[ChineseGP] 答复: latest versions of CGP and JGP variant tables?

Fri Apr 24 03:15:39 UTC 2015

Dear Hotta San, Dear All

Yes, I think Yoneya' algorithm and terms work for us.

I think if everything goes smoothly,  Han full-repertoire will be unified
set =  Chinese LGR-1 repertoire + Japanese LGR-1 repertoire + Korean LGR-1
repertoire
Which means, Japanese LGR-1 doesn't have variants and basically accepts all
the variants in Chinese LGR-1, and so do Korean LGR-1
In that case, MSS will never be used.

MSS is defined as limited sub-set of CGP repertoire, which contains proposed
variants and will be used to coordinate with J and K 
It means, some specific code points in MSS, CGP might delete them or change
their status according to the coordination result. (especially in the case
like "»ú" and "™C")
In that case,  Han full-repertoire  =  Chinese LGR-2 repertoire + Japanese
LGR-1 repertoire + Korean LGR-1 repertoire
Anyway, as Hotta mentioned, MSS is only an intermediate tool that is
introduced to speed up Yoneya's algorithm.

Regards
Wang Wei

-----ÓÊ¼þÔ¼þ-----
·¢¼þÈË: hotta at jprs.co.jp [mailto:hotta at jprs.co.jp] 
·¢ËÍÊ±¼ä: 2015Äê4ÔÂ22ÈÕ 13:23
ÊÕ¼þÈË: Íõ?
³ËÍ: KIM Kyongsok; Dongman Lee; Joe Zhang; Wang Wei; Kenny Huang; Yuri;
Yoshiro YONEYA; Zhang Zhoucai; <qichao at cnnic.cn>; Zhiwei Yan;
KoreanGP at icann.org; ChineseGP at icann.org; JapaneseGP at icann.org; ? ??; ???;
???
Ö÷Ìâ: Re: latest versions of CGP and JGP variant tables?

Dear Wang Wei,

I appreciate your effort in making the way easier and faster.
I think I understand the purpose of the introduction of MSS, and you have my
sympathy to some extent.

Please see my observation and comments below.
I need your and everybody's feedback.

===

Here, let me use the technical terms in Yoneya's algorithm we tentatively
agreed in Dallas.

<premises>

(1) Japanese LGR-1 definition is almost finished by JGP
     - Japanese LGR-1 repertoire      : intersection of JPRS repertoire 
                                        and MSR
     - Japanese LGR-1 variants        : no variants
     - dispositions in Japanese LGR-1 : all characters in Japanese 
                                        LGR-1 are 'allocatable'
     - Japanese LGR-1 WLE             : no rules in WLE

(2) JGP has tentatively decided that Japanese LGR-2 would basically 
    accept all the variants defined in Chinese LGR-1 and Korean LGR-1
      (the appropriateness of this decision just needs to be confirmed 
       after CGP and KGP define the variants)

(3) Han full-repertoire =  Chinese LGR-1 repertoire
                         + Japanese LGR-1 repertoire
                         + Korean LGR-1 repertoire

<then let me conclude that>

(A) Definition of MSS is not necessary for JGP.

    This is because
     - Japanese LGR-1 repertoire is already defined using JPRS 
       repertoire and ICANN MSR
     - Japanese LGR-1 doesn't have variants

    To put it in the other way around, Japanese LGR-1 is not affected 
    by the definition of MSS, no matter how MSS is defined.

    This means JGP does not object the current MSS proposal if (1)(2)(3)
    stand, no matter how MSS is defined. 

(B) Therefore, MSS can be defined as the union set of current MSS 
    proposal and Japanese LGR-1 repertoire, if CGP and KGP also agree. 
    And the definition of the variants within MSS can be discussed 
    between CGP and KGP. JGP will observe the discussion and the 
    result.

    Here, if Korean LGR-1 doesn't have variants and KGP decides to 
    basically accept all the variants in Chinese LGR-1, it's pretty 
    straight-forward to define the RootLGR Han variants, because 
    variants are exactly the same as those defined in Chinese LGR-1.

(C) JGP just wants to see the Han RootLGR repertoire and variants 
    that can be automatically generated using Yoneya's algorithm and 
    wants to confirm that there is no big problem in the generated 
    Han RootLGR.

(D) CGP's proposal using MSS seems to be an ingenuity to speed up and 
    to avoid big backtrack of the LGR development, I think it works 
    if C/J/K repertoires are defined soon (actually CGP and JGP have 
    already defined their draft repertoire).

(E) How about using Yoneya's algorithm and terms (such as LGR-1 and 
    LGR-2) in explaining CJK activities and its output to the 
    community and using MSS in speeding up the development?

    This is because 
    - it is easier to understand
    - it looks more neutral/fair to all C, J, and K
    - MSS can be regarded as an intermediate tool that is introduced 
      to speed up Yoneya's algorithm in the case CJK is confronting 
      now.

Regards,
Hiro

On Thu, 16 Apr 2015 23:38:18 +0800 (GMT+08:00)
Íõ? <wangwei at cnic.cn> wrote:
> 
> Dear All
> 
> After the discussion in today¡®s CGP meeting, Prof. ZHANG and Mr. QI Chao
has produced a Minimum Shared Set (MSS) for C, J and K.
> 
> The MSS is consisted of the following code point subset:
> 
> A: IICore Base (9810, all fall in MSR)
> B: Historically Registered Hanzi under .CN/TW/HK (7000/5292/4191, all 
> fall in MSR)
> C: PRC National Normalized Hanzi along with their traditional variants 
> (8105/3489, some out of MSR)
> 
> MSS=A+B+C-D-E
> 
> D: code points in C, but not in MSR
> E: Stroke or Radicals (31) LIDU?(5)  Out of BMP(62)
> 
> 
> finally, we got 12531 valid code points aggregate a Minimum Shared Set,
based on which C, J and K could define their own variants, or adopt others
variants.
> 
> After all three GPs finish the variants generation and coordinaiton on 
> MSS, each GP still has the chance to add independent code points (with no
variants) into MSS seperately, to from its own final repertoire.
> 
> STEP1: MSS generated and accepted by three GPs, a mininum code points set
shared by all of us.
> STEP2: variants generated and coordinated based on MSS
> STEP3: Add each GP's own independent and unrelated code points, form the
final formal repertoire.
> 
> Does this process work for you? please feel free to give your comments.
> 
>  
> 
> Regards
> 
> WANG Wei
> 
> PS: special thanks for the hardwork of Mr. QI Chao
> 
> 
> 
>  
> 
> -----ÔÊ¼?¼þ-----
> ?¼þÈË: "KIM Kyongsok" <gimgs0 at gmail.com>
> ?ËÍ??: 2015Äê4ÔÂ16ÈÕ ÐÇÆÚËÄ
> ÊÕ¼þÈË: "hotta at jprs.co.jp" <hotta at jprs.co.jp>
> ³ËÍ: "Wang Wei" <wangwei at cnic.cn>, "Dongman Lee" <dlee at cs.kaist.ac.kr>, 
> "Joe Zhang" <joezhang43 at hotmail.com>, "Wang Wei" <wangwei at cnnic.cn>, 
> "Kenny Huang" <huangksh at gmail.com>, Yuri <takamatsu at jprs.co.jp>, 
> "Yoshiro YONEYA" <yoshiro.yoneya at jprs.co.jp>, "Zhang Zhoucai" 
> <joe.zhang at unihan.com.cn>, "<qichao at cnnic.cn>" <qichao at cnnic.cn>, 
> "Zhiwei Yan" <yanzhiwei at cnnic.cn>, "KoreanGP at icann.org" 
> <KoreanGP at icann.org>, "ChineseGP at icann.org" <ChineseGP at icann.org>, 
> "JapaneseGP at icann.org" <JapaneseGP at icann.org>, "? ??" 
> <mjkim at kisa.or.kr>, "???" <isanjo at empas.com>, "???" 
> <partizan21 at naver.com>, "KIM Kyongsok" <gimgs0 at gmail.com>
> Ö÷?: latest versions of CGP and JGP variant tables?
> 
> 
> Dear everybody:
> 
> 
> KLGP is trying to prepare a repertoire and a variant table.
> 
> 
> I would like to get the (or almost?)  latest versions of CGP and JGP
variant tables.
> 
> 
> Thanks very much in advance.
> 
> 
> --
> 
> ? ??      KIM, Kyongsok
> 
> 
> 
> 