Saturday, February 9, 2019

A4.4.35 meta: o pokroku

Alright! I have been looking forward to writing this post all day. And finally I get to do it!

I've been thinking a lot about my language learning "pipeline." This is one of my favorite things to think about.

There are several phases in the pipeline:

1. EXPOSURE
Watch a movie, read (part of) a book, work on a translation, write an email, read an email, text, talk with people, etc. This is the most fun part.

2. PLAY WITH THE WORDS
Try (and mostly fail) to keep a meticulous list of unknown words and phrases. Find their frequencies in the cs SkELL 2.2 corpus. Perhaps there are flaws with using this corpus, but it's certainly better than no corpus!

3. RE-EXPOSURE
This is the step I have been really struggling to figure out! The fact is, I read a lot and talk to a lot of people. The vast majority of the time I read, there are new words (and sometimes ideas!) that I don't know. I put these words into the pipeline and then sometime later I figure out what they mean. That's not good because I read so much more fluently when I know what the heck the words mean. I had been recording myself reading various texts. I think it would be more effective studying for me to record myself rereading texts. This will require two things: keep better records of step 1. I don't always write down exactly what I'm reading. I just kind of go for it. The flexibility of the pipeline is important. It can remain flexible, but I should create a tab on my table that records what I'm reading the previous week. The second thing that I need to do is be more diligent at spending time daily in this re-exposure phase.

4. QUIZLET
This step is fantastic but there are some flaws in it as it currently stands. First, I don't typically get the corrected translations of 10 sentences to a willing collaborator until Thursday. That's too late in the week. I need to get it to them by Monday or Tuesday to allow them more time to record themselves. Second, I have been making the false assumption that more is better. It really is not. As soon as I stopped taking 50 question quizzes in that pády app it became easier for me to see and remember patterns! I have to, therefore, be much more choosy with my sentences.

Good sentences for Quizlet are:
- short!
- natural Czech constructions! These sentences will stay in my head for weeks, months - hopefully longer. I can easily pattern other sentences on them.
- something I'm likely to say! It is really stupid to talk about the state of something's airspace or compulsory liability! I intuitively know that I don't talk about those things!

I think it might be easier to recruit willing, computer literate volunteers to record themselves saying 10 sentences rather than 32+. My best collaborators on this end speak clearly and get it done in a relatively timely manner. I have only had one woman record her voice, which is a huge hole in my language production. I am a woman. I want to sound like one. I guess it won't make an enormous difference to my production. But it would be great to get a variety of voices. On a daily basis I get 2+ messages from willing collaborators who want to help me out. We'll see if the spirit is willing and the flesh isn't incompetent.

5. USE IT IN CONTEXT
This step is ridiculously fun. So far all my words have been high enough frequency that it's possible to slyly slip them into the text of an email (or some of them into one email, others into another) or into a chat conversation. I try to avoid it with people who might be reading this blog because, well, that's...that's awkward. It's like slapping something with a sticker that says, "OUR FRIENDSHIP IS A LANGUAGE EXPERIMENT." No thanks! My friend Petr slyly slips the words into our conversations, though. I didn't notice it at first, but then when I did, it cracked me up. It's fully okay in reverse that way. There is absolutely no way for me to signal to new collaborators that I would like them to do that. It's not a natural way to use language.

6. RE-TEST
This step should be quick and easy. It includes gathering my test mistakes and correcting them. The test from a native speaker is usually really short and simple. I have been doing this on Mondays.


Actually, let's see the breakdown for how things have been going:

1. EXPOSURE - Continuously, every day
2. PLAY WITH WORDS - Monday-Wednesday
3. RE-EXPOSURE - Not happening consistently
4. QUIZLET - Thursday-Saturday
5. USE IT IN CONTEXT - Friday-Saturday
6. RE-TEST Monday of the following week

I want to try to get things to work out more like this:

1. EXPOSURE - Continuously, daily, but slightly more purposefully. I will start to keep an "exposure" journal. That sounds really risque to be honest. Haha.
2. PLAY WITH WORDS - Monday-Tuesday
3. RE-EXPOSURE - Monday-Saturday
4. QUIZLET - Wednesday-Saturday
5. USE IT IN CONTEXT - Wednesday-Saturday
6. RE-TEST - Saturday

But I realize that I'm an idealist and things don't often, or even usually, turn out how I plan them. That's okay.

Now I'm going to attack my really nerdy pipeline spreadsheet. It is really addictive to "perfect" a spreadsheet.

No comments:

Post a Comment