are of
> course formally correct language tags).

But that wasn't really what I said above. I said you couldn't use the
Suppress-Script value for 'en' by itself to conclude that "en-fonipa"
denotes a Latin-based orthography, because the logic wouldn't hold for
"ru-fonipa" and similar. Obviously, writing "ru-Latn-fonipa" would
resolve this. But should users be encouraged to add extra subtags to
provide background information on the origins of a phonetic alphabet?

I agree with Addison that this sort of meta-information makes more sense
in a Comments field, or even in the "additional information" section of
the registration form. I went ahead and sent Michael the 'unifon'
paperwork for submitting to IANA, instead of delaying it further,
because we can always add Comments fields later.

Doug Ewell | Thornton, CO, USA
http://ewellic.org | @DougEwell ­

More information about the Ietf-languages mailing list