Blog Archive

Friday, July 25, 2014

A brief history of the Japanese writing system

Unless you're studying Japanese (I assume most of you are!), you may not realize that Japan uses three unique writing systems, hiragana, katakana, and kanji. If you count English which is a mandatory class in school, this brings the total count to four writing systems used.

An example of a sentence using all four writing systems could be:
"I went to Pizza Hut but there was unbelievably no Tabasco or red pepper."  
(Most would likely use ピッザハット for Pizza Hut, but this is just an example.)

Here I'll explain briefly how this came to be and how Japanese adopted it's writing systems.

A well respected writer and authority on Japanese once said: "Only one predominant language of one major nation remains today without clarification of its origins — Japanese".  While Japan uses kanji which originated from China, the full origin of the spoken language itself is unknown. Infact, Japan didn't have a writing system until it was introduced by Chinese around 50 AD, however most Japanese people remained illiterate until about the 5th century AD, (Japan's literacy rate today is nearly 100%)! With the import of Chinese characters, kanji started to be used to write Japanese words, and around 650 CE a writing system called man'yōgana was invented that used Chinese characters for their sounds opposed to their meaning to represent Japanese. It was through man'yōgana that kana developed, which at the time was most often used by women who were denied higher education.

(This graphic shows how kanji evolved into hiragana by means of man'yōgana. Kanji is in black on top, cursive style man'yōgana in red, and hiragana in black at the bottom.)

Katakana also evolved from parts of the man'yōgana by buddhist monks who wanted to simplify it even further:

Today hiragana is used to write Japanese words that have no kanji and words that kanji cannot be remembered for or haven't yet been learned, while katakana is used for foreign words, loan words, often used for onomatopoeia, used to show the difference between a 訓読み (Japanese reading) and 音読み (Chinese reading), to show emphasis, some company names, etc. Kanji is mostly used to write nouns, adjective stems and verb stems, and hiragana is used for particles, verb endings, and the phonetic readings for words written in kanji.

There are some kanji that were created in Japan and thus won't be known in China, called "国字", translating to "national characters", and other more obscure kanji used in names alone (人名用漢字), that might not be known unless it is in your name or someone's name that you know. To graduate from high school,  students are required to learn the "常用漢字", or every day use kanji which consists of 2,136 characters. 1,006 of those are taught in primary school (known as 教育漢字), and the other 1,130 are taught in secondary school. There's no definitive number for how many kanji there are, however the "大漢和辞典", a comprehensive dictionary written in 1917 contains about 50,000 characters. A similar Chinese dictionary contains 85,000 characters, however many are not common use in any country and are obscure variants or archaic forms. Today there are about 2,000 to 3,000 common use kanji in Japan, with more scattered here and there which are used occasionally.


When Japan first adopted hiragana, there weren't the same diacritical marks which are the symbols used to "voice" a kana into another sound. There's the 濁点 meaning "muddy mark" (the circle voicing used to change は (ha) into ぱ (pa) as an example), and the 半濁点, or "half dakuten", which is the voicing mark which looks much like quotes and can be used to change a kana like か (ka) into が (ga).  Colloquially the handakuten is known as a "点々", or "dot dot", and the 濁点 is known as 丸 or "circle". The introduction of the handakuten actually came from Portuguese missionaries in the 16th century who were baffled at how Japanese people just "knew" when a word was to be voiced and when a word would not. For instance, before the handakuten and dakuten were implemented, there was the "不濁点" -- a modern spelling of a word like "いがた", which used a handakuten to voice the か into が, would be spelled "いかた゜", using the 不濁点 at the end showing the reader the word has a voicing, except it would be up to the reader to know which part of the word to voice, in this case the ka to ga. Back then there was no difference between ハ行 and パ行, (the ha he fu hi ho and pa pe pu pi po sounds), and so it was the the introduction of the handakuten and dakuten that differentiated them.

Here is a clip from 日本人の知らない日本語 (Japanese the Japanese don't know) talking about this:

As a note, only the ハ行 kana can use both dakuten or handakuten, making a ha into either a ba or pa, though when used on ka, ki, ku, ke, or ko, it represents the sound of "ng" singing, which is a sound used in many Japanese dialects. In katakana only, the dakuten can be added to ウ creating a "v" sound which is used in modern loanwords, like "ヴィスタ" (vista).

Hiragana and katakana used to have two extra kana each, which are now near obsolete, ゐ (hiragana "wi"), ヰ (katakana "wi"), ゑ (hiragana "we"), and ヱ (katakana "we"). It's thought that ゐ (wi) and い (i) indicated different pronunciations until around the Kamakura/Taisho period where they both came to be pronounced "i", likewise for ゑ (we) and え (e).
These kana are still rarely used today, for instance Nikka Whiskey uses ヰ in it's name, "ニッカウヰスキー". They are also used in Okinawan and Ainu languages (two other languages related to and spoken in Japan).


A katakana "を" (wo) also exists though rarely used, which looks like: ヲ.  Often if an alien、 robot, foreigner, or something along those lines is talking in a book/manga, their speech is converted to all katakana to convey the alienness/accent, so this could be a case where ヲ would be modernly used.

Here's a screenshot from 塊魂 (Katamari Damashii) where katakana is used to convey this:  
 「カンタンデショ?コレナラ 小サイ王子ニモ デキルデショ?」
 "Isn't this easy? Even a little prince can do it with this."

Wednesday, July 16, 2014

Improve your character writing by enabling raw squigs

One of the main goals with Skritter is to help you learn to write characters. We believe that using Skritter's unique combination of real handwriting, corrective feedback and spaced repetition is far superior to using just paper and pencil. Not to mention that it's more fun!

As is the case with any review system, there is a distance between what we practise and what we want to use it for. Practising writing characters with Skritter on the web, iOS or the recently launched Android app is similar to writing characters on paper, but it's not exactly the same. In this article, we're going to start looking at some of the differences and how they affect your studying. Examples used here will be in Chinese, but the general principles are of course the same for Japanese; this article is for all Skritter users!

Using Skritter vs. writing on paper

Before we get started, however, it should be noted that most of the differences between using Skritter and writing on paper are quite small and a majority of them are to your advantage, such as alerting you to incorrect stroke order or telling you that a stroke is written in the wrong direction. Paper doesn't care about any of these things and even teachers need to pay attention to notice. Skritter does this automatically!

When deciding how Skritter interprets the strokes you write, we have to make a decision how strict the interpretation should be. It's obvious that we can't require you to write exactly like the model character, but neither can we allow strokes that differ too much. The current state is a compromise between the two and we constantly update character models and interpretations to enhance the learning experience.

Don't let Skritter help you too much! 

Still, if your goal is to write clear, correct characters, there are situations where Skritter will help you too much. This is fine if you pay attention, which is one of the main reasons for writing this article: we want more users to pay attention. One of the problems can be easily overcome by activating a function in Skritter called raw squigs, but let's look at the problem itself before we solve it.

Skritter attempts to determine whether the strokes you write matches the model or not, but a computer program can't know what you intended to write. Since there needs to be some margin for error built into the program, strokes will be interpreted as correct even if they are somewhat out of place or slightly wrong in some other way. In most cases, this is okay.

The problem is that sometimes Skritter interprets a stroke in an entirely different way than you intended, thus helping you a lot. Let's look at two examples:

Example 1: Stroke position: 嚣

In this example, Skritter moves an incorrectly placed stroke to another position than the one we intended, while still treating it as a correct stroke. This gives us a very important hint how to write the character, a hint that wouldn't be available to us if we wrote on paper. We thought there was a 口 to the left, but in fact there isn't and Skritter shows us that by moving the stroke as shown above. If we use this hint and get the character right, we are effectively cheating ourselves. We wouldn't have been able to write this character on paper.

 Example 2: Stroke orientation: 写

In the second case, we intended to write a right-sloping dot as in the first sttroke in 穴, but Skritter interprets this as as the first stroke in 冖, which is sloping in the other direction (and is slightly off-centre as well). Thus, we indented to write one radical, but it turns out it's not the right one. Skritter tells us that we were wrong and shows us the correct stroke by moving and rotating the stroke. If we would have written this on paper, we might have ended up up with an incorrect character (one with a dot above 冖), so we should treat this character as forgotten (incidentally, the traditional character for 写 does have a dot: 寫).

Improve your character writing by enabling raw squigs

Apart from being vigilant and paying attention to these and similar cases, you can also turn on a function called "raw squigs". This is a function that show you what you actually write rather than snapping your strokes to the model character. This means that you won't receive any hints about position or stroke direction, thus taking your reviewing one step closer to the way characters are written without feedback on paper or elsewhere. You can see what Skritter looks like with raw squigs turned on in the picture to the left. Note that our strokes (light blue) are still visible. The model character is only shown once we finish the entire character. If you want to know more about this and see some animated examples, I have produced a video on YouTube about raw squigs.

If we look at our example with 写 above, we would run into problems very quickly and it would be easy to spot them. Just look at the picture to the right. If we try to continue writing the character as indicated (left-sloping dot where the cursor is), Skritter won't accept it and will eventually show us the correct stroke (light blue). Since that stroke is very different from the one we intended, it's obvious that we are wrong here and we should treat the character as forgotten.

How to enable raw squigs

Enabling raw squigs is easy. In the web application, simply open the settings page by clicking the cogwheel in the top right corner and then check the raw squigs box in the bottom left corner (see screenshot below).

For mobile devices (iOS and Android), go to "Settings" and then enable raw squigs.

Enabling raw squigs will make your characters look uglier and you will decrease your retention rate a bit, but since you are judging yourself more harshly, you will end up with better handwriting skills. Naturally, if you don't think correct handwriting is very important, you don't really need raw squigs.

You can learn a lot about characters, both writing and recognition by using the default settings, but if you want to be able to write characters without help, you should turn raw squigs on. In my opinion, decreasing retention rate and feel slightly worse about my character writing is a fair price to pay to improve my character writing.

Wednesday, July 2, 2014

Sensible character learning challenge 2014: The Big Finish

challenge14-4After 101 days, the sensible character challenge 2014 has now come to an end. It's time to hand out the last prizes and wrap up the challenge. If you've participated, I hope you have learnt as much as I have, and if you haven't, I'll make sure to highlight the things you really should read.

I recommend reading the first article (Sensible Chinese character learning revisited) as well as the "what have I learnt" sections of the other articles. These articles were posted on Hacking Chinese, but I will write much more about character and vocabulary related issues here on the Skritter blog later.
  1. Sensible Chinese character learning revisited
  2. Sensible Chinese character learning challenge 2014
  3. Sensible character learning challenge 2014: Milestone #1
  4. Sensible character learning challenge 2014: Milestone #2
  5. Sensible character learning challenge 2014: Milestone #3
  6. Sensible character learning challenge 2014: The big finish
Prizes for the big finish
Since this is the end of the challenge, there will be even more prizes!
  • Skritter extension - One week free extension will be awarded to all active participants. If you want your free extension, you need to have been active in the challenge, all you need to do is join this group and you should get your extension (provided that you have been active, of course, meaning a bare minimum of joining the challenge, posting a progress update for this milestone, along with regular use of Skritter in June).
  • Hanzi WallChart posters - Three sets of posters worth roughly $50 each will be distributed randomly among active participants. These posters aren't only informative, they look cool too! You can see the posters here.
  • Glossika Chinese products - Glossika offers a range of products for Chinese learners and three participant in this challenge will receive one product of his or her choice for free. You can find more information about both Glossika and their products on the official website.
Winners are determined the same way as for previous milestones, i.e. randomly, but weighted for activity in the challenge (basically anything I have a chance to notice, including posts here, on Hacking Chinese, social media and so on), with a particular focus on progress updates.

I will announce the winners here on Sunday (July 6th), so you have a few days to post your updates. Note that only people who have officially joined the challenge are eligible. 

Your progress update

There's no fixed template, just write whatever you want to write in any way you see fit, but here are some examples:
  • Have you reached your goal for the second milestone?
  • What (if anything) are you going to change?
  • What have you learnt by participating in the challenge?
Note that activity in the challenge is completely unrelated to whether or not you have succeeded! Failing to reach your goal, thinking about why you failed and what you should do about it is perfectly acceptable. 

My progress update

I have reached my goal, I now have more than 5800 individual characters in Skritter! Naturally, I spent some significant time learning the last few hundred this month and some of them haven't really sunk in, but they have all been studied and learnt. his is what my challenge history looks like:
And this is what my Skritter progress page looks like:

How many characters do you need to know?

My goal for this challenge begs the question of how many characters one actually needs to know. The simple answer is that it depends on what you mean by "need". If you mean to be able to read most modern Chinese texts without having to look up many characters, you need far less than the 5800 I'm close to here. In fact, you can get very far with around 3000 characters and 4000 will make you comfortably literate (I'm now ignoring the fact that literacy of course includes other things than knowing characters, such as knowing words, grammar and so on, but that's not the point here).

So why did I think it was interesting to learn an additional 2000 characters if it isn't very useful? I did it for two reasons. First, I wanted to feel what it was like learning characters again. I haven't spent significant time learning characters for many years and this challenge was interesting because it made me realise some things I hadn't noticed before. I will write about these things later (some of them are already mentioned in the milestone reports).

Second, it's a mental challenge and quite fun. Even though I haven counted the exact time I spent on learning 1800 characters, I'm pretty sure the average is no higher than half an hour per day. That means about 50 hours or about two characters per minute. This might sound extremely efficient, but then keep in mind that most of the time, learning a new character is a matter of associating two characters that I already know with a new meaning. If it's a perfect phonetic-semantic combination, it becomes even easier (learning a character like 浬, nautical mile, takes just a few seconds to learn). Also, spaced repetition is very efficient.

Learning characters is not like learning random facts

When I started learning Chinese, I remember being a bit confused by people who said it was difficult to learn lots of characters. I mean, learning a few thousand isolated facts isn't that hard. What I didn't understand back then was that learning 5000 characters isn't like learning 50 characters a hundred times. The main problem when learning new characters isn't to learn how they are written and what they mean, but to keep them separate from the other characters you already know. Thus, even though character learning certainly becomes easier in some sense, it also becomes a lot harder, but for different reasons.

Future challenges
We are huge fans of challenges and you will definitely see more of them in the future. Some will be like this one, where we try to learn things together and share useful tips and insights, others will be very different, such as the character guessing challenges we have been running on Twitter for some time now (see a few examples in this forum post). If you have any cool ideas for challenges, let us know!

Stay tuned...

We will announce the winners on Sunday by updating this article, so make sure you post your progress report before then. Stay tuned!

...and the winners are

It’s now Sunday and it’s time to declare the winners:
  • Carla (both prizes for her wonderful graphics)
  • Doug Stetar (Glossika product of your choice)
  • Georges (Hanzi WallChart poster set)
  • Luke (Glossika product of your choice)
  • All active participants: Free Skritter extensions
I have sent e-mails to the winners. If you are an active participant and want your Skritter extension, please join this group and tell me. Any prizes left over from this challenge will be handed out in future challenges, stay tuned!

Tuesday, June 17, 2014

Skritter for Android Beta Release

The wait is over, and Skritter for Android is finally here! The app has come a long way since the alpha version released late last year, and now after several several weeks of internal testing, we have decided that the time to release this as a public beta has arrived.

This is a beta, so we need your help!

What we release today is a beta version, which means that even though there are still many kinks to work out and features to implement, we would rather make the app available today so that users with Android devices can finally get their hands on the app and help us make Skritter the perfect learning tool. Since we can't test the app on all available devices, by using it you provide us with valuable information so that we can continue to update and improve it. If things are not currently up to your ideal expectations, don't fret, as these updates will come often!

Devices that can run the app

The app is targeted for all devices with Android 4+, and we've tested Skritter using the following:

  • Samsung Galaxy S3 & S4
  • Samsung Galaxy Note 2, 3, and 10.1
  • Nexus 5, 7, 10
With tons of devices (5461) using Google Play, it might take some time to increase support for everything, especially older device models. If you run into problems, please mention the device you're using in your feedback. We are pushing the boundaries of what HTML5 can do with educational apps, and this relies on a fair amount of third party support, so sometimes we have to wait for other parties to implement fixes to their own technologies before we can keep advancing toward uncharted waters! 

Enough-- now tell me how to get the app!

Follow these instructions to install the app on your Android device. It's important that you follow these steps carefully and make sure you complete each step before going to the next!
  1. Join the public Skritter Beta Testers Google community with the account you have linked with the Google Play Store.

  2. Become a beta tester by visiting the link corresponding to the language you are studying:

    Chinese (click “Become a Tester”)
    Japanese (click “Become a Tester”)
  3. Either click "Download Skritter from the Play Store" from the testing agreement, or use the links below:

    Download Skritter Chinese
    Download Skritter Japanese

    Note: You can’t find the app by searching the Play Store yet, so must use the methods mentioned above to download it successfully.
  4. Start the app, sign in using your Skritter username, and enjoy! Finally, Android users are no longer tied to their computers-- go forth and Skritter away anytime and anywhere! Please note that the first time you login your entire account will be downloaded, which can take some time, but you only have to do this once!

  5. If you're new to Skritter and want to try out the Android version, you can create a new account from within the app.
  6. Report feedback in the Skritter Beta Testers Google community you joined in step one. The more information you provide, the easier it will be for us to fix the problem!
  7. Keep an eye out for updates. We will update the app often based on the feedback we receive and each update will be announced in the Google community. If there is an update available, it will show up in Play Store > My Apps > Installed.

What's new?

The Android version isn't meant to be identical to either the online flash version or the iOS app, but it's of course meant to help you learn Chinese and Japanese equally as well. Here are some new features in the Android app:
  • Guided learning mode for new character writings.
  • Support for Bopomofo (注音符號).
  • Focused learning with hidden study timer and due counter.
  • Landscape mode for tablets and larger devices.

Android? HTML5? Cross-platform? Open source?

The Android app is directly related to the current open-source HTML5 project. It is built on top the powerful Skritter API for data interactions and uses CocoonJS to help accelerate canvas drawing. By pushing the envelope with HTML5, we strongly believe that it will allow us to better expand and add features in the future. If you're a developer and ever wanted to see how Skritter works (or perhaps even contribute), we warmly welcome you to check out the GitHub project!

Saturday, May 31, 2014

Sensible character learning challenge 2014: Milestone #3

One of the most powerful ways of staying motivated is doing things with others, preferably during a limited amount of time with a clear goal. That is exactly what the sensible character challenge 2014 is about. Even though the challenge has now reach its last phase, it's still not too late to join, just set a character-learning goal that you feel is achievable before the end of June and you'll not only boost your own learning, you'll also have the chance of winning some great prizes, including character posters, language learning products and free time on Skritter!

If you want to know more, please check the post that launched the character challenge. If you want to sign up, all you need to do is set your goal for this month and include that in a comment. The rest of this article will be for people who are already in the challenge. I will write a little bit about my own experience and also encourage you to write about yours. Once everybody's had a chance to post their progress reports, prize winners will be announced!

Prizes for milestone #3

Here are the prizes available for the third milestone:
  • Skritter extension - One week free extension will be awarded to all active participants. If you want your free extension, you need to have been active in the challenge, all you need to do is join this group and you should get your extension (provided that you have been active, of course, meaning a bare minimum of joining the challenge, posting a progress update for this milestone, along with regular use of Skritter in May).
  • Hanzi WallChart posters - Two sets worth roughly $50 will be distributed randomly among active participants. These posters aren't only informative, they look cool too! You can see the posters here.
  • Glossika Chinese products - Glossika offers a range of products for Chinese learners and one participant in this challenge will receive one product of his or her choice for free. You can find more information about both Glossika and their products on the official website.
Winners are determined the same way as for previous milestones, i.e. randomly, but weighted for activity in the challenge (basically anything I have a chance to notice, including posts here, on Hacking Chinese, social media and so on), with a particular focus on progress updates.

I will announce the winners here on Friday (June 6th), so you have a few days to post your updates. Note that only people who have officially joined the challenge are eligible. Also note that people who join the challenge now will have to wait until the end of the challenge (June 30th) before becoming eligible.

Your progress update

There's no fixed template, just write whatever you want to write in any way you see fit, but here are some examples:
  • Have you reached your goal for the second milestone?
  • What (if anything) are you going to change?
  • What have you learnt by participating in the challenge?
Note that activity in the challenge is completely unrelated to whether or not you have succeeded! Failing to reach your goal, thinking about why you failed and what you should do about it is perfectly acceptable.

My progress update

Again, I seem to have overshot my goal, but this time it wasn't because of a bad goal, but because I spent a lot more time using Skritter than I thought. This is partly because I've been using the alpha test version of the Android app (which is working well enough to use that instead of the online version for my own learning). It's also because I went to 雲林 in southern Taiwan for a gymnastics competition and spent lots of time on buses and trains. Can you think of a better way to while away the time than learn lots of characters? I certainly can't! As a result, I cleared my goal for May with relative ease:

Milestone #3 (goal): 5340
Current status (May 31st):5409
End of challenge (June 30th): +366 (5775 total)

I will also share some important insight into learning characters.

Lesson #1:Spread it out

One of the major benefits of using your phone to review characters and words is that you can learn Chinese or Japanese wherever you are, whenever you have a few minutes to spare. It only takes a few seconds to start and you can easily interrupt your learning with no ill effects if something more interesting happens around you. This is much harder to do with any of the major skills listening, speaking, reading and writing. For instance, if you just have two minutes to study, it doesn't make sense to start reading a new chapter in a book or listen to a new podcast, but you can certainly clear a dozen reviews in that time!

Therefore, whenever you can, spread your reviews out through out the day. Don't review tones if you can speak with a friend instead. Don't write characters if you can read a book instead. Don't practise definitions of words if you can listen to a podcast instead. If you want to learn a lot of characters, such as if you are in this challenge, this is even more important! This is about time quality, something I've written more about here in case anyone wants to know more. If you pay attention to your daily schedule, you will find that there are lots of slots to review characters that you probably weren't aware of!

Lesson #2:Add context

Jake has written an awesome article here on the Skritter blog about something he calls "list overdose". He describes it as follows:
List overdose (or simply LOD) describes the ingesting or constant studying of vocabulary lists in quantities greater than are recommended or generally practiced. LOD may result in very little actual linguistic improvement (emphasis added).
I personally have a somewhat ambiguous relationship to this, because I think that you can use word lists quite effectively, provided that you are combining it with real-world usage and large volumes of input. So, when I say that I'm adding so and so many characters from a list, that's not the only thing I'm doing! I'm also reading tons of Chinese and listening to even more.

If you still want to add characters or words directly from a list instead of gathering them in the wild, I think it's very important to put them in context. This is relatively easy:
  • If it's a character component, add a few of the most common characters
  • If it's a character, add a few common words it appears in
  • If it's a word, add an example sentence that fits well with the word
This will make sure that you don't end up with a brick yard instead of a house. Sure, knowing just one way of using a word doesn't mean you know that word perfectly, but it is a lot better than not having any clue at all of how it's used!

Stay tuned...

I will announce the winners on Friday by updating this article, so make sure you post your progress report before then. Stay tuned!

...and the winners are...
  • Hanzi WallChart posters: Lili Woodlight and Jeremy (I have forwarded your info to the company)
  • Skritter free extensions: Everyone active is eligible, join this group and tell me
  • Glossika learning Chinese product: 愛美 (I have forwarded your info to Glossika)
Good luck for the final stretch of the challenge!

Wednesday, May 28, 2014

Vending Machines in Japan (自動販売機)

Vending machines (自動販売機; じどうはんばいき) are a convenient way to grab a small snack or a quick drink to keep you going throughout your day, but in Japan vending machines are on a entirely different level-- you can get a variety of things that you probably would had never imagined to be in a vending machine. You can of course get snacks you wouldn't normally find here like tuna mayo Doritos (and a plethora of other fun flavors), or maybe some canned bread, though strangely I might prefer my bread un-canned.


There are also vending machines that help out in a pinch:
Leave your house without your umbrella and it's going to rain? Just grab one from a vending machine.

Notice you have something on your tie right before work? No problem if you have one of these on the way.

I personally don't like going grocery shopping, so for those like me, this provides a convenient way to quickly pick up some items instead of having to go to a store and wait in a potentially long line:

And for those times when you don't feel like cooking or spending much money on a meal, 自動販売機! If you were near a vending machine that had 肉まん, 焼きそば and たこ焼き, could you resist them for less than 4 bucks each? 

I've only really come across hot food vending machines in America, primarily pizza or french fries. Have you come across any out of the ordinary vending machines where you live?

Wednesday, May 14, 2014

How Skritter Helped Me Stop Worrying and Love Writing Characters

I have studied Chinese for almost seven years now, but I have only been an active Skritter user for the last year or so. Today I want to share with you the story of how Skritter helped me stop worrying about writing characters and start loving them instead, how I went from being quite sceptical about Skritter to using it on a daily basis. Even though I study Chinese, I believe Japanese learners might also find my story interesting. Let's start from the beginning.

Chinese was the first foreign language I started learning purely out of interest. I have learnt English and French as well, but that was mostly because they were more or less fixed parts of our curriculum, and I only did what I was required to, without much extra thought going into the process of how to learn. Learning Chinese was different from the very start; I wanted to learn as much Chinese as possible each semester, so I wasn't satisfied with just adopting any method I stumbled upon. 

Paper, pencil and spaced repetition

My approach to vocabulary learning during my first year or so was very smart and incredibly stupid at the same time. I basically came up with a spaced repetition system on my own (similar but more primitive than the one Skritter uses to determine when you should review a word), without ever having read about it. This is smart because it turned out to be quite effective for learning, but it was incredibly stupid because I could have found much better systems simply by searching online.

Of course, it didn't take long before I realised that my manual system was limited. It became hard to keep track of thousands of words and I needed something better. This is how I learnt about spaced repetition software. I'm not going to go into details here, but for many years I learnt writing characters mostly by using free programs and ordinary handwriting on paper. When I felt lazy or had nothing to write on, I would trace characters on any surface using my index finger or simply visualise how they were written.

This works very well and is far superior to any kind of massed repetition approach where you simply write the same characters over and over again. I didn't think that my system was perfect, but I thought that it was quite good. I had of course heard about Skritter and knew what it was about, but I simply didn't think that it would be much different from the system I already used. It just sounded like a fancy version of a traditional approach.

Enter: Skritter

Then I arrived here in Taipei and got to know Jacob Gill, who introduced me to Skritter more thoroughly. He also sold me his old writing tablet so that I could write properly (using the mouse is pretty far from writing by hand and is quite annoying). I must admit that I was a bit sceptical, but since I'm always open for trying new things when it comes to learning methods, I decided to give it a try. That semester, I took a course in teaching methodology and language learning theories, a course mainly designed for native speakers with an in-class handwritten midterm exam.

Even though my Chinese was good enough to take active part in the course itself, expressing everything I had learnt up to that point in handwritten Chinese during a three hour test was still very scary to say the least. I had hardly written any Chinese by hand since writing the personal letter I submitted along with my application for graduate school, and that was first written on a computer and then copied to paper character by character!

An experiment in learning to write Chinese characters by hand

This was the perfect opportunity to see how well Skritter worked. I decided to divide my preparation for the exam into two parts: content and handwriting. The first part was about knowing the things I needed to pass the exam, including various approaches to language learning and how they related to each other. I decided to do this part mainly on my computer, mostly in Chinese, but also in English. The goal was just to acquire the knowledge I needed. The second part was about being able to use handwritten Chinese to show my teacher what I knew. I decided to use Skritter only for the second part.

I spent around two months doing this and it felt pretty good. I made sure to keep up with homework assignments and kept records (on my computer) of everything. I also practised handwriting using Skritter for the entire time. I didn't spend much time writing characters by hand apart from jotting down some keywords during lectures.

How did it go?

To be honest, I don't recall the exact score, but it was around 85/100, which I was very happy with indeed. Now, I don't think you're very interested in knowing about the score, but I think you might find the result of the handwriting experiment more interesting. On the whole, it went very well. I learnt many things that semester:
  • Reviewing characters can actually be fun – The most important advantage with Skritter over my previous method probably isn't that Skritter is more efficient , but that it's more fun and not a little addictive. Perhaps this is because I'm old enough to think that writing things on a screen is cool, but it really is more fun to receive immediate feedback from a program than to write the same characters on an uncaring sheet of paper.
  • Learning stroke order has never been easier – Even though the most important thing is to write clearly, following a standard stroke order is important. Writing on paper, I received no feedback at all, so by using Skritter, I sorted out hundreds of small stroke order inconsistencies. Some of my previous errors were acceptable, but others were not. In cases where Skritter and I didn't agree, I learnt a lot by looking things up..   
  • Writing on a screen is closely related to writing on paper – As I have described above, I didn't write much on paper at all during the experiment (and I haven't done that since then either, for that matter), but I still managed to write almost everything I wanted on the exam. I wrote neither quickly nor beautifully, but I wrote legible characters that communicated what I wanted to say. In other words, my practice with Skritter transferred very well into the real world. Of course, my hand hurt quite a bit after the exam, but that's only to be expected.
  • It's important to learn to write both characters and words – This ought to be obvious for most people, but I focused mainly on writing single characters, combined with a fair amount of writing on a computer and reading related books and articles. Still, what I found hardest on the exam wasn't how to write a certain character, but rather figure out what character was part of a certain word (I knew how to pronounce it and would have recognised it easily). I sometimes needed to try several options or think for a bit before finding the right answer. Words are the main carriers of meaning, not characters.

More than a year has passed since that exam and I still stick to mostly the same strategy, which means that Skritter is now part of my daily routine and has been so for some time. I think I have already mentioned the most important reasons why I use Skritter, but there is one more thing I'd like to bring up. I have a confession to make.

I don't like writing characters by hand

I remember what it was like when I wrote my first characters seven years ago. It was magical. It was exotic. I loved it. Now, I regard characters mostly as a tool for communication and I think that typing does the job so much better than handwriting that I really don't want to spend much time writing by hand. However, since I'm a teacher, I think I ought to aim for a knowledge level similar to that of native speakers (not in terms of writing proficiency or penmanship, but in terms of being able to recall how to write characters and words). I think Skritter is the perfect tool for me.

My current system

I'm very happy with the system I use now and it looks essential the same as the one I designed for the experiment discussed above. I mostly use Skritter for single character writings (I'm around the 5000 mark now and have set 5775 as the goal for the 2014 character learning challenge). This isn't because I don't care about words, but because I want to spend as little time as possible while still retaining the knowledge of how to write all those characters. I immerse myself enough in written Chinese outside Skritter (I read 25 books in Chinese last year, for instance) to take care of the bigger picture (words, sentences), so I think studying single characters suits me best. I get exactly what I want by spending as little time as possible (I've spent on average less than two hours a week on Skritter this year).

因人而異 – We're all different

This is my personal story and the method I described above works for me, but it's probably not the best way for everybody. I think most people would benefit from using Skritter more than I do, but very few would benefit from using it less. The only situation where you don't really need Skritter is if you don't care about characters at all and decide to go completely oral. I won't discuss such as an approach here, though.

If you read less than I do or spend less time typing, you can expand the way you use Skritter to cover what you need. The most important step is to add words rather than just characters. The only exception is when you see characters popping up in several different words, then it might be a good idea to learn those characters even if they aren't common words in themselves. Still, simply adding words from a list you've found might not be the best way to learn.

Beyond words, it all depends on what your study environment is like. Add anything that you don't think you will learn automatically by studying and using the language. If you want to improve your reading, focus on learning definitions. Tones are also important (much more important than you think, in fact). If you don't know what you need, you can just go all in and focus on definitions, readings, tones and writing, but if you do that, don't forget that Skritter is a tool. Chinese and Japanese are real languages used by real people, so go out there and communicate, either online or in the real world.

Stop worrying and start loving character writing

Now you know how I ended up being a dedicated Skritter user, but I'm interested in your story as well. How did you end up reading this article? I don't direct this question only to dedicated Skritter users, but to everyone. What is your method like and why does it look like that? Please leave a comment!