One of the resolutions I made during 2018 was to make language learning more efficient and streamlined. I was racking my brain on what repetitive tasks I do that I can eliminate. My pondering led me to autohotkey, gaming mouse with many buttons (I’ve been looking into other time-saving scripts since I posted that HG format post), sharex (I use the sound recording feature and capture region feature. I will probably not open paint for a long time! ), anki plugins, clipboard managers like ditto, and strategically thinking about how I spend my time with anki.
Sometime this year I decided to stop editing and fixing up anki cards. It dawned on me that there’s no way to make it efficient (a lot them are basic cards that I want changed to cloze cards and of course I only had 2 fields back then and pasted crap whereever and selected any note type), it sucks up a lot of time, and I have too many cards I want to fix or edit or generate cards from. I do my reviews on an ereader so I can only edit cards on the computer. That would mean I’d be editing cards instead of making new cards or reading/listening/watching korean/japanese stuff. I noticed one time I edited and fixed up 10 cards and 20 minutes passed and I had 900 cards MARKED from ankidroid. Fixing cards take longer than making cards because I have to figure out why I marked it. I don’t feel a sense of accomplishment from fixing cards especially compared to making new cards or doing them or doing something fun in my languages. I’ve gotten better at letting stuff go ever since I read Marie Kondo’s book katazuke no mahou 1 & 2 in Japanese ( the joy of tidying up). By the way, that book is really easy to read. Now I tell myself, if it’s that important then I’m sure it’ll show up again in some memorable context.
I’ve also moved ancient cards (stuff from 2012) and leeches into a new deck so that my decks are more lean. I don’t look at the leech deck or the ancient cards deck but it’s there in case I search for something in anki. I did it after I heard mattvsjapan’s youtube video about deleting a card after 1 year. I can’t get myself to delete cards unless they are truly terrible.
Spending excessive time editing cards is the anki trap that everyone should avoid. I am learning Japanese and Korean to enjoy the media and broaden my horizons and be more enlightened, not to make amazing anki decks. Making perfect anki decks is not the end goal unless you’re selling it? (is it even that profitable?) I thought I could put a reasonable amount of time and energy towards improving/editing my anki cards but I due to time-constraints it’s actually impossible. More importantly, editing/fixing up anki cards does not improve my Korean/Japanese like reading or listening does. People fall into this rabbit hole of spending too much time editing cards or making cards. Anki is a tool to help me reach my goals faster and more efficiently. I can truly say without any doubt that if I watched Korean TV from 2012 looking up every word without using anki I would be nowhere near where I am today in terms of passive vocabulary. Space repetition is effective and helpful especially when you cannot or do not want to immerse full time. If you use anki, don’t just think about how much time you spend doing reviews. Realize that the time spent editing and making cards is time that you could’ve spent reading or doing some other productive activity in your target language.
READLANG + EXCEL for generating cloze cards
I took an excel class in High School and never thought I’d use it since I wasn’t planning to go into business school or finance. It turns out excel is useful for manipulating text.
I figured out how to automate the generation of cloze deletion cards using excel and readlang.com. Of course I’ve been a long time fan of control +h (find and replace text) so I’m not completely clueless about efficiency.
For Korean I use readlang.com’s chrome addon to collect the words I come across in internet articles and although the site generates cloze deletions when you export the text, I would never cloze delete a whole word for Korean. It is way too difficult. I like to make 2 clozes for a syllable of the korean word and another cloze for the meaning or I make cloze for the meaning as the same cloze number as the syllable that is easier to remember. My choice is either to make clozes manually in anki or figure out excel and automate that stuff. I obsessively thought about it one day and then figured out how to do it in excel I ended up using a combination of concatenate, replace, and substitute. In addition to clozing the text I also concatenated html and asterisk so that the word stands out. I put in html to make the font size big and underline the text. It feels satisfying to drag my mouse to cloze and format the words exactly the way I want. Also I use naver translate and so-net so generate korean-japanese and korean-english translations so it saves me manual look-ups when I go through the sentences in excel before importing to anki. Readlang.com’s chrome addon uses google translate by default so it is a hit or miss regardless of what language you’re translating. It doesn’t even work well for Spanish.
For Japanese I use rikaisama to generate the import.txt and I paste that into excel to manipulate the text into my liking before my importing. I set up excel scripts so that the word is wrapped in asterisks and underlined, and the first or second syllable of the word in hirgana in the the sentence is clozed. Just 2 years ago I used to manually cloze stuff out using ___ before importing because I completely forgot about excel.
For Spanish I’ve using workaudiobook (using audio and srt sub files from youtube), lingoes pop-up dictionary, and readlang.com to learn it sporadically. I run the sentences through deepl and reverso contexto before I import into anki. I paste everything into excel to set up everything before importing. I use substitution or replace (can’t remember) to make the word be bold, underlined, wrapped by asterisks, and be a huge font size. I am nonchalant when it comes to Spanish so if the deepl or reverso contexto both fail me I suspend the card and move on. Most of the time they pull through since they both kick google translate’s ass. I’ve been thinking that I do not want to learn Spanish and French at the same time since it’s inefficient and I don’t have time or desire to commit to such an undertaking. I have no aspirations of becoming fluent in Spanish or French because I know how long it takes and I have a lot of stuff to read and watch in English, Korean, and Japanese. I hope to implement this method for French sometime in 2019 when I feel like I made decent progress in understanding conversational Spanish. I was also thinking that learning Spanish and French is a good strategy since they complement each other. Instead of looking at it from the point of view that Spanish will help me learn French, I think of it as it’ll make forgetting French harder. I think the former is too optimistic and unrealistic. That is the same approach I took with Korean. I said to myself there’s no way I will forget Korean and go backwards if I am well-versed in Kanji. There’s just no damn way. I proved myself correct 🙂 In fact, mass-reading Korean novels etc without knowing about Kanji seems mad inefficient to me. Yes you benefit from reading a lot but you benefit so much more when you know Kanji/Japanese. This year I read this book and I looked up 0 words when I read it (I underlined anywhere from 0 to 10 words per page. most of the pages had 3-5 unknown words. Of course there was a handful sentences where I didn’t understand the sentence despite re-reading it.) but I definitely improved by the end because I know Kanji and Japanese. It was just a matter of me figuring out the word/hanja that is hiding behind hangeul. Of course there were non-hanja based words that I didn’t know that I couldn’t do anything about except try to figure them out from context. I plan on flip-flopping between the 2 languages every x months because me doing a 20minute session in Spanish one day and then a 20 minute session in French the next day or the same day is pointless. The inefficiency bothers me.