しあわせ » 02 reading • 読む事

Kanji Sieve 0.3 Mac

ロバート — Mon, 14 Jun 2010 12:11:50 +0000

Kanji Sieve for Mac v0.3 is ready. Unfortunately due to pressing matters elsewhere this is about as far as I can go for now.
Unavoidably that means it could be a few months before a Windows version appears.
Rather than sit on the Mac version which is functional, now that I have permissions for the Chuta dictionary and Flashcard Deluxe features I decided to upload it.

The windows version has a huge problem at the moment. A core screen is just not rendering properly no matter what I try. Ugly I could live with but this is completely unfunctional, the css doesn’t render properly and the Japanese becomes mojibake. I *think* it is due to the Internet Explorer version I am using under XP, but until I can explore this under Vista or Windows 7 I’m at a complete loss. If I can find the time I’ll do my best to get a Windows version working. Although I’m now leaning towards just cutting this function from the windows version rather than chase after workarounds in Internet Explorer.

So here we are at version 0.3

What does Kanji Sieve do?
It takes a piece of text entered by the user and first it breaks down the kanji usage to show you which kanji are used by grade and how often they are used in the text. Then it submits the text to chuta.jp. Primarily this is so the text can be parsed to allow individual words to be extracted into a list. Using online dictionaries the user can then put meaningful definitions to the entries in a wordlist. Finally a tab-delimited flashcard deck is generated that can be used with the iOS app Flashcards Deluxe. If multiple pieces of text are entered a listing of the 20 most frequent unknown kanji can be generated, unique to the user and the pieces of text they find interesting.

My thesis is that the 1006 kyouiku kanji are the best target for Intermediate JSL students. The grade school groupings break these down into manageable chunks. If you follow the grade school groupings there is also a wealth of Japanese learning material available. Keywords are only a tiny part of learning a kanji. Indeed thinking about it as learning to read and write, rather than learning kanji would be much more useful. That said you need to learn by reading and using Japanese. Reading targeted at your interests, providing you with a vocabulary you might actually use. SRS systems can help but words and characters need context and to be meaningful in order to be remembered and used properly.
Kanji Sieve may help with this part of the process.

I have ideas about the process of learning individual kanji, which I hope to cover in a post some day.
I feel it needs several elements.

You need to find meaning in the characters – by breaking them down and understanding their structures.
You need to break the task down into manageable chunks. – 2000 kanji isn’t a manageable chunk.
You need a variety of methods and tasks – to stop you getting bored or only being able to function in one method.
You need to find it immediately useful – to be able to use words in real life situations.

I am very grateful to Professor Nakamura for giving me permission to use the Chuta web dictionary in this solution. For the full Chuta experience go to チュウ太のweb辞書, where you will find many more language options than are available in Kanji Sieve and I also recommend the Reading Tutor site
Also thanks to Ernie the maker of Flashcard Deluxe for permission to incorporate a link to his app.

Kanji Sieve Mac 0.3 runtime download page
You will also find a movie demonstrating the solution on this page.

FileMaker Kanji Project – progress 2

ロバート — Tue, 11 May 2010 16:02:20 +0000

Now with added Kanji Sieve. Or rather the way Kanji Sieve has evolved means I’ll be incorporating my Kanji NoteBook project with it.
Unfortunately I think the direction I’m heading in means an awful lot of work on the interface. And a lot of time I don’t have to spare at the moment, but I am working on it and am excited about the way it’s progressing.
The next Kanji Sieve will have multiple records, so you will save you pieces of text for future use. I’m dropping the keywords as completely futile for reading or understanding a text (although useful when first learning a kanji. I have a blog post in progress about this). I’m fairly confident of parsing text automatically to build word lists. This is why I’ve decided it makes sense to incorporate Kanji Notebook with it, if not in v0.3 at least by v0.4.

I will also automatically generate and manage lists for a flashcard system. Primarily as a tab separated text file so you can import to the tool of your choice rather than my trying to re-invent something that others are doing quite well. I have a basic flash card solution but as I have other tools that work better for my purposes I won’t be pursuing it.
I’m thinking about how to aggregate statistics across the database. I would like to rank unknown kanji and words by how frequent they are overall in the pieces of text you store in Kanji Sieve. This would mean needing a listing of known kanji and vocabulary that can be filtered. That might then give an indication of what words are personally more important to learn. Again maybe one for v0.4 when Kanji Notebook is incorporated.

Most of the functions I want are in place. I can make it work for my own needs but I need to make it work for other people before I release it. I also have to think now about how data will transfer across successive future builds.
In the meantime here’s a pic of the NoteBook.

Japanese on an iPod Touch

ロバート — Sun, 02 May 2010 17:15:06 +0000

I finally bought myself an iPod Touch about a month ago. I wanted something to allow me to use electronic flashcards on the move. On the train I can use my DS and Kakitorikun but it’s impossible to use pen input properly on the bounce around tube journey. With the Touch I’ve more than doubled the amount of time I can study on the otherwise mostly unproductive daily commute, although I read a bit less as a consequence. I also find myself using it at other spare moments, a couple of minutes here and there going through flashcards.As a device it’s almost what I’d hoped for since my first Palm device by Sony. Lot’s of storage, lot’s of good apps, a good screen, small and light it easily goes in a shirt pocket, good battery life, a ~~good~~ reasonable price as long as you don’t need the phone or 3G functionality.
The oversights. No pen input. For kanji learning and quick notes and just drawing a pen is indispensable. A finger just isn’t good enough; it’s like a blunt jumbo crayon compared to a 0.1 Rotring pen. That said the multi-touch interface is nice, except for all the erroneous button pushes and link clicks. And you can forget about ever using a stylus on the current screen. If they are all like the one I bought it’s not much of an improvement over a finger.
What I dream of now is for an iPod DS. two screens in a clamshell. One multi touch the other pen input.
With a user replaceable battery while I’m dreaming.
I don’t care about lack of Flash. This seems to be the new version of the OS wars. That the OS is locked down by Apple is of more concern. At least I’d like to see more interoperability between apps beyond cut and paste. I’ve been using an ad blocker for so long on my MacBook that it’s a bit ugly seeing all those ads on the web again, and there sure ain’t going to be any blockers for browsers on the iPhone OS.

The core of the experience is the Apps. Odd given that the iPod is touted as an MP3 player and the iPhone as a phone but really they are the Newton finally realised albeit without handwriting recognition.

Anyhow what’s on my Touch?

Flashcards

My first stop was to get flashcards going.
Anki was more or less dismissed. I don’t really like the somewhat unfriendly linuxy desktop application and without jailbreaking by all accounts the iPhone experience is limited. That said a lot of people like it and you may have a better experience with it than me. (Anki is good for is getting my lists out of smart.fm though)
I was very disappointed at the functionality of iFlash Touch (free, but really only an extension of the desktop app). It’s very basic compared to the desktop application. About the only good thing I can say is the sync between the desktop and the iPod is easy. It’s a shame because I like iFlash but I think the developer is too busy with his life and studies to keep up with the runaway iPhone market.
I looked at Mental Case, which has a Mac desktop client, but I didn’t want the extra expense and something didn’t quite click with me, too many bells and whistles perhaps. It does have trial and lite versions so it’s easy to see if it’s a good fit for you.
There are many I suppose dedicated Japanese flashcard programs but I find it very hard to sort of anything of quality. I’d prefer to come up with my own lists. What wanted was straight forward import and export. An SRS system preferably a bit more sophisticated than a basic Leitner card system. Text styling. Pictures and Audio support would be a bonus, I don’t use them at the moment but you never know.

Flashcards Deluxe (£2.40 , $4, 450円, €3)
I went with Flashcards Deluxe in the end. It’s fairly simple but still full featured.
It has support for three sides per card and two categories and a flag.
You can colour cards and style the text. This is all achieved with html tags so it is easy to set up your list in a text editor.
Import is via a WiFi link to a server. (Apple won’t let anyone sync over a USB cable). The author provides a webpage and use of a server for you to upload your cards. If you want you can also share your list with other users. He also gives details how to set up your own server. Export is by email or back to the server.
From the app itself you can access Quizlet.com if you don’t want to make lists yourself or duplicate effort in making lists from popular texts. Although due to sheer volume I wonder about how useful quizlet is and what quality the lists are.
You can also make and edit lists within the iPhone app. It’s completely standalone really.
It’s drawbacks. Well it could be I’m not used to the iPhone OS yet but in Flashcards Deluxe the settings and interface seem a little bit sparse and disorganised. Replacing the ? button for a cogwell icon would be a start, and reserve the ? for when you are giving extra information about the app or interface would be a start. That said after a little exploration I’ve found out most things.
It could also do with a Desktop client. However I have an idea about this for my Kanji FileMaker projects; after all it’s only a webpage away.

Smart.fm (free)
This is the other flashcard app I have installed. I liked Smart.fm when it was still iKnow. Now I don’t use it so much after the redesigns. Mostly it’s just too laborious to make your own lists and almost impossible to export them. (Although you can export via Anki and making a list from a textfile is in progress) The iPhone app is nice but has limited offline functionality, and is only a subset of what is possible on the smart.fm site.
There is another app (unaffiliated with Smart.fm) iKnow Touch (£1.20) that uses API’s that smart.fm publish to download entire lists that looks a bit more interesting, as it doesn’t rely on a constant connection. I only found it while checking links for this post.
–update on iKnow Touch– I’d save your money. iKnow Touch downloads your lists, (and the audio card by card), but that’s all. You can read them but not test yourself on them. I don’t need to sync with my progress on iKnow but it’d be nice to have flashcard functionality, or export, or copy and paste, so I can use them in another program. It’s also the first app that’s crashed on me. I can’t recommend this.

Dictionaries.

For most uses I prefer electronic dictionaries over print versions. Lighter and easier to search. As long as the battery holds out and you don’t drop them.

Kotoba!. (Free).
Can’t beat free as a price. Kotoba! uses JMDict data as do most of the Japanese dictionaries and applications you see in iTunes and on the web, especially the free ones. So mostly all that differentiates these apps is the interface or price. Kotoba! does everything I want for now, why pay more for JMDict or Edict data.
I may succumb to the uninspiringly named Japanese later on, it looks to have better cross references and a cleaner interface, but the £10 price is offputting. Across the store 60p and I’ll buy without much thought; £5 I’ll probably buy if it fills a need and looks like quality; £10 and I really stop and think; over £15 and it needs to be extra special. The immediacy of the App Store is dangerous, all those costs add up.

MSDict Beginner’s Japanese Dictionary.
( £6, $10, 1200円, €8 ) This is a good implementation of the print version the Oxford Beginner’s Japanese Dictionary, including its essays. Edict is fine for what it is, but it gives no guidance or differentiation. What it lacks is professional editing and compilation. I’m willing to pay for that and £6 is very good value for an electronic version of this dictionary. I already own the print version and thoroughly recommend it as a first E-J dictionary. It’s limited in its entries but should have most of what you may need as a beginner but more importantly it gives good examples and usage guidance. As a J-E dictionary it has the problem that you need to know the reading to use the index. You could use the Chinese ~~handwriting~~ fingerpainting recognition, but it is sometimes hit and miss. It would be nice to have multi-radical search for kanji. There are also hotlinks in the entries to allow you to explore cross references and it has a history function.
There is another version of this dictionary in the store that is the same price, has a nicer icon, but I think has put in English audio instead of the articles from the print dictionary. I’m unsure. I can speak English so the audio is not really any use to me and misses the point of a dictionary primarily designed for English speakers.

Kodansha Kanji Learners Dictionary. (£15, $25, 2900円, €20)
I usually call this KKLD but for some reason officially it’s KALD. No idea what A stands for. It is one of my favourite print dictionaries and I was delighted to find it in the App Store, it was released in late April. This has the best interface of any of the electronic dictionaries I have on the iPod. It’s a very well made implementation of the print version. It includes all the essays and appendices as well. It looks wonderful. All the kanji are hotlinked making it very simple to explore the compounds. It still has SKIP, which was a good invention for paper dictionaries, but to be honest only Jack Halpern’s dictionaries use it. Now that it’s electronic there’s no good reason not to have multi-radical as well. There isn’t a radical index which is a pity, nor are the numeric indices searchable which is a pity as I can’t quickly cross reference with Edict as I can with the paper version. There is an English index which is a huge bonus. A nice feature is you have a choice between romaji and kana for the readings, which was always a criticism of this dictionary especially by the “romaji is evil” camp. The quality of editing in this dictionary beats any other kanji reference currently available on the iPhone platform. It is expensive though in the App Store ecosystem, but compared to the print version it’s a bargain.
––update 17Jun10––
This seems to have disappeared from the UK, Japanese and US app stores. I’ve no idea why.
I hope it hasn’t been withdrawn permanently it’s an excellent dictionary and I’d like to see it supported. (I’m showing an update I can’t access in iTunes) I only hope iOS4 doesn’t break it now.

Eijiro Touch. (£5.50, $9, 1000円, €7)
Not so good. No copy. No hotlinks. The barest of interfaces. A fairly useless flashcard system. It’s a pity as I like the Eijiro dataset. I thought this looked like the best of the dictionaries on the App store that use it as the others looked overly gaudy and unprofessional. For now I’ll just stick to Eijiro on my MacBook where I can cross reference it and copy and paste. Maybe this app might improve and in the meantime I don’t want to throw good money after bad to explore the other eijiro apps on the store.
––update 25Jun10––
And now this has gone from the App store, and stops working under iOS4. Good riddance.

Learning Programs

Many of these seem a bit cheesy or are at too basic a level. By and large I’ve ignored them.

Japanese Sensei Deluxe. (£9.50, $10, 1200円, €8 )
This caught my attention though. On the face of it it’s a custom flashcard app, that introduces words ten at a time, in no discernible order other than relative difficulties over a couple of hundred lessons. So far so blah. What earned my money was the audio and example sentences. Every sentence and word has a native speaking it. Every word had an example sentence. It is all searchable like a dictionary. That alone would have convinced me. As a bonus there are some nice review games. A multi choice quiz, where the audio is also used. A matching pairs memory card game. Sentence building from the word elements. Select the correct particles game and a draw lines to match pairs game. These provide necessary variety to my flashcard studies. The interface is ok although the reviews criticise it, I think the programmers are looking for a designer to spruce it up a bit. I’ve come across some audio doesn’t quite match what’s written though which is a bigger problem. It’s rare though. Sometimes it’s a different particle. Other times the complete sentence is different from what’s printed. Both are always correct though, it’s not bad data, just oversights in matching up the thousands of sound samples. The other niggle is any word you look up is added to your flashcards, I’m not sure I want this, at the very least I’d like to be able to edit what has been added.
I bought the full version as I wanted to use its potential as a dictionary. It does have a free lite version, where you can buy the extra levels from within the app.

Fun and Miscellaneous

路線.Locky (free)
This is a very clever graphical representation of timetables for the Nagoya underground. I go to Nagoya and I’m not sure how useful this information is because the trains are so frequent but I keep this just because of it’s uniqueness.

駅.Locky (free)
Crowd sourced train and station information for Japan. Most of it is offline but you can download and store it on your phone. Obviously all in Japanese and quite hard to go through at times, but with a bit of planning it might be useful on a trip around Japan.

Shueitai Type Specimen Book (free)
This is a reproduction of an old specimen book. Doesn’t do much but the typefaces are beautiful especially the kana which tends to be much more expressive than the kanji typeforms.

Stanza (free)
An ebook reader. It seems full featured although I’d prefer to scroll rather than “turn” pages. Reading is doable but not ideal, although I used to read on my Palm and the iPod has a much better screen. Check out the free versions of Lafcadio Hearne‘s first hand accounts about Meiji period Japan as in-app downloads from Project Gutenberg. You might also search out “hana” in Feedbooks if you’d like to attempt to read something in Japanese; also available for download in-app. There are also translations of famous Japanese authors such as Natsume Soseki’s Bothcan.

Kanji Sieve v0.2

ロバート — Sat, 10 Apr 2010 18:39:56 +0000

It has taken me a little longer than I thought to get to version 0.2 of Kanji Sieve. Mainly due to getting it to look better cross platform and avoiding problems for a user that wouldn’t be an issue for me as the developer.
However, as someone actually downloaded, looked at and commented on my initial little solution I looked at Kanji Sieve again. A little encouragement will always prompt me to continue projects.This time I’ve taken a bit more care over the look for the Windows file. On the suggestion of Tom Hodgers I used the Meiryo font and allowed the user to change the font size of the sample text.

I added non-Jyouyou kanji and katakana words to the sieve. This may be a bit indiscriminate. What I’m doing with katakana is searching for runs of katakana and hoping these are words. They may not be. For non-Jyouyou I try to eliminate all roman characters, kana, Jyouyou kanji, and punctuation. What’s left over in a Japanese text should be non-Jyouyou kanji. Again strange punctuation and foreign characters may appear here. I do have some plans to try to refine this panel though.

Trying it out I was surprised at the amount of non-jyouyou a friend of mine used in her mixi diary. I would have thought a larger amount of kana and jyouyou kanji in a personal diary. I wonder if it is due to using a word processor, it’s easier to generate those kanji and presumably she can expect ordinary friends to read them easily. If she was writing by hand it might be different.

Lastly I incorporated a little hack I put together to replace kanji with keywords. I did this to demonstrate how little meaning you get from just keywords. Especially when the most popular keywords in English that appear as the first entry in Kanjidict are a bit dreadful at times. These panels may or may not survive into the next version. If my notebook ever sees the light of day I’d generate the keywords from the users input which may at least be more useful and perhaps generate an xml file with the keyword furigana as pop-ups.

Further plans
I’d at least like to solve exporting. At the moment I have an issue with the flow of records of unknown length in printouts. It may just have to be an xml export.
I may make it into a multi-record solution.
I also found something very similar at the reading tutor web site at Tokyo International University. Which has the added benefit of producing custom glossaries for articles. If I could understand how they can parse for individual words I’d implement this myself.

Download from my new permanent Kanji Sieve page.

––update 11Apr10––
I’ve corrected the oversight I made in not filtering for half width kana or full width roman characters. non-Jyouyou and katakana should work a bit better now.

Kanji Sieve – Analysing Kanji Usage

ロバート — Fri, 26 Feb 2010 18:20:10 +0000

This is a little FileMaker solution I’ve written.
It takes a piece of pasted Japanese text and analyses the kanji contained in it.

I wrote it as a quick and probably imprecise way of looking at kanji usage in texts. Probably because of the 1998 study of kanji usage in the Asahi Shinbun (Shinbun denshi media no kanji, Senseido, 1998) usually a figure is quoted of 1000 most frequent kanji account for 95% of usage. I have also seen this as 1000 characters allow you to read 95% of articles (a subtle difference) but I think this is a bit of an overstatement, (the thread below suggests 1900 kanji in order to read 95% of compounds). While doing a bit of research on this I came across several other frequency studies and an interesting thread where Jim Breen notes

…a discussion at a language teaching conference in Japan I attended in 1999, where there was general consensus that
the average Japanese adult could read 700-800 kanji…

Although I find this a bit hard to imagine, write by hand maybe…What interests me is the percentage of kyouiku kanji that are used in texts and which of the remainder of the jyouyou kanji are used most frequently.

My hypothesis is that the kyouiku kanji are a better medium term goal for JSL learners than the complete jyouyou set. The diminishing returns in terms of effort on the 939 kanji beyond the kyouiku kanji might suggest approaching these on a need-to-know basis. The old canard (by Heisigists I suspect) is that leaving out 10% of the alphabet isn’t a good idea. I don’t know. Firstly a more accurate analogy would be around vocabulary and it’s not so much that you completely ignore them but that it is possible to work around the unknown characters. And there’s a world of difference in effort between learning 3 characters and learning 939 characters. But I digress.
The Asahi Shinbun also probably isn’t the most read source by JSL learners either. It might be good to have some statistics on Amazon reviews, mixi blogs, or manga.

Kanji Sieve filters for the six primary school grades and for the remaining jyouyou kanji. It then counts the occurance of each character. This might allow you to see the most frequently occurring characters in the texts you are interested in.
Characters outside the jyouyou set are not considered.
For readability or difficulty other considerations would need to be addressed such as the vocabulary used, the length of compounds and the grammar.

If I continue to play with this I would like to add an export option, maybe allow you to collect a series of articles and see the aggregate statistics.
I would also like to incorporate it into my Kanji Notebook, to allow you to lookup kanji or add them to a study list or set of flash cards.
I would also like to see if I can extract vocabulary in the same way, but I suspect word boundaries would be an issue there although Rikachan manages it though….

Downloads
most recent version only
Kanji Sieve Page

More Heisig Musings

ロバート — Thu, 07 Jan 2010 20:22:56 +0000

I never learn, so “once more unto the breach..”

I am interested in how people learn and the problems of kanji and language in particular.
The topic rises again and again on forums so I’ve been thinking a bit more about it. Instead of a long post into threads that have strayed and grown too long I thought I’d make it into a post here.
It has stayed in draft form for a long time but between a comment about Heisig on this blog recently and trying to catch up with half finished posts, I’ve revisited it. Hopefully this will put Heisig to rest for me, it becomes a little frustrating to have my kanji studies defined in reaction to a method developed 33 years ago by someone who self-admittedly knew nothing about Kanji or Japanese when he first developed it.

First I think there may be a mismatch with what people are calling having learnt a kanji. For people who have used RTK1 “learnt” seems to be writing a kanji linked to a keyword. Quite an achievement and useful in and of itself. Another group perhaps equate “learnt a kanji” with being able to use it. ie linking it to actual words in Japanese, reading and writing. And yes RTK users will progress onto this at which time they will fully “own” those kanji.

Here may be the difference in style. RTK users are learning a larger chunk at once concentrating on two tasks, writing the kanji and an English meaning. They are deferring putting the kanji to Japanese, with the hope, or certainty, that this will be easier further down the road. Traditionalists, for want of a better term, want to have usage of a kanji all at once for the Japanese they know at a given time. They may have to deal with more at once but they deal with it in smaller chunks at a time, and have more immediate full use of a given kanji. They may concentrate more of kanji they immediately need. I also get the feeling that self-directed learners tend to use the Heisig method and those that attend classes tend towards more traditional methods. I feel it’s a bit swings and roundabouts though.

As far as I am aware, other than anecdotes and opinion there is no evidence or studies one way or another about effectiveness or time taken to be literate using any method for non-natives. (Literacy being able to read and write Japanese to a particular standard; the end of formal schooling before high school, being a reasonable benchmark perhaps.) Also while we seem to be discussing two styles of learning I’m sure there are many more styles of learning a language and many approaches. The problem is finding the approach that will work for you.

This is what I’ve found with Heisigs order. Which I followed for a while. I constantly came across kanji that as yet I have no use for and this inhibits my memorising of them. I have far more success with those I can link to Japanese I already know or see some more immediate use to. Now of course there’s no reason I have to stick to the order, but then I’d lose the advantage of taking something off the shelf rather than having to work it out for myself. Catch22 it seems. But if you take away the strict order and learn components as and when you need them it would be easier to use for kanji you need to learn, for the grade school and JLPT orders, or based on frequency of use. There are diminishing returns in learning kanji. The most frequent 1000 are used 90% of the time.

Another thing I’m noticing is the keywords. I’ve noticed a couple of things here. Ambiguity; I don’t make the same connections as Heisig. For example, Spring can have 4 meanings for me; season, water appearing from the ground, metal coil, the action of jumping. I’ve got to go the usage of the kanji to clarify it. Then I feel fountain gives a better prompt for 泉［いずみ］The other thing I’ve noticed is less common links being made. For instance 頂, “wear on head” an idea which I’d more commonly associate with 冠, made more useful links when I saw it was いただく or ちょうだい. humbly receiving. I suppose in one respect it makes these kanji stand out but it also erodes what confidence and trust I have in the author and I’m back to not being able to use something off the shelf. Nor am I alone in noticing this and I wish Kanjidic would cull the Heisig keywords or at least note them as such rather than having them as the first meaning.

Whatever way you look at it these keywords are giving meaning to a particular kanji. I’d also go further and say the stories are giving meanings to the kanji. Often meanings that don’t exist. It may not be the intention but I feel it is a by-product. This suspicion is one reason I dislike the stories. I’m more interested in the actual derivation of a kanji. That helps me more than making something up. In any case I’m not very good at making up stories like this.

Writing out kanji as the alternative to Heisig is a common misconception. Yes at first it does look like that is the alternative. Especially if you use Japanese classroom like materials with all those practice boxes.
Traditional isn’t write repeatedly and cram readings in (IMHO). Yes on Japanese school books or handouts there is usually space for 12 characters but it’s very important that a child learns neat handwriting. And yes the action of writing helps you learn it. But… The learning of meanings and readings seems more to do with hanging the kanji on example vocabulary the student already knows, combined with a definition and possible etymology of the character. Because of traditional dictionary usage, stroke count and radical are taught although I would guess this might tend to be ability to count and identify rather than rote learning. It’s more based on understanding than on rote. The characters are also broken down into specific elements much the same way as spelling in English. So in explaining 明，it might be described as 日 and 月

That breaking down is one part of the key, yes just as ol’ H does it. I just prefer to use the actual names and not use a story. にちへん、つきへん。If I can I prefer to hang it on a Japanese word. 明るい in this case. The other part is putting it to words rather than trying to remember a complete picture.

A more interesting Japanese approach is in the books by 下村昇, a Japanese teacher whose writings predate Heisig’s. His method uses pictures to explain the etymology of the forms (although he does ignore the difference of phonetic components and semantic components) This at least gives you an understanding of the character. Then having broken down kanji into about 36 basic components and 24 common radicals he can write out in a sentence of about 4 steps how to write any given character.

From an article about EFL I read recently.

Alas, the perfect method is an illusion. A chasing of the wind.

And the problem may be that on the Internet there are very many students and very few teachers. The Internet while of fantastic value is full of amateurs (like me and many blogs and forums) groping for direction in the dark. Often the value of a good teacher isn’t recognised. I think some good kanji teachers are needed.
下村昇 is one, so is隂山英男 (of Kakitorikun fame), and 加納千恵子 the author of the Kanji Book series and a researcher into teaching Japanese to foreigners is another.

Japanese Childrens Kanji Book

ロバート — Thu, 07 Jan 2010 11:32:09 +0000

These books have the usual unwieldy titles of so many Japanese books.
1行読んでおぼえる小学生必修1006漢字—低学年500漢字 and 1行読んでおぼえる小学生必修1006漢字—高学年506漢字. The idea behind them is quite simple though. You read one line per kanji to memorise if not all, at least most of its important readings. It’s similar to books written for Japanese second language learners like JLPT3 Kanji by Examples and JLPT2 Kanji by Examples.

The difference here is that the author crams all the readings into a single and maybe quirky sentence rather than the rather dry single use sentences of normal texts. The drawback being that this can use more complex or less common vocabulary and grammar, which is fine for natives but is a bit harder for learners. However as it’s aimed at primary school children, overall the language isn’t difficult.

The books also don’t fully stick to the grade school order. Book 1 is mainly Grades 1 to 3 and Book 2 Grades 4 to 6 but they are presented thematically and so are mixed up. I’d say it’s a bit more of a revision and consolidation tool in this respect. However by grouping them thematically I feel you make more connections and so remember more. Connections are the key. It’s easier to remember a sentence than a string of readings or even a single word. At the very least it should be a good resource for those that like to sentence-mine.

As you can see from the example below, there are five kanji to a page. Five sentences with the target kanji highlighted with a variety of readings. All kanji have furigana. 5 sentences per sitting seems reasonable. Only time will tell what my retention will be like. Maybe I need something like Smart.fm to help me, but while I am transcribing sentences for myself into my kanji notebook project, I don’t think it’s ok to make them public. It’s a very nicely designed book and not too expensive at 1260 円. If you want there’s even pictures to colour in to chart your progress!

The books certainly get rave reviews from parents on Amazon.

FileMaker Kanji Project – progress 1

ロバート — Sun, 20 Dec 2009 14:00:30 +0000

I’ve started making my Kanji Notebook.

The first week has seen me gather a lot of the basic data I want, some of it imported using XLST such as Kanjidic, some of it input by hand, some from tab separated files.

So I have the data on all the kanji I could want. I decided only to import the data I was interested in so many of the dictionaries and Heisig didn’t make the cut, nor did Spanish, French, Korean or Chinese. I might import this data later and allow it to be toggled on and off. One of the many reasons I’m doing this is that current dictionaries don’t display want I want how I want or give far too much information.

The radicals were only given as a number so I needed to make a table of the radical names and sub-classify a number of them. Then I needed to input an index from Basic Kanji Book vols 1 and 2 by Chieko Kano. I also put in an option to override Kanjidic’s keyword (too often derived from Heisig and too ambiguous) and use your own keyword or one from the Kanji Learners Dictionary if you input it using the index number for ease of lookup.

Next was to relearn FileMaker.
The first big task was to sort out XML input. Within the limited scope of Kanjidic I’m confident of this now.
Then I looked at the new Japanese features. Automatic furigana creation is useful especially in the sentences tables. Tategaki (Top to Bottom) representation of related tables took me a while but I figured it out in the end. Along with this I wanted to colour the okurigana of the kunyomi. Possibly it can be done with FileMaker native functions but I used an external regular expression plugin to achieve this.

The readings themselves need sub-classification into common and uncommon readings and be marked for those learnt in middle school and high school as well. I’m also thinking of omitting the prefix and suffix instances.

As one of my major ideas behind this was to be able to add notes to kanji, something I can’t do in a dictionary or a web browser, I have notes fields and automatic lookup of my favourite kanji resources using FileMaker’s built-in viewer. In time I may also be able to automatically extract the data I want from some of the dictionary resources.

Another plan I have is to be able to link it to my copy of Kanji Odyssey again to allow annotation. I think I can probably achieve this if I can place Kanji Odyssey in a known file path then get the index numbers used.

I also inputted some sentences from another Japanese children’s textbook. This is going to take quite a while and can only be for my own benefit because if I ever make this project public I’d have to delete them.

I put in some photographs I have of Japanese signs as examples of the kanji in use.

Everything is a bit haphazard at the moment. I may end up taking what I’ve learnt at some point and starting from scratch again with a better idea of where I’m going. I need to sort out interface for one thing, it’s fine for my own use but if anyone else ever wants to use it I’ll have to get a lot more organised. I probably need to think a bit more about where I’m going. I want to enable the radical lookup in Doremons Grade School Kanji Book. This uses 24 basic strokes and another 30 common radicals to describe any of the 1006 Grade school kanji. I also want to integrate example words so the same words appear for any given kanji in an example word. (ie. 勉強 would appear as an example for both 勉 and 強) I’d probably draw examples from the old JLPT lists, kakitorikun and various textbooks and maybe allow you to filter which words appear in the list. I’d also like to look at an iPod Touch version using FMTouch.

So lots to do. I wonder if I shouldn’t be spending the time actually studying….

Kanken Result 2009

ロバート — Sat, 19 Dec 2009 12:40:58 +0000

As can be seen from the above certificate I passed 9th grade in the Kanken test. やった！
I was reasonably confident however. Although I can be a little proud of my achievement it only puts me on a par with a below average seven year old, with knowledge of 240 common kanji. However I can say that the knowledge is native level. I can read the onyomi, kunyomi and tokubetsu readings and write them by hand, not just put an English keyword to them.

You get a lot for the £23 exam fee. There was the test itself that had to be hand marked and this morning I got a packet with

my certificate to go on the wall
two certificates that I think are for a school or work file
an answer sheet for the test I took
a marking sheet showing the answers I got correct and incorrect
my ranking relative to the average score on a spider diagram
an serial number and password (also in 3D barcode for a phone) to log in to the kanken site
comments and advice

And the time frame for the result is pretty good after all. I took it on the fifth of November, it was awarded on the 24th, presumably then sent from Japan and posted onward from London yesterday.

So in the end I got 136points out of 150. 120 was the pass mark. My reading was a bit below the average though. Mind you the average was between 95 and 100 percent in all the areas.
So onwards and upwards towards level 8 (440 characters) next year, hopefully in the summer rather than to have to wait a whole year. My medium term goal are the 1006 grade school kanji at level 5. Maybe above level 2 would be a lifetime goal difficult even for Japanese. The ultimate level, level one, for instance tests 6000 characters. I doubt I shall ever see such giddy heights.

I think this test is a good adjunct to JLPT. Although primarily about kanji, as it is aimed at native speakers and is all in Japanese it will test your reading and writing abilities to their limit and you can be sure they are on a par with Japanese students at the same level.

I heartily recommend this test to anyone who can find an exam centre. Unfortunately that usually means Japan. But if you cannot find one maybe you could see if your school or college would offer it, I don’t think it would be that difficult for them to apply to to it. It might also be possible for them to get assistance from the Japanese embassy’s cultural attaché.

New JLPT specifications

ロバート — Fri, 11 Dec 2009 12:28:21 +0000

I read an interesting document about the new JLPT specifications for the test starting in 2010 from the Japan Foundation.

Considering the search terms people use in Google these points are a big departure from the old test:

There are no plans to publish collections of complete copies of tests administered in past years.

and

The goal of learning Japanese is not to memorize vocabulary, kanji, and grammar, but to become capable of using them as a means of communication. The new test is to measure both “Japanese language knowledge, including vocabulary and grammar,” and “the competence required to perform communicative tasks using language knowledge.” Therefore, we determined that it is not appropriate to publish “Test Content Specifications” which includes the lists of vocabulary, kanji, and grammar.

and

Failure to exceed the minimum acceptable score in any scoring sections will result in a fail for the entire test, even if your total score is above the minimum acceptable score.

While the old levels 1 through 4 map more or less onto the new N1, N2, N4, and N5, figuring out N3 will be a challenge in the beginning. I wonder if people won’t just study for the old level 2 and take N3 if they don’t feel confident.

I suppose the many and varied JLPT sites with lists of kanji and grammar points will still exist. Maybe many unchanged. Publishing companies may start publishing their own guesses at what material is required or re-jigging their old books in new covers perhaps. I do predict a fair amount of confusion in the short term, until the Japan Foundation starts publishing some example tests or textbooks.

The requirement to pass each section in order to pass overall is new. No longer can your kanji ability carry your weak listening ability. Although the score requirements have yet to be published, I think the pass requirement might be lowered and two pass levels Weak and Good will be introduced from what I see in the booklet.

Overall it’s a more modern sounding test. The emphasis has changed to testing real world abilities rather than memorisation and exam technique. The main guideline is now the “Linguistic Competence Required”.
Ideally an oral exam and written exam would also be offered but within it’s limits (multi-choice and thousands of candidates) JLPT is trying to keep up with modern ideas about language teaching and learning.

I wonder if I’ll attempt N3 this June. I reckon I have “the ability to understand Japanese used in everyday situations to a certain degree.” What that certain degree is I’m not sure!
I wonder if SOAS will offer it.
In the meantime here’s a link to sample questions on the new JLPT test.