Hey, I've been using Skritter to study for my Japanese exams, and I'm definitely a fan of the "hide reading" option.

However, when reviewing a compound, it only hides the reading until you complete the first kanji. Is it possible to hide it a bit longer?

I've been playing around with this to try to fix it, and it doesn't seem very helpful to do what it does for Chinese, which is to show the reading for a character as soon as you're done. Most of the words with multiple kanji have the readings of the kanji mushed together, so Skritter can't tell which kana belong to which kanji. Are you suggesting that it not show the kana for any of the kanji in a group until the last kanji of that group is done?

(I've tried hooking it up so that it only hides each kanji group until you finish it, but it hardly ever fires, since there are so few words where you have kanji kana kanji kana in the writing.)

Just take the first few words that have multiple kanji from JLPT Level 2 (2009) section 5. I can easily see that,
営業 = えい+ぎょう
学部 = がく+ぶ
委員 = い+いん
主婦 = しゅ+ふ
機能 = き+のう
基本 = き+ほん
期間 = き+かん
実施 = じっ+し
For 営業, is it OK to show only the えいafter 営 is entered but withholding ぎょう?

Oh. To do those, I would implement logic which looked at the onyomi for each component character, converted them to hiragana, and tried to claim more specific kana strings. It wouldn't always work, but it could be done.

How good would it be, though?