Tuesday, March 31, 2009

Input X Output

If we weight the importance of output and input, input would comprise with 98% of the efficacy, output would be 1% and more 1% to unknown factors.

It is not that you can or not become fluent from input only. It is that the best way to achieve it is to focus at least 98% of your effort on input.

980 hours of input and 20 hours of output have much better effect than 500 hours of input and 500 hours of output.

It's not a question of belief. It is a fact.

Yes, my numbers are skewed, but it is not possible to account it perfectly.
Anyway, probably 1% for speaking practice is already way too much.

For Japanese I've read far more than 30000 sentences.
Right now, only my sentences deck account for more than 3000 sentences.

Also, I've listened far more Japanese that I have read.

I'll take some lifetimes to do the same with output.

Thursday, March 26, 2009


Well, all I can say is that I already  can speak a little of Japanese. 

I have not put it on trial yet, but I plan to do it soon.

Until now, the only interactive output I had was chatting on the net, only for a few days, months ago.
I stopped because I noticed that it had no effect on my Japanese. Just reading and listening were working very well.

But now something is changing. I'm talking with myself a lot in Japanese lately. I do not do it actively. 

I do it without noticing. 

Btw, this made me recall that before I could speak English, I ranted with myself the same way I'm doing with Jap now.

It started without form. At first there were only sounds. But these sounds kinda fit into each other. Lately these sounds are becoming words. For now, it feels that I'm fooling myself that I can speak Jap, but my hope is that these grunts will become proper speech.

Input before output. But maybe, the time for output is near.

Friday, March 13, 2009

Progress Report: The News and Output

I'm doing fine reading the news. It depends of the article. Sometimes I might not need the dictionary at all, somemtimes I'm forced to look up more than 20 items.

Listening is not going so well. It is much harder to understand it listening. It's too much information compressed in so few words. I guess now it has been 4 months since I started the news operation, but I think I'll still need more 4 months.

On the other hand, anime and dorama are becoming very easy to understand.
Sometimes I don't need make any effort at all to get probably more than 95% of understanding.

I think this is finally triggering my output. I feel like soon I'll be able to construct my thought directly in Japanese. I'll try to write something once a week in my lang-8 journal to measure how I'm going.

Finally, the anki stats:

The 3210 seen cards in this deck contain:
* 1968 total unique kanji.
* Jouyou: 1737 of 1945 (89.3%).
* Jinmeiyou: 57 of 287 (19.9%).
* 174 non-jouyou kanji.

Jouyou levels:
* Grade 1: 80 of 80 (100.0%).
* Grade 2: 160 of 160 (100.0%).
* Grade 3: 200 of 200 (100.0%).
* Grade 4: 199 of 200 (99.5%).
* Grade 5: 185 of 185 (100.0%).
* Grade 6: 180 of 181 (99.4%).
* JuniorHS: 733 of 939 (78.1%).

Wednesday, March 11, 2009

Ow, no! My Japanese is getting too good to keep playing games!

After a long and sad time without a TV, I finally manage to get one and turn my playstation on this year. I put persona 4 in there and started to play it.
The last time I was playing it was awesome. Every word was new. Every kanji had a new reading. Every expression had to be decifered. The game skills, items, moves, options? They were incomprehensible.

Now I can understand pretty much everything. Even the curvy balls like 何処 are not so curvy anymore.

So my question is. Does anki + massive listening for fun rip off the challenge from language learning?

Btw, I'm almost giving up on French. It's so hard! Japanese is so comfy!

Tuesday, March 3, 2009

Japanese names

It's been hard to find new kanji in texts. Vocabulary never ceases to appear, but kanji is kinda different.

So, I started to tackle names.

Japanese names are interesting. I'm not sure how it is yet, but it seems there are many different ways to write the same name so I'm not intending to be able to do that. I just want to be able to read them, even without 100% accuracy.

You may call me a cheater again but I fixed a file with the names of a few hundreds of japanese writers and just imported them in my deck.

Kanji Stats:

The 3100 seen cards in this deck contain:

* 1931 total unique kanji.
* Jouyou: 1724 of 1945 (88.6%).
* Jinmeiyou: 46 of 287 (16.0%).
* 161 non-jouyou kanji.

Jouyou levels:

* Grade 1: 80 of 80 (100.0%).
* Grade 2: 160 of 160 (100.0%).
* Grade 3: 200 of 200 (100.0%).
* Grade 4: 199 of 200 (99.5%).
* Grade 5: 185 of 185 (100.0%).
* Grade 6: 180 of 181 (99.4%).
* JuniorHS: 720 of 939 (76.7%).