RFC 4645bis: making 'pes' and 'prs' extlangs

Doug Ewell doug at ewellic.org
Fri Dec 5 15:39:19 CET 2008

As far as I can tell, this originated as a private request to make 'pes' 
and 'prs' extlangs under 'fa', under the not-yet-approved RFC 4646bis, 
and a second request was later added, to change the names "Western 
Farsi" and "Eastern Farsi" for these subtags to use the word "Persian" 
instead.  In keeping with a time-honored Internet tradition, nobody ever 
changed the Subject line of the e-mails to deal with the second request, 
leading to some of the confusion.

My understanding is that the extlang request (which would be an LTRU 
issue) has been withdrawn.

The existing Registry has the following entry:

Type: language
Subtag: fa
Description: Persian
Added: 2005-10-16
Suppress-Script: Arab

This would be unchanged in RFC 4646bis except to add a new field:

Scope: macrolanguage

This is an informative field that reflects the status of the ISO 639-3 
code element 'fas', which is equivalent to the ISO 639-1 code element 
'fa', which (because it exists) is used as the BCP 47 subtag.  It has no 
direct bearing on how this subtag is to be used to create language tags.

There is no suggestion to change this record further.  The suggestion, 
rather, is to change the following two records which are proposed for 
RFC 4646bis, and which correspond to other ISO 639-3 code elements:

Type: language
Subtag: pes
Description: Western Farsi
Added: 2029-09-09
Macrolanguage: fa

Type: language
Subtag: prs
Description: Dari
Description: Eastern Farsi
Added: 2029-09-09
Macrolanguage: fa

Note that the "Macrolanguage" field is also informative and has no 
direct bearing on the use of these subtags in BCP 47 language tags.

These are the only uses of the word "Farsi" in the RFC 4646bis Registry 
(which is included in RFC 4645bis).  There are other subtags with 
descriptions like "Judeo-Persian" and "Old Persian (ca. 600-400 B.C.)" 
which are not part of this discussion.

"Farsi" is the local name for this language, or group of languages, 
while the word "Persian" is generally preferred and more commonly used 
in the West.  However, there is some controversy over this, especially 
involving recent immigrants and writers still living in Iran.  While 
"Persian" is the preferred name, it would probably be excessive to say 
that the name "Farsi" is absolutely wrong.  (I personally take no stand 
on this.)

If desired, these descriptions could be changed in the RFC 4646bis 
Registry in one of two ways:

1.  ISO 639-3 could be persuaded to change their names.  The Registry 
almost invariably uses ISO 639 names, so the corresponding changes to 
the Registry would occur more or less automatically.

2.  There could be a request on this list (ietf-languages) to add the 
"Persian" names as additional Description fields for these subtags, in 
addition to the "Farsi" names taken from ISO.  Note that there are no 
"primary" or "preferred" names in the Registry; the ISO 639-3 reference 
name (when applicable) is listed first in the Registry for ease of 
cross-referencing only.  Note also that this request could not be made 
until the subtags actually exist in the Registry, i.e. until RFC 4646bis 
is approved by the IESG.

Doug Ewell  *  Thornton, Colorado, USA  *  RFC 4645  *  UTN #14
http://www.alvestrand.no/mailman/listinfo/ietf-languages  ˆ

More information about the Ietf-languages mailing list