Modern Chinese IME's use the pinyin-based predictive input methods (#1) as the primary method, together with a wubi-style method (#2) as a secondary input metho

Modern Chinese IME's use the pinyin-based predictive input methods (#1) as the primary method, together with a wubi-style method (#2) as a secondary input method. E.g. in the Sogou IME, if you type the 'u' (which can never begin any pinyin syllable), you can then type letters to denote strokes or common components, such as h for heng 一 , s for shu 丨 , p for pie 丿 , k for kou 口 , s for shui 水氺 or 氵 , etc. Can be useful to distinguish between characters for commonly used syllables such as 'yi', 'ji', etc. Regarding #5, I found it's difficult to switch from normal pinyin to shuangpin (double pinyin) in one go, and none of the shuangpin methods or software provide staged switching. E.g. a shuangpin IME could allow me to use 'i', 'u' and 'v' to type 'ch', 'sh' and 'zh', but continue to use the existing vowel combinations (e.g. 'ue', 'ao') for stage 1 pinyin/shuangpin typing, and then offer at stage 2 shortcuts for some of the vowel combos only when I had gotten into the habit of the stage 1 shortcuts. Of course the choice of shortcuts available would then be constrained by the staged learning path, but because there's so many different shuangpin IME's available, each with differing key assignments and none of them standard, there's probably still room in the market for a new key assignment that caters to such staged learning.