Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Question aside. Does anyone know straightforward ways to create anki decks?

Last time I checked on this it was like a lot of effort to put on and got discouraged to build my own decks to improve my italian. I was expecting anki decks to be constructed easily, in its most basic form from plain text files (html, markdown, etc) with references to maybe resources (audio, images, etc) just from a filesystem directory but is not that simple.

I can't recall the name of a Python tool that allowed you to create decks programmatically but I found it way too much effort to use it and I couldn't find other good alternatives (maybe this has changed recently, don't know).

Does anyone recommend a good, simple and straightforward tool to create decks/cards? (I'm using FreeBSD).



There's a Python library called genanki that has a minor learning curve but isn't terrible to figure out. My process is to maintain a gsheet with data I use to generate my flashcards, then manually download the CSV data, and run a Python script to convert everything into the Anki format.

If I'm not mistaken it's also possible to import CSV data directly into anki without the intermediate Python script


Yes, this is the one I previously found (thanks for sharing). Found it ridiculously complex for what an anki card seems to be... I mean from the docs I found this:

> "...You need to pass a model_id so that Anki can keep track of your model."

At that time I wondered myself: "Why on earth do I need to keep track of a model id for an anki card?" then quickly move forward on the docs without paying too much attention and finally put it aside...


Sounds like you got tripped up by digging too much into the specifics of models. While a powerful concept, they're not necessary to understand if you just want to generate basic front-back cards. Simply create a `MODEL_ID` static constant in your file and never touch it again.


Anki can import csv files.

That said, making decks is extremely tedious. It’s better than a premade deck for learning language vocabulary, but it’s time consuming.

You shouldn’t make vocabulary cards programmatically. If you’re going to invest time into learning a card, you need to be absolutely sure that it’s correct. Otherwise you’ve wasted your own time.

When I was using Anki daily, I’d reserve about 20 mins or so in the evening to input new words.


I heard this many times before... If there's an easy way to create the cards I'm willing to put some effort on it. Italian and spanish are not that far away languages so most of the time I really know the meaning of the words (with of course the help of translation and a dictionary sometimes) I learn but I need to remember them.


Well if you have a source of vocab you can do it.

Honestly CSV import is the way to go. That leaves you with how to get cards in. That depends largely on how you are working with stuff.

Some bespoke options include:

- The Kindle dictionary lookup history is an SQLite file on the kindle. Scrape that and pull it out

- Write a CLI command to lookup words in some dictionary website and stick that somewhere

- Just have a text file open somewhere where you paste words, and then you do the lookup later.

I think there's a lot of ways to make things easier, especially if you're working with another language with an open source dictionary. You need to do some one-time elbow grease to get it in, cuz what people want is different.

Another thing you can do if you like just going wild: just download somebody else's card deck. You can set things up to show you (for example) 10 or 5 or even 2 new cards a day. You will end up being shown a bunch of new words. You can just absorb those over time and see how that helps.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: