I recently decided to add a couple of the most common traditional characters but there seem to be a couple of bugs. What I did was enable both trad and simplified, then went and studied only the HSK1 list after resetting it to the beginning. The errors I've seen are:

1) I saw some words added from lists that were not HSK1, I think 1 was from my old queue.

2) I got the trad writing of 老师 as new which was what I expected. But then several reviews later I got the trad definition prompt, but the readiness said it was only 90% ready and I had studied it 8 months ago when I only started adding trad items yesterday. It seems like it is confusing the definition review for the trad and simplified versions?

Thanks for your help!

It will study any words that have been covered in the sections of HSK1 that you've done, not just ones that were first added from HSK1. So those other words you saw were in HSK1, also, but perhaps showed "Added from:" as your old queue?

The way that Skritter treats the items is that there's only one item for a word in terms of scheduling when the answers are the same, even if the writing is different in traditional. It just switches between showing you the simplified and the traditional as a prompt. So you're adding the traditional form to any reading, tone, and definition items you have, and you're adding new writing items.

In the case where the traditional variant has differerent readings/definitions, then you get totally separate scheduling items for those, too, but that only happens when multiple traditional characters have been combined into one simplified character.

For number one, I don't think that is what is happening. I reviewed the HSK sections I've added, and I don't think any of them have the words I saw come up. For example, I'm getting asked for the trad 夜裡, even though that isn't in the HSK 1? and the only place I can see it coming from is a section of the old queue I haven't studied yet? I also noticed because I was getting asked for the traditional version of an extremely obscure word that I think I once added to the queue and haven't seen since. (I can't remember what word it was, but it was for "dealer" like in poker, I don't think that is in HSK1...)

For number 2, it sounds like it is doing what it is meant to, but I'm not sure I agree with the reasoning? I'm trying to learn how to read and write the most common trad characters, but these are also the simplified readings I know really really well, with intervals in the months to years. So I won't get quizzed on being able to read the traditionals for that long? This is less helpful when I'm trying to get ready for a trip to Taiwan in a month. This would probably be a pain on the backend, so if I'm the only one that has this complaint I can just add them to Anki, but it does seem like it is unnecessarily limiting the amount of learning that can be done...

It seemed to us when designing the system that you'd best learn simplified/traditional character equivalence by actively studying the writing, which is different, and that passively studying the readings, definitions, and tones would be a side benefit. Separately studying the reading and tone of 里, 裡, 裏 would be inefficient since they're the same between all variants, but you do study the writings and definitions separately.

It works a lot better when you've started the characters' simplified and traditional forms at the same time, since then alternating the forms on prompt of reading/definition/tone items is much more balanced. You're right that it doesn't satisfy the needs of a long-term simplified learner wanting a lot of short-term traditional practice.

If that's what you want to do, you should turn off simplified reviews for a while and study individual sections, which will pull in even far-off words for review and prompting you with just their traditional forms.

Thanks nick for the always responsive help! Studying one list seems to work just fine. I'm still not sure how those random ones made it in but it seems to have stopped happening.

Thanks!

I'm following the above strategy, setting to traditional then only studying a single list (HSK 2 now). I'm getting a bug where it seems to not remember the answer I just gave? I.E. answer a prompt, then get the exact same prompt again several items later. Like for Liang3, the first time got the writing as a "new" word, answered it incorrectly. Then several prompts later, got asked for the writing again, only this time it was last reviewed several months ago?

Maybe has something to do with too few items to cache?

Perhaps... if you check your network latency indicator in the lower right after finishing one of these words, do the "saving count" numbers go down right away, or do they persist for a while?

If you can send me some screenshots of this happening on the first and then second reviews (I know it's tricky to predict the future, for which items this will happen to), I might be able to figure this out better.

I almost hesitate to distract you from the iphone app...

I'll send you an email with a couple of screenshots, but here is what I noticed. The first time, I'll get asked on just the writing (new), get it wrong. Then several items later I'll get asked for the tone & writing (with the tone interval of 5 months showing for last reviewed). At this point I don't know if the first writing got recorded? If I miss a new word they don't usually show up so fast. If I answer it correct the second time, (since I just saw it) does this count as getting it right the first time? Is this the intended behavior, seems like they should just be clumped once?

Also maybe add which part the "Studied X month's ago" is for?