Sunday, April 22, 2007

Google's IME

(Updated on 4 May 2007)

I now use Google's IME (Input Method Editor) for Chinese, released about a month ago, much more than Microsoft's. As well as pinyin input, Google's IME lets me enter the character strokes for the many characters I don't yet know the sound of. Very handy for foreigners learning Chinese. The keyboard assignments are intuitive: h for 一(heng), s for 丨(shu), p for 丿(pie), n or d for 丶(na or dian), and z for 乙(zhe). Because most foreigners learning Chinese would know the pinyin names of the four main strokes, having memorized "heng, shu, pie, na" till it hurts, it's easy to type h, s, p, n for these four strokes. And 乙 looks like z.

But only 6 keys are used. It would be nice if some extra common sequences of strokes were added to the 20 unused keys. For example, k for 口(kou), r for 日(ri), y for 月(yue), etc. This would be a lightweight addition, and easy to learn. We could learn them incrementally, always having the choice of stroke-by-stroke, or entering those components. Nothing heavy-duty like Wubizixing, which requires us to learn all of it before we can use any of it.

The IME lets us toggle between English and Chinese using the shift-key. Perhaps other natural languages that usually require an IME will be plugged in later, eg, Japanese, Korean. Maybe the IME will soon let us toggle between languages that don't usually require an IME, eg, English and French keyboard assignments. Also cool would be if we could use Google's IME to enter Unicode symbols and punctuation easily. Some possible keystroke sequences for the Latin-1 (0x0080 to 0x00FF) tokens:


À



A`











à


a`










Á


A'










á


a'
Â



A^











â


a^










Ã


A~










ã


a~
Ä



A:











ä


a:










Å


Ao










å


ao
Æ



AE











æ


ae










Ç


C,










ç


c,
È



E`











è


e`










É


E'










é


e'
Ê



E^











ê


e^










Ë


E:










ë


e:
Ì



I`











ì


i`










Í


I'










í


i'
Î



I^











î


i^










Ï


I:










ï


i:
Ð



D-











ð


d-










Ñ


N~










ñ


n~
Ò



O`











ò


o`










Ó


O'










ó


o'
Ô



O^











ô


o^










Õ


O~










õ


o~
Ö



O:











ö


o:










Ø


O/










ø


o/
Ù



U`











ù


u`










Ú


U'










ú


u'
Û



U^











û


u^










Ü


U:










ü


u:
Ý



Y'











ý


y'










Ÿ


Y:










ÿ


y:
Þ



|P











þ


|p










Œ


OE










œ


oe
Š



Sv











š


sv













TM










©


(c)
®



(r)











§


SS










ƒ


/f










ß


|3
`



_`











¡


_!










¿


_?













_,




,,














...













'6













'9




"6














"9










×


_x










÷


-:
¢



c|











£


L-










¥


Y=













_.




_-














__













-|













=|




_<













_>










«


<<









»


>>
¤



o*











¦


||










¬


-|










´


_'
ˆ



_^











˜


_~










¨


_:










°


_o
¸



_,











ª


_a










º


_0










¹


_1
²



_2











³


_3










·


_.










µ


|u
±



+-














%0










¼


1/4










½


1/2
¾



3/4


Perhaps the Google IME could become the software most people worldwide use to enter Unicode, not just Northeast Asians entering CJK characters.

No comments: