Has anyone converted completely to Japanese keywords for RTK studies? I started last year with the keywords from wrighttak, see http://sites.google.com/site/wrightak2/ I was less than thrilled with the conversion and abandoned it part way though. Part of the problem was the word missed context.

My approach to Japanese keywords RTK studies will be RTK lite (actually KO2001 order). In creating the deck I am going to include for each fact:
- hiragana word
- Kanji word
- intersect a high-frequency list for common compounds for this kanji
- Include a sentence from core collection using one of those compounds.
- KO2001, RTK index, RTK keyword (to bring back memories of my story if I forget how to remember this kanji)

My question to everyone, if they have converted, do you have any issues, tips, etc?

I am also curious about this.. I'm a fan of RTK-2, which tests kanji compound -> reading. I thought that at the same time I would make the conversion to Japanese keywords (kana -> kanji) in my RTK-1 deck, but Heisig makes some convincing arguments in RTK-2 that reviews of Japanese vocab should only be kanji compound -> reading. Now I'm not sure if I should do Japanese keywords at all...

I think if you can remember the keywords easily, then it's okay to use them, but I don't think they're any more useful than English keywords. I don't see the point, for the most part.

Some time after slogging my way through RTK, fail rates on kanji cards went way up. Why? I was forgetting English keywords due to associating kanji to real Japanese instead. At that point, maintaining the English connections seemed like an unnecessary chore, so I started over with wrightak's deck since I was too lazy to make my own. ;p

Wrightak's deck has a few inconsistencies, but overall I'd say it's pretty decent. One of it's problems is that as a pre-made deck, every word won't trigger your memory. Sometimes, you might have to substitute with something you feel you can remember better.

About the context problem - I figured I'd solve this by using multiple kana words on the question side and highlighting the part of the word whose kanji I was looking for - but again, I wasn't sure the potential benefit was worth the effort, hence premade deck.

Either way, I guess the goal is to associate kanji directly to words; that is, to burn the bridges instead of spending energy to maintain them. And still be able to write, thus not quit kanji reviews. Also, having the reading is an extra memory cue I find quite worthwhile.

At any rate, good luck in making your deck. Please consider sharing it when you're done. It sounds like it'd be fancy enough to be more useful than wrightak's, which could benefit others who are interested.

That's actually a pretty good reason to switch to Japanese. I've been tempted with a few characters (like 中) that are far easier in hiragana, but so far I've been too lazy. Of course, it's just point and click now, so I've been doing some touchups to the keywords lately.