More seriously I agree fully with this but here we’re talking about words taught in the very few levels, where the average student will know a couple hundred words. At this point WaniKani will not even have taught you words like “to buy”, “thank you” or even すみません.
That’s why the reality of stamp usage in Japan is a bit irrelevant in this context IMO. The average student will be learning ultra basic Japanese at this point. I think 切手 would make more sense at level 30 or something.
I apologize for being a bit curt yesterday, I just had a very similar discussion on the bunpro forums a little while ago and it frustrated me. There we were arguing that words like “antibiotic” or “ophthalmologist” didn’t make sense in the N5 deck and like clockwork some people started arguing that they encountered them regularly in Japan.
I’m sure that’s true but that’s not my point, the size of the vocabulary you encounter day-to-day in Japan is vastly greater than the 1000 words in bunpro’s N5 deck or WaniKani’s first ten levels.
In this context with such a constrained word budget I don’t think it’s reasonable to teach words like ophthalmologist, antibiotic or stamp unless in the case of WaniKani they’re here to reinforce the kanji knowledge, but as I was arguing in my first comment I think 切手 is a bad choice for this due to the surprising reading and meaning. At least 眼科 and 抗生物質 can be remembered from the kanji alone…
Mmm. In general people are really bad at guessing how frequently they actually encounter a particular word or bit of grammar – either you remember a couple of times you saw it recently and it feels like it’s common, or else you don’t happen to be able to bring an instance to mind so it must be rare. I think you have to look at the numbers from some hopefully representative corpus.
The other part of this is that an ideal system ought to be open-ended, so maybe it has some pretty solid guiderails to start you off, but at some point you’d like it to be possible to match it to what you personally encounter, whether that’s daily life in Japan or isekai light novels or the fact you’re working through Genki I. That way, if 切手 or 眼科 are useful to you then you can spend study time on them and if they’re not you can ignore them until they do seem more relevant.
Yeah there we have the core issue with “fixed decks” like WaniKani but in my opinion this really becomes a problem once you go beyond the top ~2000 words or so. Early on there’s a common core of vocab like “to say, to buy, number, place, people, husband, door, happy, sad, slow, idea, etc…” that you want to learn regardless of your objectives.
That’s why overall I think WaniKani generally works well until level 30 or so but after that it’s just frustrating to be forced to learn words and kanji you don’t necessarily care for at the moment while being prevented from learning those you want from latter levels.
They’ve actually improved a lot on that this past year or so. Very common words like 俺 or 誰 for instance used to be far into the course and have been moved up a lot. Meanwhile they kicked some rare words way down.
Of course due to the kanji-first approach there are always going to be oddities in the vocabulary order and selection, that’s unavoidable.
No worries, did not come across that way to me at all. Also I loved the cassette and Walkman reference. What made it even better is that I was reading that on the train ride back home having just been to the retro record store and having just had a conversation with my wife about the original Sony Walkman I had while we were walking back to the station. For the record (pun intended) the outing was not to visit that shop. We had gone out for a special lunch in that area and just happened to pass by it.
Here to throw in a vote supporting 切手 being in the vocab list.
I crushed through the first 12 levels or so of WaniKani in preparation for a two-week trip to Japan last year, having done zero prior study in Japanese. It saved my rear on a number of occasions, particularly when trying to navigate through various 駅 to the correct line/platform. 切符 was easy for me to remember because I knew of another “odd” — but weirdly kind of related in my brain? — usage case of 切: 切手. Very helpful when making a hurried search for the nearest 切符売り場 so I didn’t miss the soon-arriving 新幹線 and have to wait an additional 45-60 minutes for the following one.
(Yeah, I know you can buy the tickets ahead of time, but I purposefully left my schedule a bit fluid between a handful of set-in-stone items.)
But note that you only needed this trick because 切符 is so far on wanikani (level 30) that you didn’t have the chance to learn it yet, meanwhile it’s in bunpro’s N5 deck as well as Tango N5 for instance (it’s also tagged N5 on Jisho).
That’s one of the oddities we were talking about in previous comments, it doesn’t really make a lot of sense to separate those two words that much normally as far as vocab is concerned, but because WaniKani goes kanji first it makes sense that 切手 would be taught very early (due to the extremely common kanji) while 切符 would be postponed because 符 is a lot less useful overall.
切符 appears to also be more common overall than 切手, although not by a huge margin.
船 (a ship) is a very common word.
体 (a body) is a very common word.
船体 (a hull) is a word that teaches you the on reading of both 船（ふね）and 体（からだ).
It’s not always necessarily about learning the word “hull” (which, even on its own, isn’t the most commonplace word in English), learning the additional word also just burns the 2 kanji into memory even more than knowing them individually would. I think.
As for postage stamp, it follows the same logic.
切手 (cut hand) might not make much sense as “postage stamp”, but it teaches you some basics of Japanese, namely: words can mean wholly different things than their kanji composition, but the readings are the same (… often).
for a similar example, but a more useful word, see: 皮肉.
A friend of mine taught me the word before Wani did as an example of “absurd” words, and I’ve remembered it ever since.
But it does the great job of teaching you that whilst also retaining a perfectly normal reading (hi + niku) which means it’s a very easy word to remember at least 1 half of.
In that sense it’s probably better than 切手 (きって).
I do think that words like せんたい、ひにく、したい (死体) help you to realize a lot about the meaning of compounds and such, how to read words, what they often “truly” mean beyond a simple single-word meaning.
an example of this: 理. Ok, it means reason. Makes no sense to me.
心理？地理？Ok, now it makes sense to me, they’re the “logic of x” denoting sciences in one way or the other. Even if it’s a bit odd that geography is in your first 1000 words, they’re easy to remember words that help your understanding of a(n apparently somewhat relevant?) kanji: 理.
Or another example I ran into earlier today:
者 vs 身. At first these are hella confusing, they look similar, they mean the same thing.
医者 (doctor) 学者 (scholar) 心身 (mind and body).
Do I need to know shinshin this early on? Nah. But it does help me understand that the latter is about a body, and the former is about “people who do x”.
Hah! Well, how about that, eh? Here I was thinking the two were related because they were both small cut-from-paper things that proved you paid for something, and I wasn’t far off the mark. I just didn’t know one was derived from the other!
Language has the tricky job of having to be able to express all of human experience and the human condition, and well, the idea that there is an ideal system is something of a pie in the sky if you ask me
I’ve had the displeasure of working in software development for my entire adult life and I assure you, there is no ideal system. Your system will displease someone at some point or another. It’s more of an issue with choosing who you’re going to please, really, and a one-size-fits-all solution like WaniKani is going to inevitably rub some people the wrong way simply by casting a wide net
That mirrors my own experience in software development but I think the best systems have sane defaults while remaining configurable for advanced use cases.
Wanikani has mostly sane defaults but is entirely inflexible.
Anki has very dubious defaults and is generally user-unfriendly, but it’s highly configurable.
The One SRS App to Bind Them All has yet to be created I’m afraid. JPDB seems interesting though, but I haven’t used it myself.
And it will never ever be created, because people have conflicting ideas of what their ideal SRS app, or well, anything app is like. If the drummer requests no cowbell and the bassist says that he’s got a fever and the only prescription is more cowbell, how do you reconcile these fundamentally opposed requests?
Of course the perfect SRS app is out of reach, but having used a bunch of those over the past decade in my studies of several languages there’s definitely a lot of room for improvement before we hit intractable conflicts of opinion.
You could improve Anki’s UI without really removing anything. You could add options to WaniKani without changing the default behaviour. I used LingQ for like a year when I studied Russian and while I found the concept really interesting, the software was absolute trash and it wasn’t cheap…
I think the real issue with language learning is that there’s no real money in it. Or at least, not in doing good language learning. Duolingo is the case study for this, it’s pretty clear that monetization incentives actually detract from making a good language learning app.
No company is going to sponsor a language learning app either. And you’re not going to make bank from Patreon. It’s pretty miraculous that Anki is as good as it is in these conditions frankly.
