Learning Hanzi the Skritter way

The biggest problem with self teaching is that you tend to focus on the stuff you like, or the stuff you find easiest. For me that manifested as being able to read thousands of characters and being able to type a great deal of them with a computer but with a complete inability to write a single hanzi character by hand. You could certainly rationalise that there’s no great need to write by hand but with the love of reading Chinese came also the desire to be able to write it by hand. Months went by and I still neglected  this skill but increasingly I was writing characters into Pleco to look them up and I think that removed some of the fear.

I had heard good things about Skritter, a flash-based web site which teaches writing characters in an interactive way. With the knowledge that electronic systems can be hugely helpful in learning (my flash card progress as a case in point), I still wasn’t taken with the idea of drawing characters with a mouse. So right off the bat I ordered a Wacom Pen & Touch, signed up to Skritter and got stuck right in.

Eight days in, studying on average an hour a day, Skritter says my learned characters is about 200. That seems like good progress to me but to be fair a lot of the time I just need to bridge this gap between recognition and writing. For simpler characters the first time I write them it’s sufficient to remember. Since my word list is based on the first university text book, it’s basic stuff and I expect to continue to make rapid progress. So far it’s hugely enjoyable, less daunting than I thought it would be and Skritter itself is a joy to use. There’s word lists for loads of popular text books, HSK levels and stuff like that. I particularly like the progress stats too. They’ve also added Japanese recently just in case you fancy studying the wrong language

I’ve always had an interest in radicals as a tool for learning hanzi. Armed with a sufficient knowledge of radicals you can often guess what a word is which saves constant look-ups when trying to read Chinese. Learning to write has turbo charged my radical skills no end and of course this helps with writing too, once you’ve mastered the strokes for a particular radical they show up again and again elsewhere. Another thing I found is that after hearing much said about the importance of stoke order, I have to say I’m a little baffled by it. Not because I don’t think it’s important but because it seems to conform to simple rules 95% of the time – I’m hardly having any  trouble with stroke order at all. Maybe that’s a side effect of my learning strategy to break everything down into components and systems.

So what of Skritter? Overall it’s pretty awesome. Having your strokes snap into place is gentle guiding hand and I find that my strokes get more correct just by trying to avoid Skritter correcting me so much. The buzz of doom when getting something wrong or pressing the show button (admitting defeat), is a subtle but important feedback. I hate that sound and I try to avoid it.

If I have a criticism it is that the stroke recognition is pretty basic. It’ll often go ahead and fill something in which you had completely wrong, while other times it seems remarkably hard like the second hook stroke in the food radical. Skritter is not a hard task master, it’s easy to game the system and just wing it through characters because of how easily it will match the strokes but nevertheless I find that returning to characters I was guessing my way through shows that I’m getting better at them anyway. I think a bit of discipline helps here, I often hit erase and start again if I fudged my way through.

There’s improvements I’d make certainly. I’d really like a radical breakdown for example, probably a better character look up system perhaps with some licensed content from a dictionary. I think a better system for just practising strokes would be good for rank amateurs like me. I’m still utterly hopeless at lots of types of strokes and short of just hitting erase and trying again, Skritter isn’t too helpful. I really am genuinely hopeless with a writing instrument, my hand-written English is diabolical. If I’m honest there’s probably some mileage here in the old fashioned sheets of paper and writing out countless copies of the same character, just to get a better accuracy and neatness to my hand-eye coordination. It’s probably unfair to lump my disability in this area at the doorstep of Skritter as if it should help me here too.

I’ve pointed out to the Skritter team that I think it should definitely mark you as wrong for getting a drop the wrong way around and other obvious recognition errors. I still don’t know how much of an issue this really is, like I said I find I get better at returning to these characters anyway. The concern is that with a very large vocab list you don’t tend to return to the same ‘learned’ characters again, so a weakness in your knowledge might not get thrashed out by being so lax. I think I need more than eight days to tell on that score.

I’d also be ecstatic if it would actually support pressure input, it seems a shame you’re doomed to be incapable of imitating the thin/thick strokes of the cursive font you are imitating. It’s like it’s taunting you, using a print face would leave me feel less cheated in that regard but I think you’d lose a lot of the demonstration of stroke order/direction that is implied from a proper hand-drawn font. There’s also a substantial difference between the font you write to and the print font that shows up when looking at characters in a compound word – which is a minor issue but one I’d rather wasn’t there.

That said there’s a lot to look forward to with plenty of development going on. Skritter is a great little company with a personal touch. The boss sent me a hand written card when I signed up and it’s super easy to get in touch with concerns, the forum is a nice little community. If you sign up for a year it’s something like $7 a month and in the scheme of costs (hundreds I’ve shelled for Pleco and dictionaries for example), I think you’d be hard pushed to find a better value tool in this area. With Flash 10.1 support coming to Android soon, it would seem reasonable to expect to be able to run this on a mobile which really would be the holy grail I think.

Oh, in case anyone is looking to sign up – if you follow this link to the referrer program if after the trial you sign up, you will get two free weeks and I get two free weeks too. Can’t say fairer than that.