Isn't it time that we developed a firm policy about the relative role of codes from ISO 639-2 and ISO 639-3, in advance of ISO 639-3 being published? I'm more and more inclining to the simple view (hopefully not a simplistic vews) that (a) if a language entity exists in ISO 639-3, there needs to be no special effort to add it to ISO 639-2; and (b) if the current specifications of a user group is currently limited to ISO 639-2, they should change their specifications so that it allows use of 3-letter codes from either ISO 639-2 or ISO 639-3, as there is no clash between them. An example of such current specifications which might need a simple amendment could be (1) the MARC21 codes, and (2) ISO 3066 or its successor/replacement. As noted above, I realise that ISO 639-3 is not yet published, but it seems to be very likely that it WILL be published in its present form or something very close. We therefore need to develop such a policy in advance of publication, and not leave it until after its publication. If such a policy is developed, it would save a large amount of time for members of the ISO 639 JAC, and we need not keep considering case after case like this one, of which there will be many thousands which could be submitted (where ISO 639-2 would merely have added a code which is already in ISO 639-3). There are also "macro language codes" (or the other other "large language group" codes) which might fit from the present ISO 639-2 (the ones like South American Indian etc, which will not appear in ISO 639-3). We should ask the Joshua Project which particular problems will occur in IT systems, particularly in the open source translation project called Zitt (http://zitt.sourceforge.net), and in the GNOME desktop (http://www.gnome.org) which they mention. This is especially so given the current confusion and disagreemeents on the IETF list about the actual use of specific IETF language tags. What views do other ISO/JAC members have? John Clews ---------------------------- Original Message ---------------------------- Subject: New ISO 639 proposal: Gong; Kagoma - Discussion (until 2005-01-14 ) From: Håvard Hjulstad <[log in to unmask]> Date: Mon, December 13, 2004 3:24 pm To: [log in to unmask] -------------------------------------------------------------------------- Dear JAC members, Please see the submission below. The submitter refers to Ethnologue: http://www.ethnologue.com/show_language.asp?code=KDM. The identifier "kdm" is also the one that is used in draft 639-3. However, for some reason the submitted request asks for "gng" (which is in use in Ethnologue and 639-3). Ethnologue claims that there are 6250 speakers only of this language. In Linguasphere the item is probably the one that is encoded 98-IAB-f OR 98-IAB-fa The following represents the encoding in Linguasphere: 98 = BENUIC phylozone 98-I = HYAM+TAROK set 98-IA = HYAM+IZERE chain 98-IAB = HYAM+YESKWA net 98-IAB-f = Kagoma 98-IAB-fa = ka-goma [gwong, gyong, kwong, agoma] (Ethnologue has the same alternate names.) If an alpha-3 identifier is assigned for 639-2, it should be "kdm". DISCUSSION please (until 2005-01-14). Best regards, Håvard ~~~~~~~~~~~~~~~~ Håvard Hjulstad Standard Norge / Standards Norway P.O.Box 242, NO-1326 Lysaker (Norway) +47 67838645 (direct) fax: +47 67838601 http://www.standard.no/ mailto:[log in to unmask] ~~~~~~~~~~~~~~~~ -----Original Message----- From: Rebecca S. Guenther [mailto:[log in to unmask]] Sent: 10. desember 2004 16:01 To: Håvard Hjulstad Subject: New ISO 639-2 code (fwd) Another one from the same group. I guess this is Kagoma in Ethnologue. ---------- Forwarded message ---------- Date: Thu, 9 Dec 2004 17:34:57 -0500 From: WWW generic account <[log in to unmask]> To: [log in to unmask] Subject: New ISO 639-2 code This data was submitted on: Thursday, December 9, 2004 at 17:34:57 lang_in_eng = GONG, Kagoma lang_in_fre = ref_where_found_1 = http://www.ethnologue.com/show_language.asp?code=KDM lang_in_vern = GONG ref_where_found_2 = As related by the native speakers themselves trans_lit evidence = We cannot swear that there are 50 or more documents available in Gong (the way we can for its fellow Jaba cluster language member Hyam), but the possibility is there. We would like to point out however, that we will produce documents exceeding that number, mostly of a technical nature, but not exclusively; since this language, like the other four we are working with, are minority languages, we are intent on promoting them and allowing the youth in the areas where these languages are spoken to acquire ICT skills in their own languages. addinfo = Fantsuam Foundation - http://www.fantsuam.org - is the institution recommending an individual ISO 639-2 code for Gong. Gong is a Nigerian language spoken in Kaduna State, Nigeria by about 6,360 people (Joshua Project - http://www.joshuaproject.com) and is a member of the Jaba language cluster. The native speakers refer to their language as Gong, while outsiders call it Kagoma. The request for an individual code for Gong is being made by Fantsuam Foundation at this time because an open source translation project called Zitt (http://zitt.sourceforge.net) has commenced. The Fantsuam Foundation has an active ICT program training local youth who speak the five languages we intend to translate. In order for our work to be incorporated into software such as the GNOME desktop (http://www.gnome.org) - the first piece of software we are localizing - we have been asked by the GNOME development team (specifically Christian Rose) to acquire individual codes for each of the five languages (the others are Gworok, Hyam, Jju and Tyap - separate submissions will be made for each). In addition to localized open source software, our work will also produce the following: -Gong glossary -Gong grammar -online Gong IT dictionary -online Gong-English IT dictionary -online Gong-English general dictionary -website (http://zitt.sourceforge.net) in Gong -Gong translations of documentation associated with each localized software -basic/introductory ICT training/how-to documents in Gong -information on the Gong people and their land (as the information available online is quite scanty and not in Gong) One of the reasons we are translating open source software into Gong is to encourage publishing, both on and offline, in Gong. We intend to lead the way, as demonstrated by the list above. request_addition = ISO 639-2 only 2_code_suggestion = 3_code_suggestion = gng submit_name = Uchenna Agbim submit_email = [log in to unmask] submit_status = I am the project manager/coordinator for the Zitt Open Source Translation Project which will be doing the translation work. The Zitt Open Source Translation Project draws nearly all of its members from the ranks of Fantsuam's ICT students.