words: 4327 (30% unknown)
So, there are now 3020 words that I "don't not know." But a lot of them are floating around in a miasma where I can easily confuse them or forget them.
Like "takviye" (reinforcement) and "tavsiye" (advice).
On the other hand "rastgele" (random) and "muadil" (equivalent) I never forget. Why? Rastgele is a bit unusual looking, but there are lots of mua* words.
How are things going over all? Şöyle böyle.
Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: B2 by 20190401 via flashcards
You do not have the required permissions to view the files attached to this post.
2 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: B2 by 20190401 via flashcards
4448 words (22% unknown)
As mentioned before, my reading has improved markedly from use of these flashcards. I still read at a beginner level, but now it's advanced beginner etc. I read most of the illustrated classic for children Sefiller before I got bored. I still need another 3-5000 words and have decided to get them from "intensive reading of subtitles."
I went back to Oyunbozan. On maybe half the subtitles I don't even need to look anything up, I can just read them and read them pretty quickly. This is part of my haul from the last couple days:
manşet -- headline
nakli -- transplant
rezil olmak -- to be disgraced
patınaj -- skid
sabote etmek -- to sabotage
dayak -- beating
saygılar -- respects!
haklısın -- fair enough
kraliçe -- queen
düşünsene! -- imagine!
Listening ability is now lagging far behind, though.
As mentioned before, my reading has improved markedly from use of these flashcards. I still read at a beginner level, but now it's advanced beginner etc. I read most of the illustrated classic for children Sefiller before I got bored. I still need another 3-5000 words and have decided to get them from "intensive reading of subtitles."
I went back to Oyunbozan. On maybe half the subtitles I don't even need to look anything up, I can just read them and read them pretty quickly. This is part of my haul from the last couple days:
manşet -- headline
nakli -- transplant
rezil olmak -- to be disgraced
patınaj -- skid
sabote etmek -- to sabotage
dayak -- beating
saygılar -- respects!
haklısın -- fair enough
kraliçe -- queen
düşünsene! -- imagine!
Listening ability is now lagging far behind, though.
0 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
4756 words (18% unknown)
I need to get a couple thousand more words from somewhere. I tried getting them from a dictionary, but I invariably chose bookish words that are not normally used in conversation. I tried Turkish newspapers, but their articles are interspersed with a lot of graphics and links (at least what I looked at). It's distracting.
Then I tried intensive reading of subtitles which allowed me to watch Oyunbozan while using mining of words as an excuse. This has the advantage that any word I find is 99.9% guaranteed to be a word in common parlance. The disadvantage is that it's very very time consuming. At this point I'm getting less than 100 words per episode, and actually rolling all the scenes back and forth takes more than two hours. I just don't have the time at present.
Today I tried BBC Turkish. Nice compact articles. I can copy and paste chunks into Google Translate, figure out which words I don't know, done. This is probably the most efficient way.
I need to get a couple thousand more words from somewhere. I tried getting them from a dictionary, but I invariably chose bookish words that are not normally used in conversation. I tried Turkish newspapers, but their articles are interspersed with a lot of graphics and links (at least what I looked at). It's distracting.
Then I tried intensive reading of subtitles which allowed me to watch Oyunbozan while using mining of words as an excuse. This has the advantage that any word I find is 99.9% guaranteed to be a word in common parlance. The disadvantage is that it's very very time consuming. At this point I'm getting less than 100 words per episode, and actually rolling all the scenes back and forth takes more than two hours. I just don't have the time at present.
Today I tried BBC Turkish. Nice compact articles. I can copy and paste chunks into Google Translate, figure out which words I don't know, done. This is probably the most efficient way.
1 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
4883 words (4% unknown)
When I started this experiment, the reviews were taking 45 minutes to an hour. Over the last few weeks, though, they ramped up inexorably. It's taking close to two hours now. Partly this is due to brain overload and too many failed reviews, partially it might be some weird problem with the Quizlet black box (I'm learning to appreciate the transparency of Anki, where I can at least see what's going on). And partly it's just boredom ... I've been going for three months now. My brain is starting to fight back and say "I want to watch some TV instead."
I have seen substantial reading gains. I can take a BBC Turkish article and read it and have some idea what it's about. I was a real A0 prior to this experiment, so that's a definite win. I think the listening gains are more marginal, but then how would I really know. I've watched maybe 10 hours of Turkish TV total in the last three months.
The big problem is that I'm about to run out of new words, and will have to go back to mining them. This will add another 30 minutes or more to my 2 hours of reviews. I've already dropped the dabbling experiment completely (will post on that thread to close it off later), but Turkish is now cutting into Russian and Italian time pretty significantly. I console myself that it's temporary.
At this point I'm shooting for 6000 words. I'll mine the rest of them out of BBC Turkish, which will give me reading practice ahead of the big GLOSS ODA. I'm pretty confident I'll be able to get a 1+ by then ... the question is can I get a 2? Hope springs eternal.
When I started this experiment, the reviews were taking 45 minutes to an hour. Over the last few weeks, though, they ramped up inexorably. It's taking close to two hours now. Partly this is due to brain overload and too many failed reviews, partially it might be some weird problem with the Quizlet black box (I'm learning to appreciate the transparency of Anki, where I can at least see what's going on). And partly it's just boredom ... I've been going for three months now. My brain is starting to fight back and say "I want to watch some TV instead."
I have seen substantial reading gains. I can take a BBC Turkish article and read it and have some idea what it's about. I was a real A0 prior to this experiment, so that's a definite win. I think the listening gains are more marginal, but then how would I really know. I've watched maybe 10 hours of Turkish TV total in the last three months.
The big problem is that I'm about to run out of new words, and will have to go back to mining them. This will add another 30 minutes or more to my 2 hours of reviews. I've already dropped the dabbling experiment completely (will post on that thread to close it off later), but Turkish is now cutting into Russian and Italian time pretty significantly. I console myself that it's temporary.
At this point I'm shooting for 6000 words. I'll mine the rest of them out of BBC Turkish, which will give me reading practice ahead of the big GLOSS ODA. I'm pretty confident I'll be able to get a 1+ by then ... the question is can I get a 2? Hope springs eternal.
5 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
- reineke
- Black Belt - 3rd Dan
- Posts: 3570
- Joined: Wed Jan 06, 2016 7:34 pm
- Languages: Fox (C4)
- Language Log: https://forum.language-learners.org/vie ... =15&t=6979
- x 6554
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
Happy New Year.
ACTFL Proficiency Guidelines and samples
https://www.actfl.org/publications/guid ... lines-2012
https://www.actfl.org/sites/default/fil ... sments.pdf
ACTFL Proficiency Guidelines and samples
https://www.actfl.org/publications/guid ... lines-2012
https://www.actfl.org/sites/default/fil ... sments.pdf
1 x
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
Thanks, reineke!
I looked at the samples. Intermediate is pretty easy. Probably the sample is not "intermediate high", but ... yes I could probably pass with a 1+ today.
On the other hand, the advanced sample is tough. I don't know if it corresponds to a 2 or a 2+. I can get some of the gist, but I'd fail a test on it ... at the moment. Unfortunately, there's a problem with some unknown vocab, but the bigger problem is the grammar is harder. Anyway, I have 10 more weeks.
I looked at the samples. Intermediate is pretty easy. Probably the sample is not "intermediate high", but ... yes I could probably pass with a 1+ today.
On the other hand, the advanced sample is tough. I don't know if it corresponds to a 2 or a 2+. I can get some of the gist, but I'd fail a test on it ... at the moment. Unfortunately, there's a problem with some unknown vocab, but the bigger problem is the grammar is harder. Anyway, I have 10 more weeks.
1 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
This experiment terminated abruptly.
Over the course of the experiment, the time needed for flashcard reviews grew from 45 minutes at the beginning to 2.5 hours at the end. Partly this was due to the success rate going down under the barrage of 70 new words a day, partly it was just boredom and exhaustion. I killed off the dabbling experiment to keep this experiment running longer, but it wasn't enough. Towards the end I was skipping Russian and Italian -- Turkish flashcard review was what I did for my whole day's study.
I hit a couple back to back days where due to work I was unable to complete my reviews, and then within five days I realized I was never going to be able to catch up.
I also realized the goal of hitting a B2 reading level was going to be unobtainable. All these flash cards did give a big boost to my reading level (maybe from A1 to B1) but to move higher I would have to understand Turkish grammar better, and that was outside the scope of the experiment.
Pros:
Flashcard cramming will boost reading ability noticeably, particularly from low beginner levels.
The boost to listening comprehension is smaller, but still noticeable. It doesn't help with deciphering sentences, but various key words do pop out at you.
Cons:
Tedious
The words seem to be held at a superficial, temporary level. I feel like I've already lost a lot of them, and if I were exposed to no Turkish at all for a month they'd all be gone. Of course a lot of my cards never got into the mature stack, so there's that.
Conclusions:
It doesn't suit my nature. It could work for others. I would only recommend it for type A personalities with a lot of patience who have some explicit goal like passing a test in a given time frame.
Going forward, I plan to continue with the lazy slug approach I first tried out with Italian. And I'm free from any desire to be more efficient or finish faster or take any exams.
Over the course of the experiment, the time needed for flashcard reviews grew from 45 minutes at the beginning to 2.5 hours at the end. Partly this was due to the success rate going down under the barrage of 70 new words a day, partly it was just boredom and exhaustion. I killed off the dabbling experiment to keep this experiment running longer, but it wasn't enough. Towards the end I was skipping Russian and Italian -- Turkish flashcard review was what I did for my whole day's study.
I hit a couple back to back days where due to work I was unable to complete my reviews, and then within five days I realized I was never going to be able to catch up.
I also realized the goal of hitting a B2 reading level was going to be unobtainable. All these flash cards did give a big boost to my reading level (maybe from A1 to B1) but to move higher I would have to understand Turkish grammar better, and that was outside the scope of the experiment.
Pros:
Flashcard cramming will boost reading ability noticeably, particularly from low beginner levels.
The boost to listening comprehension is smaller, but still noticeable. It doesn't help with deciphering sentences, but various key words do pop out at you.
Cons:
Tedious
The words seem to be held at a superficial, temporary level. I feel like I've already lost a lot of them, and if I were exposed to no Turkish at all for a month they'd all be gone. Of course a lot of my cards never got into the mature stack, so there's that.
Conclusions:
It doesn't suit my nature. It could work for others. I would only recommend it for type A personalities with a lot of patience who have some explicit goal like passing a test in a given time frame.
Going forward, I plan to continue with the lazy slug approach I first tried out with Italian. And I'm free from any desire to be more efficient or finish faster or take any exams.
7 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
- reineke
- Black Belt - 3rd Dan
- Posts: 3570
- Joined: Wed Jan 06, 2016 7:34 pm
- Languages: Fox (C4)
- Language Log: https://forum.language-learners.org/vie ... =15&t=6979
- x 6554
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
I am thoroughly impressed. You managed to keep my interest notwithstanding the title which is crammed full of detestable things (except for the language itself which simply does not interest me). Since we've discussed the problem of testing extensively here, would you care to complete that Arealme test and the 17 minutes test? Both should only take 5-7 minutes of your time or two Peppas max.
Thank you.
Thank you.
0 x
-
- Blue Belt
- Posts: 821
- Joined: Tue Oct 06, 2015 1:19 am
- Languages: ru it tr
- x 2221
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
arealme: 4320
17 minute test: 2200
I do feel the decay rate is fierce when you quit. I probably should have tested a couple weeks ago. It's like "Flowers for Algernon."
17 minute test: 2200
I do feel the decay rate is fierce when you quit. I probably should have tested a couple weeks ago. It's like "Flowers for Algernon."
2 x
Ещё раз сунешь голову туда — окажешься внутри. Поняла, Фемида? -- аигел
- reineke
- Black Belt - 3rd Dan
- Posts: 3570
- Joined: Wed Jan 06, 2016 7:34 pm
- Languages: Fox (C4)
- Language Log: https://forum.language-learners.org/vie ... =15&t=6979
- x 6554
Re: Turkish Cramming Experiment: ILR 2 by 20190401 via flashcards
February 2018
Turkish
I recognized the cognates and I guessed everything else. One question had Mustafa/Kemal/Ataturk as 3/4 possible answers.
I can assure you that after 10-30 hours my ear would still only be getting used to the language and the score wouldn't budge. On the other hand I was wondering about the potential usefulness of SRS/Anki. This was a pretty disappointing result especially when you consider you'd have a much lower chance to catch these words when listening.
Korean
I blindly clicked around (towards the center of the screen) since I cannot read Korean. I retook the test and I spread my answers around. I got 1410.
Polish
Czech: 4830
Dutch: 4750
With Czech, I understood some of the headwords and several synonym/antonym options, but I apparently still made the wrong choice. Some international words did not seem to have corresponding antonyms or cognates. It is possible I misunderstood some words. In Polish, the written word for "sun" may be misinterpreted as meaning elephant by speakers of other Slavic languages. That's the only Polish trick I know.
English + German did not earn me a very high score with Dutch.
I would have expected a higher result for Czech and Polish given my background in two Slavic languages. I would say that the test developers were generous with some easily recognizable international cognates. On the other hand I didn't see many easy interslavic cognates. Generally speaking, Romance and Latin cognates are easier to spot than Slavic cognates.
In any case, currently I find written and spoken Polish opaque. I used to think written and spoken Russian and spoken Portuguese were opaque when I started listening to these languages. Spoken and written Spanish felt transparent* from day 1.
*European Spanish was less transparent.
6/3/18
Polish
Total root Polonization time: 10+ hours? Polish is less opaque. Edit: No, I did not learn 10,000 words in 10 hours. The result reflects the fact that I can now navigate Polish a lot better than before.
I wrote that comment because Ani was like "Whaaat?!" That score reflected the ability to sort of follow some audio.
After around 500 hours of playing with Russian I got this:
Russian
My conclusion is that you are better off without SRS/Anki. Period. If you want deep roots that come alive in contact with input you'll need to do a lot more than flip flashcards.Or a lot less depending on how you look at it.
Turkish
I recognized the cognates and I guessed everything else. One question had Mustafa/Kemal/Ataturk as 3/4 possible answers.
I can assure you that after 10-30 hours my ear would still only be getting used to the language and the score wouldn't budge. On the other hand I was wondering about the potential usefulness of SRS/Anki. This was a pretty disappointing result especially when you consider you'd have a much lower chance to catch these words when listening.
Korean
I blindly clicked around (towards the center of the screen) since I cannot read Korean. I retook the test and I spread my answers around. I got 1410.
Polish
Czech: 4830
Dutch: 4750
With Czech, I understood some of the headwords and several synonym/antonym options, but I apparently still made the wrong choice. Some international words did not seem to have corresponding antonyms or cognates. It is possible I misunderstood some words. In Polish, the written word for "sun" may be misinterpreted as meaning elephant by speakers of other Slavic languages. That's the only Polish trick I know.
English + German did not earn me a very high score with Dutch.
I would have expected a higher result for Czech and Polish given my background in two Slavic languages. I would say that the test developers were generous with some easily recognizable international cognates. On the other hand I didn't see many easy interslavic cognates. Generally speaking, Romance and Latin cognates are easier to spot than Slavic cognates.
In any case, currently I find written and spoken Polish opaque. I used to think written and spoken Russian and spoken Portuguese were opaque when I started listening to these languages. Spoken and written Spanish felt transparent* from day 1.
*European Spanish was less transparent.
6/3/18
Polish
Total root Polonization time: 10+ hours? Polish is less opaque. Edit: No, I did not learn 10,000 words in 10 hours. The result reflects the fact that I can now navigate Polish a lot better than before.
I wrote that comment because Ani was like "Whaaat?!" That score reflected the ability to sort of follow some audio.
After around 500 hours of playing with Russian I got this:
Russian
My conclusion is that you are better off without SRS/Anki. Period. If you want deep roots that come alive in contact with input you'll need to do a lot more than flip flashcards.Or a lot less depending on how you look at it.
You do not have the required permissions to view the files attached to this post.
2 x