magamo wrote:

Tobberoth wrote:

magamo wrote:

I think blackmacros meant a learner uses SRS so 90% of information is supposed to be retrievable at any given moment. And he or she is not that serious about these numbers, I guess.

Probably, because doing that in an SRS would be torture.

Doing RtK1 in 20 days require about 100 new kanji each day. That means that on that 20th day, you're probably averaging 300 kanji reviews each day. Add in the kanji that you fail and you'll have a lot more.

That's still bearable though. So let's add BOTH KO books in 30 days... the reviews would be insane, probably over 700 reviews each day, maybe 1000 or more. No one could learn in such a situation, certainly not enjoy it.

1000 reviews! I definitely don't want to do it!

By the way, if we assume that an SRS algorithm is optimized so that you review information as less frequent as possible, learning the same amount of information in another way is bound to be less or equally efficient. Otherwise, the premise that the algorithm is more efficient than usual rote learning is wrong. In other words, if it seems to be unfeasible with SRS, then you'll fail harder without SRS. The 3 months limitation may make a difference, though, because SRS isn't good for short term memory.

Exactly, that's why passing JLPT1 in 3 months is ridiculous. It's not even possible in an SRS where the whole point is to study as much as possible with as little effort as possible. I don't think SRS is much weaker in short term memory than any other system, you should only enter stuff you know, and once it's entered, you should never fail it (well, retention limits are still there, but 90% retention is still much higher than what you'll get in most other techniques).

jmkeralis wrote:

I think the movie reference you guys are looking for is The Perfect Score, a movie made by MTV about 6 high school seniors who steal the answers to the SAT (with Scarlett Johansson).

If you figure out a way to pull it off, though, I'm totally in.

Nah... I figure it's gotta have the word "JOB" in it. It's what hammers down that imagery of capery. The Perfect Score is like, what basketball, football, that chick anatomy class. It's too vague, you know. Wait.... The Perfect Score... Job....

Nahhh.....

Aijin wrote:

Can someone explain 'SRS' to me? I've never come across the acronym before.

Oh, sorry. I just couldn't imagine a member of this forum didn't know SRS...
I think you get the idea better if I explain it in your first language:

えっと、フラッシュカードって知ってますよね？一昔前には日本でもあったと思うのですが
表面に単語を書いて裏に翻訳書いたりするやつです。あれのコンピューター版とでも言えば
わかりやすいかも。

別に「単語と訳」のようなペアでなくてもよくて、どんなことを書いてもいいですし
音声、画像も取り込めます。なかには「裏面」という概念をとりはらって、「面」が二つ以上
あるものもあります。たとえば、表面に外国語文章、裏面１に音声、裏面２に言い換え、
裏面３に辞書の定義、みたいな使い方もできます。

特徴的なのは、どのSRSもいわゆる忘却曲線というものから理論的に計算して、
「ちょうどカードの内容を忘れたころ」に復習させてくれるようになっています。
たとえば、新しくカードを作ると、きっとそれを翌日復習することになりますし、
そこでうまく覚えていたら、次は3日後、その次は1週間後、1ヶ月後、のように
どんどん理論に沿って忘れるであろう瞬間にだけ復習すればよいというシステムです。

ここまで聞くと当たり前に聞こえるかもしれませんが、びっくりするぐらい暗記が楽に
なります。SRSをはじめに広めた人の端的な説明だとYou forget about forgetting.
（だったかな？SRSでカード化しなかったから忘れちゃった！）

以前AJATTという一年半で日本語が流暢になったというブログを紹介しましたが、
そこにも詳しい解説がありますよ。一度使えばわかるのですが、語学に限らず
大抵の勉強は圧倒的に楽になります。まあ、誰にでもうまくいくとはいいませんが、
上手にSRSを使えない人はものすごく損をしてるだろうなぁとは思います。

日本では本当にマイナーですね。知ってても秘密にしたいという人も少なくないですし。

ちなみにこのサイトは、SRSで漢字を覚えよう！という趣旨ですので、知らない人がいるとは
思わなかったんです。ゴメンナサイ。

Last edited by magamo (2009 June 02, 3:01 pm)

mentat_kgs wrote:

Academia does its best to avoid it

I was in full-time education for sixteen years and never heard it mentioned once. I don't know why it's such a well-kept secret...

mentat_kgs wrote:

I wished I knew about it about the time I entered college.

Yeah me too....

Aijin wrote:

Research has proven that memory consolidation can take decades before information is stored more permanently in the neural networks. It may be possible to memorize many characters and their meanings, but to keep it at a level beyond working memory you can't simply spend 20 days learning them and then hold that information firmly for the next 70 days. There would have to be constant daily practice of all the information, and there simply isn't enough time.

SRS allows you to learn anything forever - with as few as 7 reviews per year!

Aijin wrote:

I've visited most of the Japanese departments of the world's finest universities, and none of these feats are done by students, nor are these methods used by professors, who have devoted their entire lives to the pedagogy of the language.

Most of the Japanese departments of the world's finest universities?!

1. How long did you stay there?

2. There are many effective methods unknown for many people (or most of them):

SRS, L-R, Iversen, Ilya Frank, Shadowing, Scriptorium, Heisig's RTK, MM, etc.

Just because many people (students or professors) do not know about something, does not mean it is not effective.

Aijin wrote:

I don't know of any Japanese scholars who could keep up such a high level of memorization even though it's their native language.

Do Japanese scholars know everything?

Did you forget that the Japanese need 12 years of study to learn the kanji, while many of us mastered it in few months - 2 years?

Aijiin wrote:

Can someone explain 'SRS' to me? I've never come across the acronym before.

Read this and you will change your mind about memory and memorizing:

http://www.supermemo.com/english/princip.htm

http://www.supermemo.com/articles/horod.htm

http://www.supermemo.com/articles/kowal.htm

Forget about forgetting!

harhol wrote:

I don't know why it's such a well-kept secret...

mentat_kgs wrote:

I wished I knew about it about the time I entered college.

mentat_kgs and harhol, you stole my line...

Tobberoth wrote:

Aijin seems to be completely fluent in English. I doubt your Japanese essay makes any difference tongue

But I can't explain it better in English! Me speaks poor engrish.

Aijin wrote:

Very interesting. It's strange that I've never heard of this technique in language programs before. Even the psychology courses I've taken haven't mentioned this even in passing. I'll try to read up on it a bit more and then give it a shot tonight. Is it meant to be used for learning, or simply for recall/recognition; should I only add information once I've learned it/become familiar with it, or should I add unknown information to it as well?

You can use it to learn, but that's ineffective. You should enter stuff you know and it will make sure you won't forget it.

That does not mean you have to train facts over and over before entering them, it just means you have to make sure you fully understand the facts. For example, you could take a random sentence and enter into your SRS, then fail it constantly until you know it. However, it makes more sense to read a sentence and make sure you understand every part of it, maybe even make some mnemonics and then enter it. A regular SRS will assume you will remember the fact for 3 days after the initial review, so it's most effective with stuff you're familiar with.

That's why we use stories and the heisig technique on this site. Simply entering the kanji into an SRS is theoretically enough, but making a proper mnemonic saves a LOT of time and effort.

Aijin wrote:

Very interesting. It's strange that I've never heard of this technique in language programs before. Even the psychology courses I've taken haven't mentioned this even in passing. I'll try to read up on it a bit more and then give it a shot tonight. Is it meant to be used for learning, or simply for recall/recognition; should I only add information once I've learned it/become familiar with it, or should I add unknown information to it as well?

My understanding is that it's recall/recognition. But if you use it for language learning, you'll run across the same words/phrases/grammar points/ again in a year, and so the information you put into SRS will naturally be "learned." If you use it for math formulas and didn't use them to prove moth problems, then they'll never be learned though you can recall them all.

By the way, my first language is JAPANESE if anyone didn't know it. That's why I posted the complicated thing about SRS in Japanese...

Last edited by magamo (2009 June 02, 4:10 pm)

magamo wrote:

Aijin wrote:

Very interesting. It's strange that I've never heard of this technique in language programs before. Even the psychology courses I've taken haven't mentioned this even in passing. I'll try to read up on it a bit more and then give it a shot tonight. Is it meant to be used for learning, or simply for recall/recognition; should I only add information once I've learned it/become familiar with it, or should I add unknown information to it as well?

By the way, my first language is JAPANESE if anyone didn't know it. That's way I posted the complicated thing about SRS in Japanese...

Oh, well that's completely different. I pointed it out because I found your post to be condescending since I thought English was your first language.

(I mean, think about it, working in Japan as a westerner and being completely fluent in Japanese, and when you ask a Japanese person about something they go "Well, I think you will understand it better in your first language" and then they answer in broken English).

How cool, at least 3 people on this forum has Japanese as their first language... one might wonder why such people visit a forum such as this though. (Of course, you're all more than welcome )

ahibba wrote:

Did he answer in broken Japanese?!

Not at all, I was just giving an example of why it would be condescending.

ahibba wrote:

1. Aijin. 2. magamo. Who is the third?

Masaman.

It's been said many times on this forum, I'll say it also: "It's both comforting and depressing at the level of quality English being offered up by those that have it as a second (or later) language."

Magamo, thanks for the Japanese description of an SRS. It might encourage guys like Alyks to come back to this forum more often if there were more posts and discussions in Japanese.

Ahibba, don't confuse what we do with RTK (writing with recognition being an English keyword) with what the Japanese school system teaches (writing, pronunciation, use in words, etc). In addition, that's just the school system formally covering it. Something tells me Japanese school children know many of the kanji before they're taught them. Only reason I'm saying this is your post seemed a bit arrogant (also I said such things in the past and regret them now in hindsight).

Last edited by Nukemarine (2009 June 02, 5:33 pm)

I'm not so sure this  is such a good forum (or any forum really) to be mining English from.  Though some of the non-native users here have better grammar than those of us who grew up in an English speaking country, most of the active users are not native speakers.  We are not nit-picky about it though (I would have been torn apart long ago).

I hope none of my convoluted, dry sentences have ended up in some well meaning English student's SRS ...