Hello, I made a Kanji dictionary from Remembering The Kanji based on the components and keywords, its a very useful resource for those who have done the RTK book.

For now the dictionary are 100% complete for Spanish users, but collaborators are needed to happen in English, I want to know if there is some people interested in this project. Other RTK languages like French or German are welcome.

The link is http://www.kanji-sama.com

Regards.

There is already a kanji dictionary based on RTK - KANJIDIC...

Incidentally this is also why KANJIDIC isn't very useful as a kanji dictionary.

For English, your might want to take a look at Katsuo's google spreadsheet. The last column lists the primitives for each kanji. Note that Katsuo included some variations of the names as well. (btw The link is from the first stickied topic in "The Japanese Language" section.)

This is a bit different than kanjidic, Jarvik. This allows people to look up kanji by primitive or primitive combinations.

Jarvik7 Wrote:There is already a kanji dictionary based on RTK - KANJIDIC...

KANJIDIC is just a text/xml file... it contains the Heisig codes but, that doesn't mean it's "based on RTK."

Jarvik7 Wrote:Incidentally this is also why KANJIDIC isn't very useful as a kanji dictionary.

Huh? I've always found it quite useful myself, though of course it depends on the software.

daremo Wrote:

Jarvik7 Wrote:There is already a kanji dictionary based on RTK - KANJIDIC...

KANJIDIC is just a text/xml file... it contains the Heisig codes but, that doesn't mean it's "based on RTK."

http://www.csse.monash.edu.au/~jwb/kanji...tml#IREF07
KANJIDIC is, in fact, based on RTK.

daremo Wrote:

Jarvik7 Wrote:Incidentally this is also why KANJIDIC isn't very useful as a kanji dictionary.

Huh? I've always found it quite useful myself, though of course it depends on the software.

KANJIDIC's uselessness (as a dictionary) is based on its inaccuracy and complete lack of any sort of detail. kanjigen sometimes takes 3+ pages to define a kanji whereas kanjidic usually uses one or two words. KANJIDIC does have useful applications, but not for when you actually want to look up a kanji to find out its meaning.

Thora Wrote:This is a bit different than kanjidic, Jarvik. This allows people to look up kanji by primitive or primitive combinations.

I'll admit I didn't see this function on the site, but most KANJIDIC software I've used does the same thing. Ex: jQuickTrans, JEDICT, Jim Breen's website. The only difference is that they don't do it by primitive name, but visually. Most Denshijisho also have this function, using the readings of the primitives.

[Katsuo]

KANJIDIC is based on the JIS X 0208-1990 specification, and contains 6,355 kanji. (There is also a supplement "KANJD212" based on the JIS X 0212-1990 specification with an extra 5,801 kanji.)

KANJIDIC includes data compiled from many sources. Perhaps 1~2% comes from RTK.

Two things from RTK have been added. They are:
1) Reference numbers from 1 to 3007
2) Most (but not all) keywords from 1 to 2042 (as part of the "meanings" field).

I converted KANJIDIC into spreadsheet format, which makes it easy to select data for personal projects. As the file is large I split it into three parts. See approx. half-way down in this post for links.

*Note that KANJIDIC has been updated since I made the spreadsheets and now includes data on JLPT level and one or two other things. I'll try to add this data soon.

Kanjigen is indeed awesome (I just downloaded an EPWING and got it working on my mac using コトノコ) but being that it's in Japanese, that alone makes KANJIDIC extremely useful. Beginners often need to understand the meanings of kanji and they aren't going to be able to use kanjigen. (Unless there is an English translation of kanjigen that I am unaware of).

Tobberoth Wrote:Kanjigen is indeed awesome (I just downloaded an EPWING and got it working on my mac using コトノコ) but being that it's in Japanese, that alone makes KANJIDIC extremely useful. Beginners often need to understand the meanings of kanji and they aren't going to be able to use kanjigen. (Unless there is an English translation of kanjigen that I am unaware of).

There is no English version of Kanjigen unfortunately. The closest thing is the New Japanese-English Character Dictionary by Halpern. It is also available in electronic form, but I've never seen it on any 闇サイト. Kodansha's Kanji Learner's Dictionary is also available in electronic form, but only on denshi jisho afaik.

Thanks all for the comments, I'll fill the English database whit the Katsuo's google spreadsheet, then anybody can use it and propose fix to primitives if needed.

Any other comment are welcome.

[Katsuo]

Note that this file (the one Thora linked to above) does not list primitives in the same way as in the RTK books. It's intended as a quick look-up and often includes sub-parts (i.e. parts of primitives) as well. It also includes some names that RvK members made up, such as "Spiderman" and several of my own (e.g. I call 力 "Arnold"). Those extra names should not get in the way, but they do increase the amount of space (or number of fields) needed.

In my own database I also include Japanese words for radicals (primitives). So for example, when I want to find kanji containing 糸 I can type in either "thread", "spiderman", or "ito". I have just now added that radical data to the same file in case anyone finds it useful.

The radical data is not complete, but I often find it helpful, e.g. when it is shorter than the English word, or when I've forgotten the primitive name, or just because it's the first thing that comes to mind.

To summarize, the data is not intended as a reference, but as an aid to quick multi-look-up. It could do with tidying up, but may be useful to some in its present state.

Thora Wrote:This is a bit different than kanjidic, Jarvik. This allows people to look up kanji by primitive or primitive combinations.

Jarvik Wrote:I'll admit I didn't see this function on the site, but most KANJIDIC software I've used does the same thing. Ex: jQuickTrans, JEDICT, Jim Breen's website. The only difference is that they don't do it by primitive name, but visually. Most Denshijisho also have this function, using the readings of the primitives.

But I think the reason for Montefori's lookup is to be able to lookup by RTK primitive name. None of the other resources offer that afaik.

While RTK primitive lookup is useful for a number of reasons, I think beginners should also try to get familiar with looking up kanji visually using the more universal radicals/elements. RTK primitives don't always correspond to those element/radical shapes. (Katsuo's lists include a comparison.) Here's two element lookups:

wwwjdic's multi-radical kanji lookup. Contains about 252? (> traditional 214 radicals). Cons: sorted by stroke count only.

Ben Bullock's multi-radical kanji search by location. Pros: Sorted by location: top, enclosure, left middle. Then by stroke count. Then by frequency. Resulting kanji appear as each element is selected. Contains about 243? (> traditional 214 radicals.) Cons: fewer elements?

Kanjigen isn't suitable for learners. All the etymological and obsolete usage stuff could confuse even advanced learners. Probably even Japanese people. =] While it's great that Mr Killjoy ;-) reminds us of kanjidic's limitations (and suggests more suitable options), kanjidic remains a great free resource for learners imo.

Back on topic - I hope people will help tidy up the entries in Montefori's English spreadsheet. Incentive: Approaching the kanji from this different angle (identifying the primitives) will help solidify your kanji memories!

Hi, Katsuo's database version is done and 100% working, but I still want to do an official RTK database and put the 2 versions available, because it can help students to remember keywords of composed primitives and its the mode that search engine was designed.

While in actual DB version you can find any kanji in 1 search whit some primitives, whit an official DB you need to search first the composed primitives (if you only remember basic primitives) and then go for the kanji, should be more didactic and its still pretty fast, whit practice can be very accurate.

I agree whit the importance of understand the use of a traditional dictionary whit radicals, elements, stroke counts and all those things, but its bit hard to use also for a Japanese. I like Heisig's philosophy about simplicity over the traditional metods, but I recognize that Kanjidic or others like it should be the bible for a very advanced student.

Comments and corrections are welcome, search my email in Kanji-sama.