My summer of Skritter

[li3wei1]

In return for three months of free Skritter, I've agreed to share my experience. Here's my first report.

 

First, I should explain that I am unusual in that I learned Chinese a long time ago, traditional characters, lived in Taiwan for four years and got a lot of conversation time in, then left it for many many years. About four years ago, I started relearning it, this time with simplified characters, and focusing on reading (mainly because that was what I had available). So my writing is really, really weak. I can write an essay without too much difficulty on a computer, but writing by hand on a blank sheet of paper is another story: maybe a few hundred characters, then I get into trouble. I've had problems with even basic characters: 最，说，现。I've had a go with paper flashcards a few times, but generally give up after a while.

 

I'm also a newcomer to the smartphone phenomenon. I finally gave in and shelled out for a phone and Pleco, so that I could take advantage of the bits and pieces of free time in my life: waiting in lines, on the bus or train (or waiting for them), getting to sleep, getting back to sleep, etc.

 

And finally, I don't like spending money. In a big way. So the offer of three free months was too good to pass up. Naturally I decided to put all my other study plans on hold and just binge out and learn as many characters as I could, so I wouldn't have to pay for more Skritter when my three months was up.

 

I started my account on 17 June, and unfortunately I was in the middle of a very busy period, when I had very little time to sit down and read manuals. I went straight for the Android beta version, and it took a few tries to get it working and configured to do what I wanted, which is just writing. I'm still not sure that I'm using it optimally or correctly, but I've found a groove, and I'm learning.

 

I downloaded the HSK5 list, figuring that while there were probably plenty of HSK1-4 characters that I needed to work on, there would also be many that I already knew, and to some extent it didn't really matter what order I learned them in. I'll go out on a limb and hypothesize that learning to write any character will help me remember how to write all of them, or at least learning to write any one character will improve my ability to remember 10 other characters that share components with, or that I confuse with, that character.

 

So Skritter is telling me that up to 30 June (I'm not sure why it's not giving me up to 4 July, when I'm writing this), I put in 4.7 hours of study, and 'learned' 306 characters and 226 words, with a retention rate of 93.4% (I have no idea how this is measured). 1757 reviews (total words + characters), at an average of 7.1 per minute, apparently. I'm sure quite a few of those were words or characters that I could already write. And many more have been ones that I only needed a little nudge on, but anything that can give me that nudge on a few hundred characters in a couple of weeks is way cool. Sitting down for a session, and reviewing the characters that I've gone from 'no clue, show me' to 'yes, I think I can remember this' in the last week or two is very satisfying, because on the whole I do remember them.

 

Most of the 4.7 hours has been while travelling. Being too cheap to shell out for a graphics tablet, I stick to the Android. For me, the best Skritter time is waiting for a bus or train. Once you're on the train, there's time to pull out a book and read, and on a bus, it's a bit jittery for Skritter. It's not all that easy to use lying down, either, I find it very tiring if my finger has to point upwards. I also find it difficult to do more than about 30 minutes before getting bored or mentally exhausted. Now that I'm not so busy, I might try to spend several sessions of 10-20 minutes per day, whether travelling or not, to plow through the thousands of characters I need to learn.

 

One problem I'm finding is that the program actually helps me. Let's say I'm trying to write 到, but all I can remember is 至. I start writing 至, and it puts it over on the left side, reminding me that there's another component on the right. It also tells me when I'm done, when I may think I've still got a few strokes to go. Or it will 'interpret' a stroke for me, even when it's in the wrong place, shape and direction. So 'false positives', where it tells me that I've got it right when I don't really, are as common as 'false negatives', where it tells me I got it wrong because the bus hit a bump or I had low blood sugar. But I can live with this.

 

On the whole, I'm loving it, but probably not spending as much time on it as I should. I've got a lot to cover in three months.

[lechuan]

Thanks for the review @li3wei1. Try turning on the "raw squigs" option. Then it will not give you hints or 'interpret' your strokes. Instead it keeps what you draw, and allows you to compare it with the 'real' character after. I found it much more useful as a learning tool once I did that.

[imron]

Yep, definitely agree with using the Raw Squigs setting.

[li3wei1]

Time for an update. I've had it for a little over a month now, and I'm using it every day, but for less than I'd hoped - it's telling me I spent 46 minutes per day over the last week. In all, I've studied 18 hours, done 6847 reviews (words + characters), and 'learned' 496 words and 572 characters. As I said above, many of these will be things that I used to know, or knew as traditional characters, or at the very least have seen many times. Nevertheless there are many that I'm still having trouble with.

 

The only times I've gone over an hour on a single day have been when I was travelling, or at a music festival. This is my problem with this sort of thing. Even knowing that I've got it for a limited time, and I should take maximum advantage, I'm using it for less than an hour a day. If I paid for a year's subscription, no matter how much I paid, I'd be saying "I'll do it tomorrow, I've got the rest of the year". It's like a fitness center membership - money down the drain. If they figured out how to charge on a usage basis, so much for each review, say, then I would use it as much as I needed, knowing that I was paying a fair price. I'd also be motivated to try harder to learn, to do fewer reviews per character.

 

I switched to raw squigs, and that does make it harder.

 

I'm thinking of making a new list, by going through some paper flashcards and pulling out the characters that I can't write, rather than just using HSK5, which has a few that I can.

 

I've cut down on my other activities, such as reading and Pleco flashcards, because I know I can come back to those when my free summer is over. I have noticed, though, when I do read, that I'm recognising characters from Skritter. So it's definitely doing good. And it is a real pleasure to be able to produce a word like 糟糕, which I couldn't have started a month ago, after not seeing it for a few days. Right now I'm working on 隔壁, among others, a word I could use comfortably but probably not recognise unless the context was really helpful.

[li3wei1]

While my three months isn't up for another 9 days, this is a good time for a wrap-up, because for the last week or so I've let it slide, to the point where I have a backlog of 494 items, going up by about 50 per day.

 

All in all, I've put in 48 hours, done 13,385 reps, and learned 1009 characters and 949 words. According to Skritter, that is. I was working through the HSK5 list, and I'd say I could already recognize 80% of those, and produce maybe 15-20%, and many of the others I'd had considerable exposure to. And, while it says I've learned that many, my retention a week from now, or even right now, is probably about 80% or so (that's production, not recognition). Still, my character production has definitely improved.

 

I would say that it was definitely worth the time that I spent. It's probably the most efficient way of learning to write characters (though the SRS programme could be tweaked a bit for greater efficiency), and learning how to write characters helps you recognize them and appreciate their meaning and interconnections. I appreciate that it is not necessary to write characters these days, for most people anyway, and choosing not to spend a lot of time on writing is a viable option, but a) I need to know how to write them and b) I enjoy knowing.

 

Was it worth the money I would have been spending? It worked out to a little less than one US$ per hour, for me anyway. Summer is a hectic time for me, and I may actually have been able to use it more during the school year. For most of the time I struggled to get in an hour a day. I actually did best when I was on holiday, spending hours on the beach. Looked at another way, it was about 4.5 cents per character 'learned'. Compare this with, say, buying Heisig and working your way through it. Probably more time, certainly less money.

 

Because my free subscription was time-limited, I tried to devote all available time to it, so I stopped reading and Pleco flashcarding (except when conditions were too bumpy to use Skritter). So this says something about how little study time I actually put in. I think Skritter would be much more effective as part of a balanced diet of reading, talking, listening, etc., so if you're going to get your money's worth out of it, you should be spending at least two hours a day studying in some way, probably 25-40% of that on Skritter. Of course, if you have lots of money, go for it. It's probably better value than paying more than $10/hour with a tutor, though you're comparing apples and oranges, because you're learning different things.

 

I could get another 6 months for $60, and I'm strongly tempted, because I may actually use it more during the school year – my daughter's out of the house, I spend more time on trains and buses, the vegetable garden isn't screaming for attention. And I haven't quite finished the HSK5 list.

[li3wei1]

I've noticed a big jump in my ability to get through Chinese text. I was doing a lot of reading in the spring, but then I stopped pretty much all Chinese activity apart from Skritter for the three months I had it. That ended in September; I noticed the improved reading sometime in Oct/Nov. Maybe there's a connection?