Regarding the japanese single char. auction

[Rubber Duck]

Well this at least might help to demystify how one version of the Japanese Keyboard works:

http://translate.google.com/translat...hl=en&ie=UTF-8

Guess I prefer this version though:

[jose]

BTW, Where's Craig? He sure would have a word to say on this...

[rhys]

Originally Posted by touchring

Precisely, everything is now speculation.

But we've progressed a long way since RD first sold the kanas nearly 2 years ago, for those of you who could remember that sale - i remembered it well as i placed in a bid, only to forget to renew my bids!! :o

btw, aren't most single kana.jp reserved?

Now that you mention it touch, I only own them in .com. In .JP I was mistakenly thinking of some single character kanji I own...

コ.com
キ.com

Hardly authoritative but another anecdotal source

http://www.japan-guide.com/e/e2052.html

"The most common input method is to enter Japanese words in their romanized writing (for example: toukyou). The input will be automatically converted into hiragana as you write, and can then be switched to corresponding kanji by pressing the space key."

There is also this blog entry from a Japanese blogger named Mari who I often enjoy reading. She isn't exactly clear in describing this but it is clear that she uses the same romaji input method that I do.

Here is her entry in Watashi to Tokyo

"We Japanese use PCs which can type Japanese and have Japanese keyboards.Like I wrote before, we have four character sets. The alphabet, Katakana, and Hiragana are okay. But when I type Chinese characters, I type its "phonetic sound" by alphabet (Roman method) and I have to convert it until I get the one I mean because there are many Chinese characters."

[Rubber Duck]

Thanks Reece, anything is a help but the Watashi article which is in English is some three years out of date. Just does'nt seem to be any real hard up to the minute information available.

[rhys]

Originally Posted by Rubber Duck

Thanks Reece, anything is a help but the Watashi article which is in English is some three years out of date. Just does'nt seem to be any real hard up to the minute information available.

Dude, Input Methods are not anything that will change very quickly. Three years is fresh for things like keyboard layouts and Input Methods. The pace of change is glacial at best. No one who is already happy with one method is going to go and throw that away to learn a new method unless it is significantly better AND easy to learn. If, by chance, new typists are being trained in grade school on some funky new keyboard or IM I would wager we won't see them on the demographic radar in large numbers until you and I are enjoying retirement.

For example, I'm a touch typist on a querty keyboard someone would have to pay me a year's salary upfront to start typing "the quick brown fox jumps over the lazy grey dog" again on another keyboard layout.

[touchring]

Originally Posted by rhys

But when I type Chinese characters, I type its "phonetic sound" by alphabet (Roman method) and I have to convert it until I get the one I mean because there are many Chinese characters."

Seem that pinyin is easier for kanji input! all different sounds!

申 shen
黄 huang
攻 gong
效 xiao
项 xiang
工 gong
候 hou
恒 heng
宏 hong

[mgrohan]

Talking about Chinese input. Anyone know what is the main Chinese input nowadays? Anyone seen any statistics for pinyin/五笔 wubi, etc

[Rubber Duck]

Yes, but remember, just because they look like Chinese characters, doesn't mean they are pronounced the same in Japanese or indeed mean exactly the same thing. Romaji and Pinyin are quite different I believe.

Is Japanese a tonal language like Mandarin? Do you need the numbers after the Pinyin to determine the pitch and hence the correct meaning?

Isn't Pinyin language specific anyway. Does it not just work for Mandarin? Do the Cantonese not have a different PinYin system?

Originally Posted by touchring

Seem that pinyin is easier for kanji input! all different sounds!

申 shen
黄 huang
攻 gong
效 xiao
项 xiang
工 gong
候 hou
恒 heng
宏 hong

[touchring]

Originally Posted by mgrohan

Talking about Chinese input. Anyone know what is the main Chinese input nowadays? Anyone seen any statistics for pinyin/五笔 wubi, etc

i did a search, sougou pinyin ime seems to be very popular. i downloaded and installed the software, here's a glimpse of how it looks like:




The ime also can allow the extension to be entered w/o switching of keys.

[Rubber Duck]

Not sure why this should ever have been an issue. The keyboard mode should have remained as Latin throughout. Unless you have a Wubi, is there really such a thing as a Chinese Keyboard?

Japanese is total different than Chinese. They have their own phonetic alphabet.
The Chinese don't. You either use Latin to generate PinYin for the IME or use Wubi to character compilation for direct entry. With Japanese, the IMEs seem to work off Hirigana, inputting in Romaji is just another way of generating the Hirigana.

[rhys]

Originally Posted by Rubber Duck

Not sure why this should ever have been an issue. The keyboard mode should have remained as Latin throughout. Unless you have a Wubi, is there really such a thing as a Chinese Keyboard?

Japanese is total different than Chinese. They have their own phonetic alphabet.
The Chinese don't. You either use Latin to generate PinYin for the IME or use Wubi to character compilation for direct entry. With Japanese, the IMEs seem to work off Hirigana, inputting in Romaji is just another way of generating the Hirigana.

Technically it is not an alphabet. It is a syllabary, a set of written symbols that represent (or approximate) syllables, which make up words. BTW, Japanese is not tonal to answer a previous question. And to summarize where this whole thread has gone - romaji is used as the primary means of generating hirigana/katakana/kanji when a querty keyboard is being used. There are separate input modes for hiragana and for katakana though you can generate katakana from the hirigana mode too. The kanji can be generated from either mode.

[Rubber Duck]

Originally Posted by rhys

Technically it is not an alphabet. It is a syllabary, a set of written symbols that represent (or approximate) syllables, which make up words. BTW, Japanese is not tonal to answer a previous question. And to summarize where this whole thread has gone - romaji is used as the primary means of generating hirigana/katakana/kanji when a querty keyboard is being used. There are separate input modes for hiragana and for katakana though you can generate katakana from the hirigana mode too. The kanji can be generated from either mode.

Yes you are correct on the syllabary point and it is not unique in the respect. Indian languages use syllabries as well. Even the Arabic alphabet has some similar characterstics as most vowels are not explicit. Nevertheless, they are the basic building blocks from which words are made, so they have fundamental characteristics that closely resemble those of our letters, and they are very much phonetic in nature.

Thanks for the clarification on tonality. That should make it clear to most that haven't understood the point that Japanese is not just a distinct Chinese dialect. It is very much a distinct language with very distinct characteristics.

On the keyboard use side of things, it would seem that most of the anecodetal evidence seems to support your view. The picture, however, is one of a complex problem driving a lot of diverse solutions and constantly evolving user behaviour.
I think we can agree that the this mode of inputting has only evolved since the introduction of the World Processor. Clearly when typewriters were being used to write Japanese, it would not have been possible to do the conversion, and this has only happened since the introduction of IME which is comparatively recent. It would seem that the use of romaji has been primarily driven by the inability of machines to cope with Japanese characters and this clearly has often changed the nature of end product until the technology was able to give the Japanese their own language in their own script. If you typed in Romaji the reader obviously had to read in Romaji. What is clear is that apart from expressing some foriegn words or brands in Romaji there is little demand for printed matter in Romaji. It is also clear that there is no over-riding logic in inputting in Romaji other than because that is what people have been used to doing. I don't think any of can really know for sure how far things have shifted towards directly inputting the Katakana characters, or how much further it will be used in the future. It was interesting, however, that you noted that on mobile phones, it would seem that Hiragana is dominant. That would at least suggest that younger generation of Japanese are very comfortable with their own character sets and that that is the format in which the thought product of their minds is crystalising.

[rhys]

My clarifications inline in red.

Originally Posted by Rubber Duck

If you typed in Romaji the reader obviously had to read in Romaji.

Yeah, this is not obvious in fact I would say it is an erroneous conclusion. Japanese never read in romaji. What this lack of wordprocessors in Japan spawned was continued use of handwritten documents for commerce, government, and personal purposes and of professional printing services for books and magazines while the west used typewriters. You might not know this but before the widespread use of the PC in the mid-90's, there was at least a decade and a half in Japan of an electronic pc/typewriter hybrid called a "wapuro" or "wordprocessor" which was very very widely used. This is the machine I first learned to type Japanese on. So the IME solution has at least a 27 year history in Japan.

What is clear is that apart from expressing some foriegn words or brands in Romaji there is little demand for printed matter in Romaji.

True - let's say "none"

It is also clear that there is no over-riding logic in inputting in Romaji other than because that is what people have been used to doing.

Also true. But changing this will require quite a lot of time. Historical practices are not always thrown out simply because a better solution comes along. QUERTY is far from ideal for English as well but nothing is superceding it in the west.

I don't think any of can really know for sure how far things have shifted towards directly inputting the Katakana characters, or how much further it will be used in the future.

I think we can assume that so far very little. The future however is unknown.

It was interesting, however, that you noted that on mobile phones, it would seem that Hiragana is dominant. That would at least suggest that younger generation of Japanese are very comfortable with their own character sets and that that is the format in which the thought product of their minds is crystalising.

Well specifically, not hiragana so much as "kana". And your last point here is true but has never been in doubt because again Japanese never read in romaji to begin with.

[Rubber Duck]

Thanks Rhys that is most enlightening.

[touchring]

Rhys put out an important point that reading is different from talking. When people type romaji or pinyin, they are talking or translitering. Reading that transliteration is step more difficult than saying it out. The difference is more obvious when you try and read a sentence or phrase of pinyin or romaji.

[kenne]

Wonderful discussion. This is why idnforums so addictive. My search for toys somehow ended up here. I don't know what happened.

I'm trying to find a high resolution japanese cellphone keyboard (typical config). Not having much luck.

[Rubber Duck]

[kenne]

Looks like they have a "@" sign at a prominent position. Probably use email a lot on cellphone?

Also, this keyboard is less crowded than english keyboard. Japanese uses only about a dozen alphabet sound. Are these basic ones the ones represented here? Or they put the most common syllables on the keyboard, and let user toggle to get the less common ones?

Very interesting. Thanks Duck.

[Rubber Duck]

Is that a Yahoo key top right?

Kenne,

If you look here:

http://en.wikipedia.org/wiki/Hiragana

It becomes fairly obvious how each character is accessed.