My reaction is very much the same as Rebecca's. The fact that some systems "require" alpha-2 language identifiers means that there are some systems that need to be changed to allow longer language identifiers.

ISO 639-1 is up for periodic review. The review period ends in December 2007. But, please remember: Regardless of what the outcome of that vote might be, that has NO bearing on the assignment of alpha-2 identifiers. Proposals for new language identifiers are processed continuously by the JAC, and this is not a part of the issue for the review (or any subsequent revision). The JAC will not assign alpha-2 language identifiers because someone doesn't have room for an alpha-3 identifier.

Users need to be adviced that there is no difference in meaning between an alpha-2 and the corresponding alpha-3 identifier (fr = fra = fre). There is also no difference between the meaning of an alpha-3 identifier in 639-2 and the same identifier in 639-3. We cannot control how people use the identifiers, but if someone has used them in any way different from this, they have not used them according to the specifications in the standard.

Best regards,

Håvard Hjulstad
  Standard Norge / Standards Norway
  [log in to unmask]

-----Original Message-----
From: ISO 639 Joint Advisory Committee [mailto:[log in to unmask]] On Behalf Of Rebecca S. Guenther
Sent: Thursday, August 16, 2007 7:04 PM
To: [log in to unmask]
Subject: [Fwd: ISO 639-1 two-letter codes] (fwd)

I wasn't aware that this was happening. Can someone please give me the background on this? We have said that we would not add 2 character codes if there was already a 3 character code because of the instability it would cause. So where is this coming from? Why would the ISO 639 JAC not be informed about this?

My answer is that those applications that can't accept 3-character codes will need to change. 


-------- Original Message --------
Subject: ISO 639-1 two-letter codes
Date: Sun, 12 Aug 2007 21:22:20 -0400
From: DeCamp, Jennifer A. <[log in to unmask]>
To: Lutz, Richard D. <[log in to unmask]>,        Zaideman, Wayne H. 
<[log in to unmask]>,        Pet, Mandy S. <[log in to unmask]>, Wallin, 
Jay W. <[log in to unmask]>,        <[log in to unmask]>, 
Silman, Morton CIV USA INSCOM <[log in to unmask]>,        Nordin, 
Glenn Mr CIV OSD OUSDI <[log in to unmask]>, <[log in to unmask]>, <[log in to unmask]>, <[log in to unmask]>

For organizations that have legacy databases that cannot use more than two characters for language codes, your time has come.  ISO is going to be reviewing the ISO 639-1 standard on two-letter language codes.  I am on that committee.  If there are any codes that you need, please let me know.  I can be reached at the following addresses:

[log in to unmask]
[log in to unmask]
[log in to unmask]
703 347 5137

Please send me a statement of what you are trying to designate with the language code, and--if you like--a preferred letter combination that is not already taken.

Try to avoid the ISO 639-1 standard if you can.  It is very uneven in level, and the meanings of many of the language codes are undefined.
The codes are thus used differently in different situations.

Please feel free to send this email to anyone you think may be interested.