Flashcards Simplified

[ipsi]

SuperMemo is generally regarded as one of the better ones. I know it uses repitition spacing, but I'm not sure if it will do other things as well.

You're almost correct with regards to Pleco, but you can add flags to them, and that, if used properly, may allow you to do what you want. Maybe adding flags to represent chapters? But if you have 8000 cards, that's going to cause issues, as they have to be added manually in Pleco, and can't be set when importing.

Hopefully someone else can shed some light on this, otherwise I'll come back in a day or two and see if I can give you a hand.

 

[gato]

Again, please call me out on this or teach me how to use the Pleco dic better, since it's flashcard function is superior for the most part. Or if anyone knows of any other good flashcard programs for the palm, please let me know so I have something new to hold me over till' Pleco 2. My current program is circa 1998.

Did you read my first post that started this thread? I don't use repetition spacing, either, for the same reason you don't. It's meant for those with a daily study schedule and not useful if you are only studying sporadically.

 

[mikelove]

mulechicken - This should get considerably better in 2.0, since our flashcard system is now hierarchical like Flash! et al and lets you organize your flashcard categories into folders - you could create a category for each chapter, arrange them together under a single folder for the book and then easily do a review of the book or of the individual chapters.

I still wish we could find a way to make repetition spacing better for casual users, though since a well-designed frequency-adjustment system should have pretty much the same effect I suppose it's probably better to just put our efforts into refining that. I don't think it's possible to do anything but repetition spacing in SuperMemo, they're pretty seriously committed to that philosophy, though I'll admit I haven't spent too much time with their most recent versions.

 

[ldolse]

Just a thought for the casual user and repetition spacing - I get bit by this issue as well. Wouldn't a solution here be allowing the definition of a 'day' through a checkbox or something like that? Seems to me the basic problem is that it looks solely at the time passed since the last session, instead of just looking at how many sessions it's been since a given card was reviewed.

Seems to me there could be two options:
- Base repetition spacing off the calendar date/days since last session
- Base repetition spacing off an increment of one since the last session (maybe with a small upfactor for number of calendar days passed, not sure)

I suppose since a more focused user studies daily the configurable option might not even be neccessary. I haven't really studied the way the repetition spacing algorithms really work under the hood though, maybe that solution is too simplistic.

 

[mulechicken]

Thanks for the replies people. I think with my current study habits and after reading some of the other posts a bit, I'm going to try to use the Pleco Dic flash card program for a bit since it's easier to set parameters (e.g. test pinyin but not character etc.) and repetition spacing does have benefits.

Right now I'm just trying to learn ten characters a day, every day. No two character words, just single characters. What I do is pick up a book, newspaper or magazine; read through and note the first 10-15 characters I don't know or am not sure about; convert them into flash cards; then test myself a couple of times throughout the day. Only takes about 20 minutes a day to set up each new list. Since each list is only 10-15 characters, it's easy to do lots of short tests in a single day. This way, even if I only make lists of 15 characters two out of three days a year, I'm still learning almost 3,000 characters. And even if I only retain half of those, it's a still an extra 1,500 characters on top of the ones I have. Another reason I'm going to try the Pleco flashcards for a bit, is this would work well with some repetition spacing, then I'd just use Flash! to review all my old cards.

It's all about short-term achievable goals.

To test my ability to write the character i test myself on the pinyin, then "write" the character in my head, counting each stroke as I do it. Even if I'm sure I know the character, I still count out each stroke. The counting makes it a very deliberate process, and seems to make them stick better. Yea, writing it out of course is good too, but this way lets me improve my writing with hands free (walking, on the bus etc). Seems to work pretty well, let's see if I can keep it up.

Mikelove, I'm pumped about Pleco 2, it looks really good and the flash card labels should make it much more manageable for the mildly committed student, can't wait to try it out. Its also good you can leave a test and go back to it later.

In Pleco 2, can you use hardware buttons on the treo to activate "音 and 意" in addition to "right and wrong"? It would be handy to just be able to use hardware buttons for those so it can be completely one handed use.

 

[mulechicken]

Ah, never mind the last bit of my previous post. In flash card testing I can choose to show pinyin, character or definition based on how the "previous" and "next" buttons are assigned under preferences. Word.

 

[mulechicken]

Is there any way to retest mistakes the same test in Pleco dic? Is there a way to make it so, if I'm testing myself on 20 cards, it'll keep throwing the ones I get wrong back in pile and bringing them up, only ending the test after I've gotten all 20 of those cards right at least once.

 

[mikelove]

ldolse - good idea, but probably not something we can implement this late in the game, so it'll have to wait for a later release.

mulechicken - yes, and it should also be possible to remap hardware buttons to 音/意/etc in the finished version of 2.0. Though we're still trying to figure out how exactly the hardware button mapping system should work - the current way we've designed it is way too complicated to be useful. There's no good way to retest mistakes in the current version - you might be able to cobble something together using the "reviewed after" option in Card Settings along with customizing ranks to promote a card to the next rank for every right answer and demote a card to rank 1 whenever it was incorrect, but it wouldn't work very well and wouldn't let you use any other rank optimization system at the same time. In 2.0 we're adding a prompt at the end of the session which will re-test you just once on all of the cards you got wrong; we thought about looping through again and again until you'd gotten every card right once, but with some of the new test modes (especially stroke order and free-answer) that could really get annoying so we left it with just the one repeat.

 

[mulechicken]

Mikelove, thanks for the replies, Pleco 2 sounds even better than ever.

About repetition spacing. Why is repetition spacing based on days? Wouldn't it make more sense to have the option for repetition spacing to be based on number of tests instead of days? So if you spend a whole afternoon taking several tests on a batch of cards, the ones you get more correct will be less likely to show up in each test, but will still show up if you keep testing enough, as opposed to waiting a fixed amount of days before they come up again. This would also solve the problem of people being away from Pleco for a while, then having to catch up on their repetition spacing.

I like how the hardware mapping is laid out in Pleco dic 1.0 it's pretty useful. I can't stand it when when Palm OS programs don't have good hardware button shortcuts.

Glad there will be a "re-test" option at the end of the tests for learning. Would it be possible to say, when testing a list of 20 cards, limit the test to 40 questions (or have it be for a fixed # of cards with a value of 40). The test would then retest mistakes out of that stack of 20 cards, shuffling wrong answers back in as you go, but would end after 40 questions no matter what.

Mikelove do you ever have time to keep up with your Chinese? Or has all this work making dictionaries turned you off to hanzi? Just curious.

 

[mikelove]

It's based on days because the theory behind it suggests that you improve your memory by reviewing words at progressively longer time intervals; helps to gradually shift something from your short-term to long-term memory, or so they say. If you detach it from days then there's no longer any guarantee that the spacing between card reviews will get progressively longer.

Glad you like the current hardware mapping system - the main question is whether or not we can keep it that simple with a larger and larger number of available commands, but in general I prefer function-to-key mapping like we use in 1.0 over key-to-function mapping like we currently use in 2.0.

A re-test option like you describe might be possible, but it's not something we're likely to do in 2.0 - we've already got way too many options and aren't adding any others unless there's an overwhelming amount of customer feedback.

I don't get to keep up with Chinese as much as I'd like, no - hasn't really turned me off of hanzi, I'm just way too busy lately.