Top
Best
New

Posted by lakesare 3 hours ago

With 86% of the words, you might understand only 38% of an anime's lines(learnfromx.com)
48 points | 33 commentspage 2
dumberquestions 3 hours ago|
That's higher than what I would've expected.
watwut 3 hours ago||
> Coverage is the number most learning sites quote.

This person is visiting different learning sites then I do.

Also, I dont need to understand every word in sentence to understand it. Especially when coupled with visual show.

chomp 3 hours ago|
It’s an LLM, so…
watwut 3 hours ago||
It did indeed sounded like one. Noticeable by a lot of words that say nothing that would matter.
gustavus 3 hours ago||
My first anime was NGE. I knew all the words and still understand nothing.
archargelod 21 minutes ago||
I've only watched the original series, once, and understood most of it.

It's just meaning is very very vague on purpose. It's not meant to be specific. Most of Japanese art is like that.

Also, if you start learning Japanese language, you'll quickly notice that the "I'm not going to tell you anything directly" and "read the room" permeates even casual and especially formal daily conversations.

Zealotux 2 hours ago||
Really? Eva is pretty straightforward to understand once you watched End of Evangelion, Misato pretty much explains most of it in the movie.
2TOUCH 2 hours ago||
[flagged]
linguae 2 hours ago|
This reminds me of how I learned about Zipf's law. 19 years ago, I spent the summer after my sophomore year in a research lab at UC Santa Cruz as part of an NSF-funded summer research program for undergrads. My project that summer was to investigate index design for the Ceph distributed file system in order to enable faster file search. This was a hot topic at the time, especially with Spotlight being introduced to Mac OS X Tiger, and we wanted to see how to enable similar functionality for a distributed file system. I learned about Zipf's law, which states that there is a power law distribution of word frequencies in many natural languages. I remember downloading the Project Gutenberg corpus and writing a Python script where I stemmed the words and computed the frequencies of the stemmed words. The resulting distribution confirmed Zipf's law.