Apologies if this thread has been covered -- I've been away for a bit.

I'm learning mostly words. There are several kinds of mistakes one might make, but I'm curious as to how skritter's scheduler actually uses these to schedule words.

Specifically, take a few examples:

橘子

I have difficulty getting the 橘 right. However, I don't really want to enter the 子 (even though it takes very little time). However, as far as I know Skritter will always present the second character.

隔壁

A more extreme example. I have difficulty with 隔
but not with 壁. If always presented with the full word, then I am wasting time (quite a lot)...

任务，程度

Here, the characters are totally fine, but I forgot the spelling (for example, I confused 任务 and 人物 and tried the second character as 物）. I would be annoyed if this mistake led to 服务 being scheduled.

My guess is that Skritter treats words as indivisible wholes. If so, why ask to grade after each character?

阿福


PS I just went to Korea and Japan, and found it very amusing that Koreans, Japanese and Chinese pretty much had to use English to communicate with each other :-) Also, knowing 汉字 was useful in Japan, but not in Korea -- everything in Korea written in Hanja was also written in English...

Welcome back, 阿福! Korea and Japan--jealous.

If you don't want to write 子 or 壁 in those words, just press space, right arrow, or enter to skip it, or click the next button. It'll be the same as if you'd written it.

Characters themselves are graded when you're doing a word, too. So if you're writing 隔壁有耳 (gébìyǒu'ěr: "the walls have ears") and forget the writing of 隔 and the tone of 耳, then the writing and tone of 隔壁有耳 will be scored as a 2, and the writing of 隔 and the tone of 耳 will be scored as a 1 and come back first, since you need to study those more than the whole thing. (But if you haven't added 隔 or 耳 individually, they aren't activated and won't be prompted, even though their dueness is still tracked.)

I'll yet be changing it such that if you get an easy character you know really well wrong in a word, then it blames the word and just doesn't submit any review for the character. It does this already for reading practice, but I haven't applied it to writing/tone practice yet.

We originally had it (long ago) such that it'd only show you the characters in a word that were due themselves, but that led to problems. I've been thinking of ways we could bring that back as an option: to automatically skip the characters you know much better than the others in words you have seen before. Is there interest?

I quite like having to repeat characters I already know. I see it as a sort-of built-in focus on higher frequency characters. And I think that repeatedly thinking about a character in different words helps me understand it's meaning better than just reading a single-character definition.

"We originally had it (long ago) such that it'd only show you the characters in a word that were due themselves, but that led to problems. I've been thinking of ways we could bring that back as an option: to automatically skip the characters you know much better than the others in words you have seen before. Is there interest?"

Not from me. But far less automatically calling for a character review would be good (maybe the onus could be reversed somehow..). Presently, if I get the last character wrong on a writing prompt with no tone, but don't want the character reviewed alone (99% of the time), I will click "3/green/good", it'll load the next character, I'll click back, toggle the word review, click through to the next one. I do this a fair bit and it's one thing that could be streamlined... At the same time I think the "no extra click" is on the whole a good thing.

To skip the extra forward/back, Byzanti, you can use the correct button (or "V") to toggle the correctness instead of the grading button. But yeah, toggling the grading indicators in the prompt needs streamlining.

Ah, thanks for the tip!