RE: [Ltru] Re: Registry deltas

BTW, I meant to add in this discussion of inverted names and other aspects of the forms of descriptions (not registry deltas!), if there is feedback that this WG would want to send back to the ISO 639 JAC for their consideration as to changes they could make in the descriptions that would make things easier for the LSR, this is a good time to submit that as the JAC will be meeting in a couple of weeks (teleconf) to discuss various issues related to the completion of part 3 and maintenance of the various parts in relation to one another. There have been some plans to work on names (e.g. the 639-3 RA had a work item to review the names in the 639-3 draft code table to improve consistency, especially where the draft code table for part 3 doesn’t have names that are widely used, such as “Dari”).

(Also, btw: seeing the problems we have with managing description fields has for the first time made me wonder if the formal overhead of ISO 11179 might not offer some benefit. I don’t know enough about 11179 to be able to answer that; it’s just a question that popped into my head.)

Peter

From: Peter Constable [mailto:petercon at microsoft.com]
Sent: Monday, October 16, 2006 10:05 AM
To: LTRU Working Group
Subject: RE: [Ltru] Re: Registry deltas

Definitely some of the changes below would be wrong: e.g.,

- “Yi, Sichuan” is an inverted form of “Sichuan Yi”

- “Agta, Camarines Norte” is an inverted form of “Camarines Norte Agta”

- “Malay, Ambonese” is an inverted form of “Ambonese Malay”; “Ambonese” is an alternate name, but “Malay” probably isn’t a good choice for an alternate name since that would generally be understood as Bahasa Malaysia

- “Nahuatl, Highland Puebla” is an inverted form of “Highland Puebla Nahuatl”

More of these are similar; a closer review would be needed to determine which of these changes would really make sense.

Peter

From: Mark Davis [mailto:mark.davis at icu-project.org]

CHANGE TO SEMI-COLON

Yi, Sichuan    =>    Yi; Sichuan
Ayta, Ambala    =>    Ayta; Ambala
Agta, Camarines Norte    =>    Agta; Camarines Norte
Ayta, Abenlen    =>    Ayta; Abenlen
Malay, Ambonese    =>    Malay; Ambonese
Adi, Galo    =>    Adi; Galo
Tibetan, Amdo    =>    Tibetan; Amdo
Amis, Nataoran    =>    Amis; Nataoran
Amuzgo, Guerrero    =>    Amuzgo; Guerrero
Arapesh, Bumbita    =>    Arapesh; Bumbita
Asmat, Casuarina Coast    =>    Asmat; Casuarina Coast
Asuriní, Xingú    =>    Asuriní; Xingú
Asmat, Yaosakor    =>    Asmat; Yaosakor
Manobo, Ata    =>    Manobo; Ata
One, Molmo    =>    One; Molmo
Gbe, Ayizo    =>    Gbe; Ayizo
Atta, Faire    =>    Atta; Faire
Nahuatl, Highland Puebla    =>    Nahuatl; Highland Puebla
Kayan, Busang    =>    Kayan; Busang
Birifor, Malba    =>    Birifor; Malba
Pahari, Mahasu    =>    Pahari; Mahasu
Bareli, Rathwi    =>    Bareli; Rathwi
Chin, Bawm    =>    Chin; Bawm
Béte, Guiberoua    =>    Béte; Guiberoua
Bété, Daloa    =>    Bété; Daloa
Bareli, Pauri    =>    Bareli; Pauri
Itneg, Banao    =>    Itneg; Banao
Sorsogon, Masbate    =>    Sorsogon; Masbate
Karen, Pa'o    =>    Karen; Pa'o
Ayta, Mag-Indi    =>    Ayta; Mag-Indi
Kanuri, Bilma    =>    Kanuri; Bilma
...

NOTE
As I was doing this, I saw the following. It may be perfectly fine, but it looks a bit suspicious, especially since there are a bunch of "Atga" entries with different modifiers. Could this be a spelling error?

Agta, Mt. Iraya

Mark

On 10/9/06, Martin Duerst <duerst at it.aoyama.ac.jp> wrote:

[co-chair hat off]

At 02:22 06/10/03, Mark Davis wrote:
>Overall, I think inverted names are a bad idea. They are rather anglocentric, and we find that translators often do the wrong thing with them. They are also hugely inconsistent in the current registry.
>
>I think they are tolerable if (a) every single instance of a comma indicates an inverted name, and (b) all names are consistently inverted. Because of (a) one could programmatically fix the names to the correct value.

I don't think creating artificial syntax rules or some
kind of restricted (English) grammar for Description fields
is appropriate. What I can immagine is that we use semicolons
rather than commas to indicate boundaries between input
from different origins.

Regards,    Martin.

#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#   http://www.sw.it.aoyama.ac.jp        mailto:duerst at it.aoyama.ac.jp