Re: tlug: Re: Japanese input

[Kei Furuuchi <kfur@example.com>]

Taro Yamamoto writes:
 > 	Translating Japanese words into another natural language such as
 > English looks to be something peculiar. If a message writer speaks
 > French & Japanese, he would prefer using French (as far as I understand
 > the purpose of this new notation you propose). But the grammer of the
 > notation is based on Japanese,so message recievers need to know both
 > French & Japanese. Also, it is thought that Russians prefer Russian,
 > Germans prefer German, Catalans prefer Catalan, etc. . . . The message
 > receiver needs to understand various different natural languages in
 > addition to Japanese.

Again I am not advocating that my proposal is best. I want
to make a frame of reference, something to compare as to how 
special is kanji notation. If it is a French mail list, it
is French I will use with. And I am not the one who decided
to use English in this mail list. And English is somehow
favorite language in internet.


 > 	If we limit the language for the new notation to just English and
 > Japanese, it should work, but it can not express Japanese sounds
 > (pronunciations), while it is partly succeed in expressing meanings.
 > Also, unlike programming languages (and their target machine languages),
 > the mapping between English and Japanese words is not necessarily a
 > one-to-one mapping (btw, I think programming languages are just using
 > English words and they do not want to be a natural languages. Grammers
 > of computer languages are artificially very limited and usually more or
 > less deterministic. If Russians had dominated the computer industry,
 > Russian words and characters would have been used in computer
 > languages).

I meant if you can make one-to-one mapping out of kanji and
jis-code why not make one-to-one mapping out of kanji and
English word by further compromise.

I make one example. Let's say:

	"Earth girls are easy."

Word "easy" will be translated into "やさしい". But, what
will this mean? Are they kind or easy? So use a kanji to
supplement a word: 易しい. Then someone who understands
kanji and kanas, but do not know how to pronunce nor read
Japanese may understand what it means. 
This is the advantage of using kanji over using mere kanas
or romanjis. If it is so, why not use English word to
supplement a word: "easy shii". Then someone who understand
English and alphabets, but do not know how to pronunce nor
read Japanese may understand what it means.
This is the advantage of using English words over using mere 
kanas or romanji.
The advantage of using English word over kanji is that
pronunciaton of them have less same ones over kanji and
pronunciation is up to date. The disadvantage is that
English presents third largest population of users with
first Chinese and less habitual because nobody does it.

 > 	Thinking this way, I am skeptical about the effect of the separation of
 > meaning and sound in the proposed notation. I think the traditional
 > romaji notation works better, if we add some explanatory notes in
 > English to distinguish conflict words. Also I think it is possible to
 > improve the romaji notation by using punctuation marks and hyphens
 > appropriately.
 > 
 > 	For example:
 > 	"FORTRAN wa tsune ni kagaku-teki (scientific), suuchi-teki (numerical),
 > kogaku-teki (engineering) programming ni oite shuyou na (principal)
 > gengo de atta . . ." (Michael Metcalf and John Reid, FORTRAN 90/95
 > explained, p. v., Oxford University Press, Oxford 1996)
 > 

I think it is the same thing you suggest provided one-to-one 
mapping is made. you can elimiate kanji comprex part because 
of one-to-one.

 > 	Refining romaji looks to work better than inventing a new notation.
 > 	IPA can be used instead of romaji. But both romaji and IPA that express
 > sounds only do not suffice for writing Japanese texts because of the
 > abundance of conflict words. Hints to resolve ambiguities need to be
 > added. (This exemplifies the validity and efficiency of kanji in
 > Japanese). Also, IPA is like a machine language with many instructions.
 > Every sound can be expressed in IPA, but poeple will prefer a simpler
 > notation that suffices for describing Japanese sounds only, as
 > programmers like to use high-level programming languages that are
 > suitable for their own purposes. Some languages may require IPA, but we
 > can use romaji instead.
 > 

Anyway I am saying that mechanically replace kanji with
English word, retaining Kanji pronunciation too. So
computers can handle them. Then with little grammer, people
may understand the meaning without knowing how to
pronounce. This may help communicatte better in massive way.

 > > 6) Once in internet, there is no control over how to use English. And
 > > there is Japanese usage of naive and the rest of the world usage of
 > > it. I wonder which is going to win once used in internet.
 > 
 > 	Yes, we are writing English sentences in Japanized (and often
 > incorrect) ways. It is one remarkable result of the ASCII-centric
 > communication on computers and networks. Those who have been deprived of
 > their writing method by an influential foreign culture have to use the
 > language of the foreign culture, then start transforming the language
 > and penetrating into the ruling culture. It's an ironical situation.

I like to propose another:
In this situation, "thank you" in Japanese is
"arigatou". And this is only initial part of the sentence
that goes on to amount to say thank you. The actual
translation should be "unprecedent". So I like to use
unprecedent to mean thank you.

Regards,
Kei.
--------------------------------------------------------------
Next TLUG Meeting: 13 June Sat, Tokyo Station Yaesu gate 12:30
Featuring Stone and Turnbull on .rpm and .deb packages
Next Nomikai: 17 July, 19:30 Tengu TokyoEkiMae 03-3275-3691
After June 13, the next meeting is 8 August at Tokyo Station
--------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp