Briefly about me:
I am a lecturer at a university in the Netherlands, despite my WK level, I speak Japanese at an intermediate level and won second prize in a Japanese speech contest.
I have noticed that when we look for knowledge or information on language acquisition, we tend to end up with youtube influencers or commercial services (some of them are amazing). However, I was always interested in knowing what the vast field of language acquisition would say about the questions I had in mind. To what extent is mass input empirically validated as being superior for a given time unit etc. etc.
That’s why I have built (and will continue to develop) www.learnanylanguage.com. You simply add your query and it pulls up the most relevant peer-reviewed publications on the topic to which I have legal access. It then turns it into a report which is accessible to all interested users. Technical terms are explained, and examples are given.
Over the years, I have looked at this community on Wanikani as one of the most solid and intelligent communities, that’s why I am basically launching the project here.
I hope you enjoy it! There are many planned updates to come, so I cannot immediately use your feedback, but I will definitely put it on my list!
All the best,
Fatih
EDIT: thank you so much for your input - please keep it coming - it really helps.
Had some very interesting questions come in in the last 2 days:
++++ I want to give everyone who has tested the service a month free for the plus account. You can use ‘wanikanicomm01’.
+++I would have wanted to give more, but quite frankly, even for paid members, I am barely hitting break even. At this point in time, I just want to optimize the quality of the answers as much as I can.
This is interesting. I’m not sure what else to say about it, but I did a search and the results were interesting. I’ll probably play with it a little more.
I tried to give it a go, it started “thinking” for a really long time (40 minutes or so?) before crashing, and then telling me I already got my free report for the day. :')
It was interesting - although I think that the AI made some assumptions that were not necessarily inherent in my question, but perhaps I phrased my question a bit sloppily or incompletely.
i’m not a scientist (well, I am, but not in language acquisition), but my 2c: the optimum balance is 0:0, for the reason that since japanese is neither grammar nor kanji but does make use of them. granted, because it makes use of them, they often feel useful to study, so do as much as feels useful
On 1st request I got a simple message (after 5 or 10 mins) saying an Error occurred, but inviting me to try again.
2nd request (quicker?) which said I had used my 1 unregistered request for the day, but no actual response to the query that I could see.
If I had been asking a human I would have said is there an optimum, which would have invited the response that it is different for everyone I would guess.
Not really expecting a ratio!
That’s great feedback - actually these reports cost money and I was trying to find ways to reduce the cost - that might have created a bottle neck down the line. But I have fixed the issues, and will commit the new version today (in 5 hours). Would really appreciate it if you could try again - and hopefully get a really useful answer. In general, the more specific the question the better. But broad questions are also processed reasonably well.
I have read your invaluable feedback and will commit the fix for both of the problems (time out problem, reducing credit despite failure) - would truly appreciate it if you could ask your question again in 5 hours. All the best!
I don’t have the exact wording in front of me, but I was asking for research on whether you can study a language by listening to recorded lessons while you sleep, and my intent (at least) was that it would be for someone who was already studying the language, but my impression is that the AI decided to frame the question as if it were someone with no prior experience studying that language using such a technique while learning the language from scratch.
I don’t think that I said anything in my question that could legitimately be interpreted as limiting the scope to learning a language from scratch, rather I believe that the AI ran with that as part of the definition of what research to consider.
