So, the question about those is that should they have completely different meanings in the dictionary, or should the differences be handled via usage notes or something else. Currently, all of these pairs share the exact same meanings in renshuu.
I’d argue if they aren’t fully interchangeable, they shouldn’t be sharing the same entry. Otherwise it will be too hard to learn how to use them correctly.
It should be fairly easy to understand, but please read the top.
Also do note that I am still making changes to the number-based duplicates, so what is on the left (keep) and what is on the right (lose) may change.
Now, if you see things that you think should be separate, or feel you have duplicates that are a)in renshuu materials and b)not on that page, please let me know.
do not have published differences in JMDict, which renshuu is based on.
Thank you!
That‘s a bit odd with those two, isn‘t it… I‘m surprised they aren‘t different in JMDict… But maybe that dictionary, as well as renshuu, need to draw a line at some point, choosing simplicity over correctness… The difference between 会う/逢う seems to be explanation worthy even to native japanese speakers
(Found a Forbes Japan article on that, for people interested: here)
I like the new test page and think it will be good to implement, i think adding an option to select which reading to use in addition to this would synergize well, very similar to the experimental "study rare kanji" option, but with a selectable menu of which specific version to add into a schedule to work up to (with the standard version being default). that way people can still study the version that aligns with their study material or the one that they see most commonly in their conversations or reading.
Thinking about that… wouldn‘t it be a good idea to exclude „duplicates“ from cleanup that map different kanji altogether?
Basically the kanji is carrying a meaning, so that kanji is usually the reason for slight differences in meanings… Also, most likely, users added those „duplicates“ to their schedules on purpose (originating from songs, literature, textbooks, …)
Could that be a clean cut, where those number-duplicates and kanji-kana-duplicates are still removed?
ハイム: this is not possible, sorry. The renshuu lists are getting
updated with the primary version, and I cannot set up a system to let
the users choose to use that one or an alternate one. You'll be able to,
on your own, add the alternate and block the primary if you want, but
adding another layer of selection is too complex. The "use rare kanji"
is a different thing - it's working with a single term, and whether or
not to use the kanji on that term. These cases are all with multiple,
*separate* terms, and the system cannot swap one in or out based on user
preferences.
Ultimately, those, these are just the ones that pop
up in renshuu lists, and users can do changes on top of that with
whatever works best for them.
3Neko: I respectfully disagree - if
the kanji do not signify any kind of difference in meaning (and then
often the case is that the term is in hiragana by default, and so the
difference in kanji "mean" even less, then keeping both versions (or
multiple versions) in does not make sense to me.
Regarding the
meanings, if it's a difference in usage, but not in explicit meaning,
I'd tend towards suggesting those go as usage notes, not as a change in
meaning (I'm looking at あう in particular with this one). It's not a
clear-cut case, so I'll just have to make a decision in the end.
セージ:
you are reporting on terms that are already separate in the renshuu
dictionary, and so would not be covered by this in the first place. Only
ones that appear on the merge page are up for discussion.
You'll be able to,
on your own, add the alternate and block the primary if you want, but
adding another layer of selection is too complex. The "use rare kanji"
is a different thing - it's working with a single term, and whether or
not to use the kanji on that term. These cases are all with multiple,
*separate* terms, and the system cannot swap one in or out based on user
preferences.
Ultimately, those, these are just the ones that pop
up in renshuu lists, and users can do changes on top of that with
whatever works best for them.
I think this still achieves the desired effect. I can definitely understand how extra selection layers will mess things up. I'm very keen to see this implemented.
one thing I think this will encourage is more usage notes; 生む、産む are the same word in both reading and (fundamental) dictionary definition, but I've found that each tends to get used for certain contexts, though its more an implication than anything tht should be included in the standard definition. 才 and 歳 are also the same, but the kanji used seems to largely depend on the way the person was taught.
Looking at mine, a lot seem reasonable, but I would really like the ability to quiz on multiple different ways to write something (no matter whether that’s different kanji 円／丸 or okurigana 受け取る／受取る or even just hiragana-vs-katakana because I want to be able to recognize both of those as the word in question. (This also applies to stuff already in the same entry but “also written as” and rare kanji vs kana versions.)
You can study multiple ways if you want - that has always been present and will not be removed. This is just for unifying the entries in renshuu-based lists to a single version, and optionally (that's what the page is for) merging the alternate versions into the main one so your account is cleaned up. All alternate versions will show up in the dictionary, and you can add them (or do anything else) like before. :)
Hi, when I look at my list now, I will miss all the words with 御 and 程 and such. They're good for kanji recognition. But less words overall will be good.
Here are a few specific points:
- いよいよ is rarely written in kanji according to the dictionary, but on the merge page I only get the option between 愈 (main) and 愈々 (remove). If the kanji version is rare, can I deactivate it completely or get another option to have the hiragana as main?
- same with など/等/ 抔 and おばあちゃん/ お婆ちゃん/ 御婆ちゃん
- If the different kanji version for different nuances will be added to words as usage notes, the kanji helper will probably not add the rarer version?
- When I currently study a kanji, I can get example words like 暖まる from 暖. This would be removed as a separate word and instead 温まる be kept. Will this affect the examples in the kanji study? So that 暖まる won't really show up unless I specifically search for it in the dictionary?
1. While the kanji is explicitly written on the merge page, if it's rare kanji, it will still appear as hiragana in quizzes :)
2. This only affects the words in renshuu's main lists - how the terms are listed in the dictionary, and how usage notes are attached to each version will not change :)
3. There are no merging of words in the dictionary, just renshuu's lists. So if you study a kanji, the relevant example words will still appear :)
Note to all: I feel pretty comfortable with the process, so I should have the test merge available in the first half of next week for testing! You'll receive a private message (like before) when it's ready.
When you start to use the merge tool, it will make a copy of your study data for all affected words, so just in case a bug pops up, there will be a way to revert the study data for the merged terms.
