“grit” as an indicator of success


Here’s a link from khatsumoto’s twitter feed: The truth about grit.

There are a lot of interesting ideas in this article, but i think it mainly reinforces the fact that your brain needs time to make solid connections. You need to keep working and working at something in order to get good at it, and therefore you need to break through any setbacks you may run into that can sap your motivation. Your mindset should be focused on jumping those hurdles and continuing towards the goal. To do this, i think it helps to have examples of people who followed similar methods and succeeded just by doing it lots and lots.

I’m trying to find any such examples that i can, because it helps me continue if i know that it’s going to work eventually. Being a very logical person, i find it easy to work with textbooks and definitions, but language doesn’t play well with textbooks and definitions. It’s more of an absorption process, and that causes me some problems. I need to retrain my thought processes to trust in the long-term, to have more grit.

One thing that i’ve been working on today is to give up some short-term satisfaction, or at least what appears to be satisfying at first. For example, i was watching german star trek today, and it was a really interesting episode and i wasn’t quite getting all of the plot details that i wanted to. Instead of flipping to english subtitles, i just kept hitting rewind and listening to the same few german sentences over and over again. Instead of looking up the words in a german-english dictionary, i tried to figure out what the possible meanings might be in the context.

The plot is not the focus of my training, the language is. There’s no reason for me to need to know the plot, except when it helps my language learning. Really, i should be thinking of the plot as a puzzle to be solved, not something a prerequisite for understanding the episode. Learning the plot through english beforehand is just an illusion that would kill my curiousity and hence my reason to learn more from the episode. it is the result of me focusing on an end, rather than on the process.

Figuring out the plot in german and resisting the english subs took more effort, but i think it was more fun, and i eventually figured out several words that way. The easy way out was to fall back to translation, but taking the hard route paid off (in the short term too, surprisingly). It would be much more gritty of me if i could continue to do this for months. As Keith would appreciate, this intensive german-only investigation of the phrases and words gives me more context, more experiences to tie them to. The investigation itself is an experience, as is the video.

Grit seems to be composed of self-confidence, and the ability to convince yourself of the inevitability of long-term success given some short-term process. Since we can’t directly visualize what the changes in our brain will be after hundreds of repetitions, we have to create a sort of self-faith that such changes will occur. Or perhaps the buddhists would describe it as letting go of the endpoint completely and just become one with the task right now…be present. Khatsumoto might say “just keep starting”. There is no act called “finishing” that one can choose to do, one can only choose to start. Grit is when you convince yourself to keep making that choice.

what do you put in your SRS?


I just responded to the latest post over at Language Geek, where Josh asks “what do you put in your SRS?”. After my short comment turned into multiple paragraphs, i figured i’d just post it here too:

Firstly, i only put phrases or full sentences into anki, except if they’re nouns that really don’t require a sentence. a lot of nouns don’t really have much “usage” information, but it still makes me nervous to just put a single word in.

I generally get my sentences from books that i’m reading. Currently i’m reading harry potter #2 in german (”Harry Potter und die Kammer des Schreckens”), and although i understand a lot of it, i constantly come across words that i don’t know. I don’t put every single unknown word into Anki, however. I usually wait until i’ve seen a word more than once, or if there’s a certain paragraph where there were MANY words that i didn’t understand, then i concentrate more on that particular paragraph.

I do this because i put a lot of emphasis on reading without stopping. If there’s a word that continually bothers me, then i’ll use a highlighter to mark it for later, and then go back to it later when i’m working on new anki cards. Sometimes when i go back to it later, it makes perfect sense so i don’t bother.

Combined with this, i sometimes get some cards in anki that just annoy me. Maybe they seem useless now, or maybe i always get them wrong and they just bug me. these get deleted mercilessly. I know i’ll have no problems finding more words to put in anki, so i don’t worry about losing a couple of the stupid ones.

Another thing i’ve been trying lately is using electronic copies of books in order to do some statistical sentence-mining. I use an emacs add-on that a friend wrote, which makes a list of all words that i have in any cards in anki, and then uses that “known” list to find words in my book that are “unknown” and are also of a high frequency. It then gives me example sentences for those words. This way, i can work through the important vocab in frequency order, which helps tremendously.

In general, i try not to focus too much on words i “need” to know. Ya, eventually i want to know EVERY word, but i don’t need to do them all right now. There are plenty of things to learn, and i think i’ll be more effective if i do them in a better order (and usually i think this means frequency order). This is why i try do wait until i’ve seen the word more than once in my paperbacks, or i actually do frequency analysis if i have the electronic form.

Also, remember that you don’t need to worry about how much you know in total so far, you just need to keep increasing. As long as your knowledge keeps going up, you’ll get there. Don’t worry about grabbing everything right now this instant.

august week 3 summary


I’ve been a bit busy this week, so i’ve done much more listening to audio on my mp3 player, and much less watching TV and anki. I’ve also noticed that i’m starting to judge myself by my ability again, instead of by my amount of work. This is probably my biggest mistake lately.

To get back on track, I’m going to try to remind myself that it doesn’t matter what i can and can’t understand at this point in time, it only matters that my numbers are going up. I know that my method will prove successful in the end, because others have done it before me. All i need to do is keep the numbers going up, week by week.

On that note, i also notice that i’m impatient to understand everything in an episode when i watch TV. This next week, i’m going to work more on cultivating the attitude that i should enjoy whatever i can in the episode. There’s no need to worry about fully understanding every minute plot detail, because i can just go back and watch it again later once i’m awesomer. For right now, i want to just watch and enjoy it for what it is.

Overall, i like my goal for tv and for listening. I’m not entirely convinced that my anki goal is right yet, though. 150 new cards every week seems to keep me quite busy, both in adding cards and in doing reps for them. I’m thinking of using a highlighter in my book so i can find sentences for later…that should save me some time. Then i can spend more time freely reading and just flip back later to grab some things. Ideally, i’d like to spend just one or two days per week on adding new cards, and the rest of the time just relaxing with tv and doing reps.

keeping stats


Lately I’ve been reading about some ways that other people keep track of their progress. I just thought i’d clarify what i’m doing, since some people appear to be interested.

First i set some goals, which were based on some numbers i picked for the rest of the year. Then i brought them down to the level of month/week/day, just to see if the number of hours was feasible. Once i had adjusted the numbers, i wrote down what i needed to do each week in a file called “goals”.

I then started a new file called “logs” where i just wrote down the date and what i did that day, in a short format. I quickly found that when i was updating my blog, it’s much easier to get the numbers right if i keep a weekly subtotal, and then i can just add the numbers for the current week onto last week’s subtotal. If you don’t keep track of that running subtotal then it gets hard to figure out if you’ve updated the total properly yet, especially if you end up doing more work after you update.

Another thing i’ve been playing with is making graphs with gnuplot. To do this, i use a file with various info separated by spaces, with different days on different lines. then i have a gnuplot script that will graph those datapoints. I haven’t quite streamlined all of this, so it’s a bit cumbersome right now. What i’d like to do is set up something where i update in one place, and then the data gets automatically placed into the gnuplot data file and graphs get generated. But right now i just do it manually every day or two, which means i’m updating several different text files. It only takes a few minutes though, and it makes me happy 🙂

The next thing i’d like to do is to find some sort of wordpress plugin that will create little progress bars, or maybe a thermometer picture where the “temperature” goes up as i progress. So far i have no idea how to do that, and i don’t want to spend the time to figure out how. I suppose i could write something using Imagemagick, but i’d rather be working on german instead of coding. Hopefully someone else has already written it somewhere else, and i just have to go find it some day.

Another method that i recently heard about was using a spreadsheet. John over at Global Maverick likes to create a spreadsheet with milestones in various boxes, and then change the color of each box as he progresses through those milestones. I quite like this idea, since the coloring thing sounds fun, and you end up with a list of accomplishments that is visually indicated.

What i’ll probably do with this stuff is make a little shell script or something, which can just gather the stuff i do. Then i can tell it “i watched 2 tv shows”, “i listened for 3 hours”, whatever, and it’ll keep track appropriately and spit out some stats and pretty graphs. I’ll let you know if i make any progress on this, but it’s still a low priority behind actually studying german itself 🙂

august week 1 summary


It’s been one week since instating my goals, and i’m feeling pretty good.  Although there are still plenty of things that i don’t know, it’s been pretty helpful keeping stats so that i can judge my progress by numbers going up instead of just wishing i was awesome already.  It’s enough to know that things are going up, and eventually i’ll get to where i want to be as long as they keep going up.

Somehow i managed to get in 21 hours of german TV this week. My goal was 17, which would give me 70hrs total for the month. I watched a lot of star trek, along with a little bit of House, Simpsons, and MASH. MASH and the Simpsons are reasonably easy to follow, and star trek is great now, but House is pretty difficult. I think most of the jokes in House are based on sarcasm, and are sometimes subtle, so it’s a bit hard to get, but i’ll keep trying.

Reading is going well, but i didn’t make as much time for it as i wanted. I only did about 2/3 of my goal amount, but i think i can catch up this week. Harry Potter is starting to seem really familiar, and there are lots of words that i know. There are still some parts where i wonder wtf is going on, but i usually get it if i start the chapter over again once or twice.

Adding sentence cards to anki has been tremendously helpful. A lot of the words from those sentences come up all the time in my tv shows and books, which gives me yet more context and exposure. I’ve mostly been adding things to anki based on common verbs or adjectives. Sometimes it’s stuff that i found in a tv show or book already, and other times i just look at a frequency list of german words and pick out some “common” stuff that i don’t know yet, and go find sentences for them.

Here’s a good spot for finding german example sentences: http://www.mydict.com/Wort/X/ where “X” is the word you want. I found this randomly in a google search, but i hadn’t seen it before. It seems to use the contents of all sorts of magazines and newspapers to pull up examples. Although there’s a “search” area at the top of the page, it seems to link elsewhere, so i just edit the URL to put in new words. Sorta a hastle, but the real example sentences are worth it. Some of them are quite complex, so i sorta browse through until i find one that’s mostly understandable, and then i add it to anki for review.

My extra listening exposure has been a bit of a problem. I thought i’d have plenty of situations where i couldn’t watch TV, but i could have my headphones in and be listening to stuff, but it actually hasn’t come up as much as i’d thought. I have a Harry Potter audiobook playing right now as i write this, but otherwise it seems like i’m busy doing something that doesn’t quite work with headphones. I’ve lowered my goal for that, but maybe i’ll get a lot more headphone time at work this week. I think i partially made up for it this week by exceeding the TV goal, anyway. At least all the numbers are still going up.

Overall, i really like this numbers-based approach. I like getting high scores, and it gives me something to work towards. It also sorta reminds me of mudding, ie playing online roleplaying games. somehow getting my numbers to increase is pretty satisfying, but maybe that’s just my weirdness.

making goals


I’m pretty new to this whole “making goals” and “planning” thing. I tend to just jump into something, do as much as i can, and figure things out as i go. After reading a lot of different language learning websites, though, it seems that the best language learners also seem to make goals and try to exceed them, so i’m going to give it a shot.

One idea that i’ve seen mentioned in several places is that when you make goals, you have to actually decide what you want first. My experience with chinese has taught me that you can’t really make goals out of specific language milestones, because you don’t know how long it’ll take to get to those.

So i’m going to take Khatsumoto’s advice about short term goals. I’m just going to pick some raw numbers for this month for all of the language tasks i want to do: a number for words read, a number for amount of TV watched, a number for listening done, and a number of Anki cards entered. I want to read 150000 words, watch 70 hours of TV, do 90 hours of other listening, and enter 600 new cards into Anki..

To accomplish these, i need to break them down to a weekly and daily level. Per week, that’s 37500 words read, 17 hours of TV, 22 hours of listening, and 150 new Anki cards. So far this month, i’m at 7 hours of TV and 6 hours of other listening, with 92 new Anki cards added, and 6000 words read. Already i can see that i need to get more reading done to catch up, so these weekly goals are already pointing me in the right direction.

Now that i have something to aim for each week, i can also try to get that done early. I have some sort of expectations to exceed. Hopefully this will help me stay away from the temptations of surfing the web.

One more thing to remind myself is that consistency is key. Doing at least some work each day is better than doing a whole bunch all at once. If you can do at least a little bit of language work each day then it’ll add up to quite a lot over time. With that said, i’m going to go make some new sentence cards for some verbs i just saw 🙂

more on SRS and sentence mining


Lately i ran across Glowing Face Man’s blog again, and reread some of his articles. He has some informative descriptions of Sentence mining and Spaced Repetition, as well as many other interesting articles.
I was particularly inspired by what he called his “French Revolution“, where he spent 1 month trying to see how much french he could cram into his brain.

If i had to advise someone on how to do a 1-month challenge like that, i’d certainly suggest an SRS as one of the tools, but not exclusively. Judging by his description, GFM felt by the end that he didn’t spend enough time on pronunciation. I don’t think pronunciation is best learned through SRS; rather, i think you should just watch tons and tons of TV, or listen to the radio, or any other regular audio source.

I don’t think you’ll get enough content by adding a few sound clips into your SRS, because you need massive exposure to the sounds. It doesn’t particularly matter which sounds, either…you just need lots and lots of them to hit your ears. To get that, you need to have something with extensive content, and therefore it has to be interesting and entertaining to keep you watching or listening.

What SRS *is* good for, however, is strategically filling your brain with important vocab. The words of any language are not all equal in usage. Some are used a lot more often than others. Those frequently-used words are of high value, especially at the start of your studies where all words might seem quite confusing. When you get some of those frequent words figured out, it’s like a blast of fresh air and everything seems so much easier. This will give you the encouragement you need to continue on with your studies.

After the really frequent words, i think an SRS is also good for some of the seldom-used words. You may not hear them that often in a TV show, but they can still be important in certain situations. This is where you can use your SRS to give you more exposure to those words than you might have had just from TV or radio alone.

For most of these words, you should be entering them in a full sentence context. Single-word cards are probably only good for things like household objects or other lists of nouns. For anything else, especially verbs, you definitely want to have phrases or full sentences.

Words tend to have many different uses in different situations, and you lose all the subtlety when you make a flashcard with only a single word on it. Therefore, I suggest that you always try to find example sentences for the words you want to learn. For many words, you might want to add several sentences that show the different possible meanings.

The way that I do this lately, is i watch several hours of TV every day, and i also spend a little bit of time gathering some sentences. I might go through the subtitles of one of the TV episodes to find words i don’t know, or sometimes i make some notes while i’m watching and come back to it later. I also go through whatever book i’m reading and find interesting words in there and grab a sentence for those.

Sometimes i’ll open my book to a random page and just look through all the words on that page to see if anything catches my eye. Then i type the sentence into Anki to save for later. This way, I’m gradually gathering vocabulary in context, and i’m using Anki to make sure i never forget it. This gives me focused exposure to these words that i previously didn’t know that well. As time goes on, the cards become more spaced out and i don’t spend as much time on them because they’re more well-known, which gives me time to work on any new unknown words.

I think that both massive exposure and SRS content are important to balance out somehow. SRS is a type of exposure, but not quite as good as hours of intensive audio and video content. But your SRS content is a focused tool that will help you quickly increase your vocabulary, which will let you get even more benefit from the hours of intensive audio and video. Therefore, i think these two strategies go well together, as advocated by Khatsumoto on his All Japanese All The Time site. For example, check out his article about how he learned to watch the news in japanese.

So, in summary, watch a LOT of TV, and continually gather sentences in your SRS to quickly learn vocab. I think you’ll be surprised at your progress 🙂

Bootstrapping beginners, and using “Michel Thomas German”


For beginners in a new language, there’s a common problem of getting “bootstrapped” so that they have enough basic understanding of the language in order to pursue more interesting materials. Most of the intro textbooks and methods are pretty boring to me. I’d much rather be watching interesting TV shows and movies, and reading awesome books, but it’s frustrating when you start totally from scratch and try to watch an incomprehensible movie, or read a totally impenetrable book.

For german, I didn’t have this problem, since I had previous experience; when I was about 5, I spent a lot of time listening to my german grandparents, and then later when i was in highschool I took some german classes. However, my girlfriend is presently learning german with me, and she hasn’t had the benefit of previous experience such as this.

At the start, we tried using videos to give her an introduction. We would watch an episode in english first, so we knew what was happening, and then we’d watch it again in german. I found this a bit boring, since i didn’t always want to spend 2 hours watching a 1 hour episode, but it did help her quite a bit with getting the sound of german into her head right away. After watching most of Star Trek: Deep Space 9, Season 1 in german, we decided that something more was required, in order to get her some knowledge of more of the basic words.

This is where we decided to try Michel Thomas. Pimsleur wasn’t that great, because it seems to focus more on travel vocab rather than general use of the language. Michel Thomas gives a pretty good overview of how a language works and how to put together interesting and practical sentences. His method assumes that you’ll fill in a lot of the vocab gaps later, but at least you have a good base to build on.

We got a lot out of listening to the first 3 CDs of his “Beginner German”, but we were really annoyed at the “students” in the recording. They pronounce everything horribly, whereas my girlfriend had excellent pronunciation already from watching all that star trek. She could also recognize right away all the places where the students on the recording made mistakes. The other thing that annoyed us is that there was more english and badly spoken german than there was actual german, and the progress became quite slow after CD 3, so we skipped the next 5 disks completely. Yes, we skipped CD 4 all the way to CD 8. the important bits were all on CDs 1-3.

Next we came to the really valuable content: the “Language Builder”. This series of recordings does away with the idiot students and their horrible mispronunciations, and just has Michel Thomas going through tons of useful words via numerous example sentences. For each word, he gives many possible sentences that have slightly different meanings, and also reviews previous content as he goes along. All the words are super-useful general words that you use all the time. So many great verbs, in great example sentences. This is really bootstrapping gold. Having all of this content under your belt would really catapult your understanding of TV shows and books.

As someone who previously took high school german for 4 years, i was surprised to find a lot of really awesome words that i hadn’t previously learned. I think the high school courses had a certain idea of which concepts were too “hard”, and so avoided a lot of extremely useful things, whereas Michel Thomas basically doesn’t care which things you might think are “hard” grammar-wise. He just wants to give you a great grounding in the language.

So, in summary, i think i’d recommend this for beginners in german:

  • get german sounds into your head through extensive listening. You don’t have to understand anything, just listen to hours and hours of german.
  • after maybe 30 or 40 hours of listening (which could be in the background as you do other stuff), listen to the first 2 or 3 hours of michel thomas beginner german.
  • next, skip the rest of the beginner course and go straight to the “Language Builder” and listen to all of that.

Doing this will shoot you way ahead, i think. From there, you can go back to hardcore movie watching and book reading. I think you’ll be able to catch a lot more vocab naturally.

Now, what would you do with a language that doesn’t have these particular recordings? I’d say the biggest thing would be to search out a list of the 100 most frequent verbs, and maybe some adverbs and adjectives. forget grammar, just fill your SRS software with the most frequent vocab, and you should try to do this by finding example sentences for each of those words rather than just the words by themselves. Once you can pick out all the most common words by ear, then the rest of the language should start to fall in place as well.

Good Luck!