Looks like the Great Firewall or something like it is preventing you from completely loading www.skritter.com because it is hosted on Google App Engine, which is periodically blocked. Try instead our mirror:


This might also be caused by an internet filter, such as SafeEyes. If you have such a filter installed, try adding appspot.com to the list of allowed domains.


Lurks   September 21st, 2010 9:44a.m.

I shall now be tested on 电 next in... nine years apparently.

jww1066   September 21st, 2010 10:40a.m.

I was helping a 12-year-old niece study for a science quiz this morning. She had flash cards with definitions of various parts of the cell. The previous evening it had required a lot of struggle for her to remember things like endoplasmic reticulum, vesicle, and vacuole, but this morning she blew through them without any problem. Then she reviewed them again... and again. I told her it was a waste of time, all those cards were already in her short-term memory, but I think she wanted an ego boost and/or reassurance.

Which makes me think it might be fun to have an ego boost of our own: every once in a while we should cram a bunch of characters that used to be really hard just to see how easy they are now. ;)


Lurks   September 21st, 2010 7:23p.m.

Heh yeah. You know a really interesting thing, if you look at your vocab list and look at time spent, you see the words you've had the most trouble with.

Here's my top 5:

1. 麻婆豆腐
2. 售货员
3. 矿泉水
4. 警察
5. 熟悉
6. 幸福
7. 餐巾纸
8. 希望
9. 得
10. 食堂

They are mostly ancient history which probably goes to show you get better at learning. I guess that stands to reason, now I remember things by radicals having essentially learned them all.

I still have some ridiculously easy ones I keep getting wrong. Often by confusing words like 练习 and 联系 and characters like 熟 and 舒.

Some of it is context, things being harder to nail because you just don't use them, aren't really sure how they're used. I found this improved when I made lists more relevant by focusing on textbooks I was using and terms which I had imported from the zhongwen extension.

Lurks   September 21st, 2010 7:23p.m.

Alright it's a top 5 with EXTRA! :)

nick   September 21st, 2010 9:36p.m.

Two hours of my life for writing these cursed words:

遗嘱 (yízhǔ): 14 mins
积蓄 (jīxù): 14 mins
愚昧 (yúmèi): 12 mins
旅 (lǚ): 11 mins
渐 (jiàn): 11 mins
概 (gài): 11 mins
赋予 (fùyǔ): 11 mins
不 (bù): 11 mins
趋势 (qūshì): 11 mins
陵 (líng): 11 mins

I would love to build a feature that combined analysis of what you spend a ton of time on with its importance to identify those words that are just wasting your time. Then you could either choose to delete them or focus on them.

Lurks   September 22nd, 2010 5:49a.m.

Oh lots of buggers in those (不? lol). I guess your data isn't actually the sum total of the time you spent learning hanzi. :)

jww1066   September 22nd, 2010 7:50a.m.

My number one is 一 with 44 minutes, which seems bizarre. All I can figure is that it counts the time spent in words that contain 一, or maybe it's because I am often switching away from Skritter and whatever word I happen to be on when I switch away gets the full 30 seconds.


Foo Choo Choon   September 22nd, 2010 12:09p.m.

I've spent 20+ min. for each of the top 50; 15+ min. for each of the top 150.


Lurks   September 22nd, 2010 5:52p.m.

jww1066 - nah it can't count words that appear in other words, otherwise our top list would be dominated by those sorts of things. Would seem odd you'd switch away from 一 in particular?

Inaccuracy of timer when you go idle is a good point. Nick, do you think it would be possible to undo the time if it runs to the full 30 seconds? Assuming you've wandered off.

Hmm, come to think of it. It would be quite cool to plot length of time required to learn each hanzi over time. I'm doing some data processing stuff with python this semester so I should be able to get around to doing cool stuff with the the API at last...

nick   September 22nd, 2010 6:14p.m.

It does count characters that appear in other words, actually. That is why 一, 不, and 得 are in those lists.

Yes, I still have it on my list to make a much smarter AFK tracker that would do, among other things, what you describe. It is a very old to-do.

Let me know if you need other API calls or data dumps and I'll see what I can do.

Lurks   September 23rd, 2010 7:45p.m.

Ah curious, none of my single character words in my top list (of time spent) are parts of compounds. Well except for 得 maybe.

I've got an essay to write (on ancient Chinese philosophy no less) in the next few days but next week is free for cool stuff. I'll give it a bash and let you know. I want to see what sort of analysis Python's natural language toolkit can bring to bare on Chinese as well as making some visualisations. I want to try plot characters learned rate against time spent as well as that efficiency of learning over time.

This forum is now read only. Please go to Skritter Discourse Forum instead to start a new conversation!