Hi, I’m trying to use a part of my vault as a learning machine too.
We should probably not forget that french or korean or any other language must have a specific way to learn. The way you’re gonna build your vault depends on your first language, your goals, your favorite strategies to learn language, your current abilities in different languages…
Currently, my vault is using 4 language French,(L1), spanish (L2, some notes), english ((L3), lot of my research are on this language (linguistic imperialism is bad)) chinese (L4, basic level (~HSK3/4), language studying).
As first I was thinking about using a table in a specific template, and using a plugin as smart random note or Flashcard, but I’m not sure it will really fit my needs and I think everything would be lost on all the notes. I’m still thinking to use one of them but not sure how to integrate it in the flow
Currently my process to learn chinese is to add new words using inlinefields. As my research are using a lot of chinese word inside english text, every time I have a new word. I just use inline fields.
I have no specific knowledge on coding or anyhting, I’m just sharing my flow any recommendations or suggestions will be really appreciated
Basically the flow is based on a “not focused on learning” approach. Which means I’m not trying to learn I just try to learn new words when they appears on my readings.
I highlight the new words using hypothes.is or Zotero which both have plugin to export notes to zotero.
If i dont know the word on the quote I go to Pleco or google translate and check the pinyin and the translation and add it under the quote.
this is a quote about some important stuff on China using some 中文 [simplifié:: 中文]
[pinyin:: zhongwen] [fr:: chinois]
I use Refactor plugin with a templater to extract my field into a new note in a specific folder and I let an indented picture of the new note like :
- I use Dataview on my daily note to review all the new words.
pinyin as "🗣", simplifié as "🇨🇳", fr as "🇫🇷", en as "🇬🇧"
AND file.cday = date(this.file.name)
flatten pinyin and simplifié
SORT file.ctime asc
Explanation of my flow
- It’s not just a popover function, as we could do it with [[link]]. So, I still have to do the effort to check the pronunciation, the translation and the way we type the characters.
- It’s not a dictionnary. I have to actually think about the hanzi, the translation and the pronunciation every time I met the new word, until I mastered it. It’s sencence based and not vocabulary based which fit more my needs on the target language (I think…)
- Creating a new note for every sentence give me the ability to add any new language later and make it appeats on dataview later
- It gives me the ability to enter full sentence and by this way my learning is not focused on "vocabulary but in structures (it will probably be more complicated to use this flow with roman languages).
Reflexive view on the flow
- I should probably use a faster way like “L1”, “L3”, to design the language but I think I could mixed them up and destroy the database if one day I consider chinese as L3 or L2…
- There is probably a better way to process and i’m taking any advice about the process or the code (code newbie)
- A morning review of thevocabulary learned on the past fays could be another good way to review
smart random note or Flashcard could still be useful, but I’m not sure I will use them… With dataview, I have my notes every night when I’m writing my end day note
- Not sure the extraction of the note is really important. I see it as a good way to check my workflow. Like “What did I learn yesterday afternoon?” Well I have 15 notes , not only one created at 1pm I could probably use the quick add plugin but it will probably be doublon on the dataview then…
Chinese language is famous for his idioms (chengyu), so I download a list of chengyu I found on the web. (500 chengyu that I separate into different notes). I also play the chengyu version of wordle which I iframed on my daily notes
I recently did it so the flow is still under construction. but it looks to something like that:
- I have a note looking like :
[pinyin:: cuò zōng fù zá]
[fr:: embrouillé / compliqué]
from [[2022-06-01]] [[成语 - Liste]]
On my daily note I’m using spotlight, to have a random note. So, every morning, on my daily routine I check the new idiom and i add the date using natural language dates shortcut, (this is just to make it appears in some dataview that I did not prepared already)
At the end of the day during my review, I’ll add some context like "“Quote from poetry from Qing Dynasty talking about blablablablabla”
View on this flow
- I know I could use other plugin like Dice plugin, but this one fit my needs for now, so I give it a try and I’ll see where I go with this one.
- i’m still looking on the best way to develop this part
General considerations on the topic
Other ways I’ve seen on the web
- It looks like this guy is using some magic to create a real database to learn chinese. His flow looks related to twitter, and markdown, so there is probalby some inspiration to take. Still did’nt dive on it
- If your target is one specific specific language I saw mulfolk as an easier way to learn new words and implement them in his daily note
- There is the anki plugin
My daily routine learning language on Obsidian
- Open daily note :
- Start reading and highlighting papers using hypothes.is or Zotero
- End of the day check my daily note to see the new words. on Dataview
- Repeat them and try to build sentence with them, add some context
Example of french (basic suggestions)
- You’ll have to build a rigorous template for all conjugation
- You’ll need to use a really precised grammatical-function based YAML
Sorry for the long post, the typos and the bad english.
Plugin I use
Other plugins I talked about