How to use Anki without getting burnt out (how many cards a day and stuff)

Open Forum
21

Thanks for the explanation, it’s very interesting to see how you work through your method although, as you said, this is a language forum for learning via reading.

I know that ANKI has already prepared decks from many users with audio as well. Do you think they are good to use as well without having to create them? I think many are based on most frequents words anyway.

It’s interesting you have touched the mini stories because Steve says to go through them several times at the beginning. Something that I didn’t do because I’m still learning on how to best use LingQ.

I’ve had a look now on Spanish, that you mentioned before. LingQ’s Spanish mini stories have 1850 unique words.

This is not bad right?

Although you are for single words, as you said before, and mini stories are sentences, do you think that they could emulate an audio only training at the beginning? Or even an ANKI deck for the most frequent words?

For example, if I’m not wrong, like Steve suggests to repeat them multiple times?

It would be nice to understand that comparison between drilling those words in a deck and repeating them with the mini stories.

With your experience, what would you think about it?

Thanks.

22

@Davide:
“I know that ANKI has already prepared decks from many users with audio as well. Do you think they are good to use as well without having to create them? I think many are based on most frequents words anyway.”
Maybe? If you can find one that has exactly what you want and no added stuff then yes for sure, because it’s a ton of work to create your own audio for the deck even starting from a pre-prepared list of words that has exactly what you want.

“It’s interesting you have touched the mini stories because Steve says to go through them several times at the beginning. Something that I didn’t do because I’m still learning on how to best use LingQ.”
I think that the mini-stories are one of the absolute core components and one of the things that Steve focuses on hardcore. I believe he is correct.

“I’ve had a look now on Spanish, that you mentioned before. LingQ’s Spanish mini stories have 1850 unique words.
This is not bad right?”
It’s definitely good from a vocab perspective but the mini-stories IMO are not for vocabulary. The mini-stories are to get a sense of the flow, tonality, structure and grammar of the language. As Steve says “to get a feel for and a sense of the language”. I get an even further “feel” for the language by watching content in all native in addition to the mini-stories.
IMO the core awesomeness of the mini-stories is that it’s gradually building on-itself easy to understand (comprehensible input) spoken natural language.

“Although you are for single words, as you said before, and mini stories are sentences, do you think that they could emulate an audio only training at the beginning? Or even an ANKI deck for the most frequent words?”
OK so here’s my armchair philosophy: in combination with drilling single words I’m getting exposure to the real language as well through the mini-stories plus youtube and netflix.
What I’m trying to do here is mimic what a baby does. The baby is accumulating single words at the beginning at the same time as being exposed to a simplified form of the language AND the real language all at the same time. The baby at first can only understand and pick out single words. Gradually the baby can pick out simple phrases and then more complicated and so on.
I’m extending the baby period by drilling the single words the entire length up to intermediate but I’m also simulating the toddler stage by listening to the mini-stories and TPRS youtube. When I’m passed the toddler stage I stop the mini-stories and watch classroom speech podcasters (e.g. innerfrench and russian with max) at the same time as watching TV with subtitles on. In my mind what I am doing at this stage is simulating an elementary school kid in grade 1 who is going to school at the same time as watching cartoons and kids shows etc.

“For example, if I’m not wrong, like Steve suggests to repeat them multiple times?”
Yes he says to listen to them over and over. I listen to the mini-stories at least ten times each.
What I get out of the mini-stories is a set understanding of the situation in the mini-stories with the phrases and vocabulary to go along with it. The mini-stories allow you to build context and link your mind-pictures of the scene to the actual narrative.
I do believe, however, that youtube TPRS is superior for this because you can see with your own eyes, but the mini-stories still work. Mini-stories in video would be even better IMO. Something like extr@ on youtube (which I watched in French).

“It would be nice to understand that comparison between drilling those words in a deck and repeating them with the mini stories.
With your experience, what would you think about it?”
They are two complimentary things. IMO drilling the words burns a representation of the sound of the word into your brain whether you remember the meaning or not. Over time you will have either a combination of the sound and the meaning or just the sound so you either hit the requirement early by drilling or when you hear the word spoken in a scene on youtube/netflix it will be immediately burned in. Doing drilling individual words at the same time gets you a double whammy.

Reading plus listening to the corresponding mp3s is supposed to do the same thing. From what I can see, however, most people just read and don’t listen to the mp3s. I believe that is an error that will cost them in listening comprehension down the road. I also believe it’s multiples of times easier to recognize a spoken word you know the sound of in a listened to text than it is to try to match a visual representation of a word to an audio representation that you hear in a show.

Peter is trying to address this by his listen while reading approach and I think it has merit, though not enough for me to abandon my method. I actually do a (better IMO) variation which is only pick material in lingQ that has a youtube video to go along with it.

23

IMO where burnout comes from is accumulating way too many reviews as a result of adding too many new cards. The approach of adding many many new cards will lead to the reviews not covering enough of the new cards over time. Which means it will take much longer to burn them into your brain because they’re not being repeated often enough.

According to the research, you have to see/hear a word at least 10-20 times before it “sticks”. Reading proponents try to get this by reading up to 3 million words a year. I’m trying to get this by doing enough reviews that it sticks.

So new vs review is the biggest issue in anki. too many new words the number of reviews becomes unmanageable. Cutting back the number of reviews means you see the word less over the weeks and months so you don’t memorize it fast enough. It’s a balance.
What I have settled on is this: I cram as many new words in as I can in the first month (maybe 1000. 2000 if I can manage). Then I just do reviews for a month. Then I just add new words in at the weekend. Then I do reviews only during the week.

I hope this makes sense.

24

Anki is incredibly boring and the feedback definitely takes weeks or even a couple months before you notice your vocabulary and comprehension has taken a jump. So it’s very very hard mentally to do it.

I also don’t spend my life doing it. At some point the words are successfully “learned” and you no longer need to. For reference: you need 7000 words of vocabulary to have a good grasp of any given TV shows. So in my case if e.g. 5500 of those 7000 words are already “learned” i.e. the next repetition is six months out, I’d just drop doing anki altogether and focus on watching TV and youtube and maybe start doing something like shadowing to practice my speaking.

While I agree with Eric that you could do it by reading and listening mine is faster to get to comprehension of native comprehension. But there is no way to tell. The double blind would be to take my twin which I don’t have and one of us does it one way and the other does it my way.

In the end, if you’re serious about learning a language and not playing at it, you’re in it for the long haul. It’s not a class it’s a lifelong skill you will be honing for the rest of your life. (Which is probably why polyglots are all over the map. Some use methods similar to mine and some others focus on reading only etc etc). So over 2-3 years most methods will work if you stick at them.

25

Thanks for the answers, very interesting.

I agree about the ANKI decks.

I also agree that video + audio + text is superior but you need, I guess, double the time. Because maybe you should watch first, then redo the audio + text. Well I’m not sure what would be the best because it depends on the level you are at the moment of watching.

I tried Fluentu (or something like that) in the past but didn’t convince me. Probably because the platform wasn’t ok yet.

I definitely don’t try to emulate the same philosophy as I clearly have another but I can definitely see that there is a lot of merit in your work. Thanks for sharing.

So, here we have a set of things:
ANKI most frequent words
LingQ mini-stories
Youtube TPRS
Youtube/Netflix + text

We could say that at A1/A2 stage a general LingQ learner could add ANKI most frequent words + Youtube TPRS and boost faster his understanding? (or it’s too much for his mind?)
I’ve never used those Youtube videos, I’ve watched one now and it could be useful.

Mini-stories would do a slow RWL, at this point ASSIMIL would do the same?

26

I think with lingQ you have to do the mini-stories. Also I think Peter is correct in that if you are going to use lingQ as your primary then you should listen as well. I’m ambivalent towards listening at the same time or listening afterwards and I could split hairs on it but we’ve already done the comparison of approaches.

Yeah assimil could work also, so could pimsleur and dave thomas. And you’re right. It’s double the time. But I don’t think there is a scenario where you can spend half the time and get the same results. Listening and reading are two different skills. There is overlap sure, especially when both skills are at a high level but at the early stages I think they are not coalesced yet.

27

“I don’t use cloze because I disagree with the idea of using phrases.”
Hm, interesting.

There are linguists who think that words that aren’t embedded in a cotext (i.e. a sentence) or a (longer) context don’t have any meaning because it’s only the co- / context that can determine the meaning of a word. In short, a word without any context at all would be meaningless (even a reference to an “object” external to language doesn’t change that because reference isn’t a necessary condition for the functioning of language).

Apart from that, the idea that “single words” are “self-contained semantic units” that can just be transfered from one (closed) context to another (closed) context is a “myth” that has been destroyed (or I’d should rather write: “deconstructed”) by French philosopher, Jacques Derrida, and his concept of “itérabilité”.

"Undecidability formalizes the idea that there is no single way to read or understand a text. It is iterable. In this regard, Derrida believes that “without our ability to read signs outside their contexts, the question of context would not arise.” Any sign, linguistic or non-linguistic, spoken or written “can be quoted, put in quotation marks, thereby it can break with any given context, generate infinitely new contexts, in an absolutely unsaturable way.” Consequently, “this does not suppose that the mark is valid out of context, but on the contrary that there are only contexts without any absolute anchoring.”
Jacques Derrida et l’iterabilite du texte - Herve Toussaint Ondoua - Proceedings of the XXIII World Congress of Philosophy (Philosophy Documentation Center) (Translated with DeepL Translate: The world's most accurate translator free version)

In short:
No word can have an a-contextual = completely stable (“eternal”) meaning because contexts can’t be closed once and for all (-> in this sense, there are “no absolute contexts” and therefore “no absolute interpretations”!) and languages never stop evolving. And that’s why the old structuralist idea of language as a “stable and closed structure” or as a kind of invariant relational “crystal lattice” collapsed!

Or vice versa:
If words could have a stable meaning (once and for all), then contexts could be closed once and for all. Thus, language would stop evolving and be an invariant relational “crystal lattice”). However, in this case, there wouldn’t be any language at all!

In sum:
The idea of words as “self-contained (a-contextual) semantic building blocks” that can be combined to form sentences is absurd and not even still suitable as “armchair philosophy” :slight_smile:

BTW, a former acquaintance of mine wrote his PhD thesis on this subject:
https://digi20.digitale-sammlungen.de/de/fs1/object/display/bsb00065020_00001.html

However, that’s probably complete overkill for regular SLA learners. In addition, it’s also written in German, which doesn’t make this advanced topic easier fon non-native speakers…

And last, but not least, here’s a little practical exercise reg. the role of context (from: https://cleanlanguage.co.uk/articles/articles/205/1/Context-Matters/Page1.html):

For each of the following questions, identify both:

  • What context is the client invited to attend to?
  • How is the that context specified/presupposed?
    Client: “My heart is broken inside.” Facilitator:
  1. And when your heart is broken inside, what would you like to have happen?
  2. And when heart is broken, what would you like to have happen?
  3. And when your heart is broken, what would heart like to have happen?
  4. And when your heart is broken, what would you like to have happen now?
  5. And when heart is broken inside, what kind of inside?
  6. And when heart is broken inside, whereabouts inside?
  7. And where is that heart?
  8. And when heart is broken inside, what happens next?
  9. And then what happens?
  10. And where could that heart come from?
  11. And what happens just before heart is broken?
  12. And what kind of heart was that heart before it was broken?
  13. And where did the broken of that heart come from?
  14. And how old could that heart be?
28

“I personally don’t agree that in the early stages reading is optimal” (xxdb)
I agree. "Reading alone (i.e. without listening!) " is definitely not optimal at the beginning stages (see my post in the concurrent thread from today)!

However, I’d say “reading-while-listening” plus some additional listening in our “down time” is. There’s also some research on this “dual input” mode. See these articles, esp. the “references” part, for example:

But the more interesting question is (at least for me):
How “good” is your “single input” mode, i.e. focusing “on listening to meaningful content without reading” at the beginning stages of SLA?

However, I’d ditch your “armchair philosophical” idea of words as “self-contained semantic building blocks” in this context because I don’t think anyone can save this idea, which has absurd consequences.

29

+1

30

reading and listening is good for beginners even like stock beginners with super foreign languages like japanese or chinese? from 0 i can just start listening and reading or does foundational work need to be done?