I always had a bad time trying to read native content.
When you read, do you try to flow with the book? Don’t look at the dictionary unless totally necessary?
Do you try to think about the grammar in the sentence? or it’s enough if you understand the general direction of it?
I do a bit of both honestly. I read all digital and have a dictionary popup and a plugin connected to my jpdb account that highlights words I don’t know on a page so I can smoothly check the reading if necessary (or mark them as known so it gets added to my “vocabulary”), I use this for when I want to just read and immerse focusing more on quantity than quality and pick up things along the way.
Then there’s specific things I read that are much shorter where I do intensive reading and focus on checking all the grammar etc.
Similar to Sisthra, I do a bit of everything. It depends on what I’m reading and what I want to get out of it. Sometimes I’m just reading something for entertainment and getting a general idea of spots I’m finding tricky is good enough to keep enjoying the content. Sometimes I’m trying to improve my comprehension and I’ll pick a part the sections I’m having trouble with using a dictionary and grammar tools.
If it’s something I want to understand and I’m going to spend some time with I’ll usually do a quick read of a section / article and see what I can understand, what I think the overall meaning is, where the problem areas are, and what I may need to look up / improve to clear up the text. Then it turns into more of a study topic: go through and look up unknown vocab / grammar and piece together sentences that were giving problems, improve my translation, turn those things I had to look up into topics for further study or srs.
I think immersion is one of the best ways to improve languages. If find something you enjoy then turning it into a study source lets you link your knowledge to something and helps with remembering. As you find more content to go through like that it makes future content easier and adds up to making the language easier and creating a loop to fluent understanding.. it just takes a while to get there
On the first read of the content:
First read of a sentence / paragraph go with the flow then instant lookup of everything unknown through jpdb using directly this extention (which is probably the one mentioned above in the others answers). Then mine all the unknown content.
Repeat until done.
On rereads (after the mined content has been absorbed): vibe with the book :).
For me grammar is rarely the reason I’m lost in a sentence (not because I’m good at it but because for now missing vocab and / or kanji cause me more cognitive strain). I think about it actively when all the content is known but the meaning is still unclear.
my fave thing to do (although i get lazy and don’t always) is to write down words or phrases i’m unsure of in my notebook while reading and then look them up all together, and maybe if i’m feeling especially thorough, reread after looking them up.
the extra time and attention to writing out the word and the bit of it that i didn’t know is my version of SRS: i ain’t into anki farming, and i rarely revisit what i write down.
something that’s fun now that i’m almost done with WK is the kinds of lookups I’m doing are kinda differentiating: like, “i didn’t learn the kun’yomi for this kanji but here it is in a verb” or “i’m pretty confident about the meaning but unsure of the reading” or “i know how to read this word but i’m not sure of the meaning” in addition to the “i don’t know anything about this word” that i’ve been familiar with since forever
how do you use jpdb?
So jpdb is a website that collects wordlists from different media (books, novels, etc), so that you can in theory prestudy the vocabulary with their SRS system. But I use it more as a database of words I know that I slowly built up. There is an extension called jpdbreader that you can install on most browsers (even some mobile), that basically acts like other online dictionaries like yomichan but will allow you to also mark words as known by adding them to a jpdb deck. So it’s basically like Anki with adding flashcards to study later, but you can also mark a word as “never forget” and it will get marked as known but never give you reviews for it.
Note that jpdb is essentially abandoned by the developer, though, so people who don’t already have an established setup with it might want to see if https://jiten.moe/ (broadly similar, open source, actively maintained) does what they want first.
I’m still using jpdb because I’ve been using it for years already at this point and it does mostly do what I want from it; but at some point I expect I will migrate to jiten (which has some support for importing from jpdb, but not totally seamless because jpdb doesn’t expose all user data).
I personally prefer to follow with the book and avoid looking into dictionary and also feel grammar more than think about it… But that’s just me – and I’m just a small clumsy cat 
Ohhh looks like I might jump the gun, seems their reader integration is top notch. Thanks a lot for the heads up !
(edit: urghghhh having everything in new is unbearable 
i’m good with jpdb for now but it’s nice to see there’s a backup if the service discontinues
edit 2: I really like the activity of jiten and especially having up to date dictionaries, seems like a migration later would be a good compromise.
)
Thanks for the heads-up! I’m fine with it for now, but I will look into migrating into jiten eventually. I would like to avoid having to mark 10k+ words as known again lol.
Seems migrating isn’t the worst Migrate from JPDB to Jiten - Jiten to keep the known levels but one missing feature that i (unfortunately) still use is the bastardised ai translation that is still very useful at my pitiful ability.
Yeah I checked and thankfully the migration is pretty effortless! It “dropped” some 1k words but those could be some mix of blacklisted and katakana words I guess. Will still continue to use jpdb for now because I have it all set up already
Yup I’m not convinced to migrate for now…
The flow disruption on my decks is too great to jump in lightly…
(it ask to re-review my 3K+ level 6 cards… Thanks but NO 
)
But if i find how to have the same review template that i have on jpdb, vertical examples sentences, etc. it might become a possibility, one sure thing is that it’s packed with features and it seems to evolve, in the long run it will be better than a non active site.
My main issue is that it’s focused on adding lists to work on / find media content while my current flow would rather be to start from nothing and add content as I work on it.
Also not sure if I agree with the “find content adapted to your level” logic, just work on the content you want dummy !
Maybe its better suited to more experienced users that like to branch out.
To answer this question: I do basically full lookups. If there’s a word I don’t know or I’m unsure about (meaning/reading/etc), I look it up. I think of learning new words as part of the fun. Lookups is basically free SRS, the more you come across a word in the wild, the easier it will be to remember it later too.
Thanks to Bunpro I don’t really have to do grammar lookups. Not saying they’ve covered all the grammar there is, but at least the most common ones that appear in books and conversations.
I started using jpdb in 2022 for reviews, by then I had read 30 novels. When I come across a word I don’t know I add it to my vocab deck. If it’s an expression or yojijukugo I have other decks. If it’s a word I want to appear in lessons asap, I press “add to front of review queue”.
Im using wanikani for kanji, bunpro for grammar, but also bunpro vocab decks (I’m now doing kaishi 1.5k)
Do you think is better to move to jpdb or jiten for vocab and start with a deck of maybe よつばっと instead? or doing kaishi 1.5k there?
I’ve never used bunpro for their vocab srs, so I can’t answer that, but knowing the most common words is a good idea, so I’d say keep up the work with the kaishi deck.
If you’re reading Yotsuba, you could add it as a deck on bunpro or jiten (seems like jpdb doesn’t have a build-in deck) or add the words to your own deck on your preferred srs site.
What worked for me when using jpdb is that when you add decks you can choose your preferred mode for doing lessons. Thinking in terms of efficiency, it’s a good idea to learn the most common words first, either generally or in the particular deck/book. If the words are rare it’s okay to skip to prioritize more common words first. It’s not like you have to learn 100% of the words in the deck.
My guess is that that happens because the card scheduling algorithm is different and the new one wants more reviews given the past review history. There might be some tweakable config knob to help with that, but I haven’t checked.
Yes, but unless you have a way to conveniently create cards for the content you’ve picked (which is a pain if you read on paper), having prebuilt decks is handy. It’s the main reason I started using jpdb in the first place.
100% agree with you. Which is why I think it’s more suited for more experienced users that have those kind of considerations. Right now I must be full digital or it is too painful to go through material and thus having my custom decks is more flexible (as I work veryyyy slowwwly though my first novel).
Maybe the “external rating” is more reliable.
