DrJones Wrote:I'm already over 3000 readings in iiKanji, and I think I finally formed an opinion on the tool. I think you might not like to hear it, but I'll say it anyways.
We are open to any comments, as long as they are constructive.
DrJones Wrote:The order is suboptimal.
This is a bold statement. I hope you have substantial arguments to support it.
DrJones Wrote:While it lists the words according to their frequency in japanese texts, that is only "optimal" if you are supposed to stop using the tool at that point.
iiKanji's purpose is to teach the jouyou readings of the jouyou kanji, which form 94% of Japanese texts. Take it or leave it.
DrJones Wrote:If your purpose is to learn all readings, the order is far from optimal.
By "all readings" you mean jouyou + non-jouyou readings?
DrJones Wrote:Heisig users also use a wacky order to learn the kanji, but that's because the goal is to learn all of them. By ordering for frequency, you lose all the benefits of learning the readings using a logical order (by minimizing the number of "building blocks" at each step, and exposing the patterns that give internal consistency to the japanese script).
Sorry, could you explain in detail what you mean by "logical order"?
DrJones Wrote:The tool isn't as helpful as it could be, because it explicitly avoids to tell the user any relation that can exist between words and between kanji.
I don't understand what you mean when you mention the relation between word and kanji. Do you mean between meaning of the word and meaning of the kanji?
DrJones Wrote:For example, using this tool the user will know that 図 is read sometimes ず and sometimes と, but will not know how to differenciate between the two readings.
Do you mean that the user will not know which of the readings to use when he sees a new word containing this kanji? This, unfortunately, doesn't have any other solution than learning every word one by one.
iiKanji aims at allowing you to learn the building blocks (jouyou readings) as quickly as possible so that you can use them to memorize new words more easily.
DrJones Wrote:Knowing that 図 is read と could also be helpful to learn the reading of 斗, but the tool doesn't let you see that relation, so it takes longer time to learn, more "brain space", and makes it easier to forget (because there is one less "anchor" to remember the reading).
I tried using the radicals as an anchor to learn the readings of the kanji using RTK2, and it didn't work for me. There were too many exceptions. That's why I came up with iiKanji.
DrJones Wrote:The tool is redundant and ambiguous because it strictly limits itself one reading per flashcard, even when two would be better (this is also a mistake that Heisig does). 夜 can be read よ and よる and have the same meaning, there's no penalty in learning the two readings at the same time, however you create a trouble by having each reading in a different flashcard. Even when there's no risk of confusion, merging flashcards as for example 語る and 語り would kill two birds with one stone.
Limiting ourselves to one reading per flashcard is the core idea of the method.
I think that merging phonetically similar readings for a given kanji would not be beneficial (ie, ゆれる ゆらぐ ゆるい and the others for 揺), but let me know if you find some for which you think it would definitely be beneficial.
DrJones Wrote:The tool also includes plenty of rare readings for words that you'll likely never meet in a japanese text, ...
Rare readings only appear in rare words. If you want to learn up to the rarest jouyou kanji/readings, you will inevitably have to deal with rare words.
DrJones Wrote:...while there are a lot of words that are commonly used but whose kanji is no longer used and/or not on the list
Could you illustrate with an example, please? I think I'm not following you.
DrJones Wrote:If the purpose is maximizing the amount of japanese text you are able to read at a given point, this tool would have better results if it didn't restrict itself to the jouyou and all its readings.
The jouyou readings of the jouyou kanji give a coverage of 94% of all Japanese words. I personally think it is not a bad coverage at all.
If you want to actually read Japanese texts, you will inevitably come across some words which do not have jouyou readings (the remaining 6%). To deal with it, we are considering to slip in some frequent words whose readings are ateji just to increase this percentage a bit more.