One of these days I’ll get back to writing scripts (I had the bug for a while, but all my other interests are clamoring for attention).
Anyway, here’s an idea that I’d love to see implemented. Maybe I’ll get lucky and @Gorbit99 or @Kumirei will implement it before I get around to it!
I’d like a little “daily senryu” plugin for reading practice. I’m not sure whether I’d rather see it on the dashboard or on the review summary page, but I’d like it to show the Japanese by default and force me to click or at least hover to see the full translation (Japanese with furigana and English translation).
I wrote about senryu in another thread. That post links to a list of 300 Japanese-only senryu, enough for almost a year without repeats. There are other lists (with many duplicates from this one).
My first stab would be to take that list and use some scripts to automate the translation. I’d create a big JSON object with the original, original+furigana, and English translation. I’d probably just store the randomly-ordered list in the script and either step through it in sequence (keeping the user’s current index in browser storage) or, even easier, use a hash of the current date as the index. I’d use deepl or something similar for the translation, and sanity-check the translations manually.
It might be better to store the database external to the script so that new senryu could be added and translations corrected without having to create a new version of the script, but I’d definitely go with the simplest solution first. [A github repository for the database might be nice (PRs to submit changes).]
If somebody does decide to implement this (one can dream) please let me know so I can feel better about procrastinating!
I’m not against the idea, but just in case, I’ll voice some of my concerns.
Imo it’s easier to create new versions, than keeping the database in a github repo. For one, new versions automatically get applied daily usually and, since this script would be a daily senryu script, you wouldn’t need to update the database more than that. Also greasyfork has a strict policy against links that could change over time. Honestly I would just use the userscript thread as a sort of submission page. Someone sends in a poem and the translation and if it’s alright, it goes in.
Greasyfork has a limit of 2mb, I believe, as I’ve seen, one of these is around 1000-1500 bytes, the rest shouldn’t be much longer than a few kilobytes, so you could theoretically store about a thousand poems. Should be fine.
Also, you would need to find poems that are at least share alike or share with credit, you can’t just for example use a book and put in the senryu from that.
A senryu is 5+7+5 syllables. If each syllable was a 4-byte kanji (unlikely) then the longest possible would be 68 bytes, no?
Because senryu are so short and expressly not formal, serious poetry I wasn’t going to worry to much about attribution, but it’s always good to have a reminder. Contest winning entries might be copyrighted, for example.
That’s why I said I’d run it through machine translation and just manually review the results.
Many are pretty straightforward and machine translation is good enough. Some I can read (like that one) but the machines don’t quite get. A few are just beyond me with or without machine translation!
So true! (Although in the various book clubs we’ve discovered many … let’s say … sub-optimal translations
Sure, but I had the feeling that the twist-rule was still in place.
Oh, interesting - so you’re reading it as one sentence where the circle qualifies the birthday? I read it as two sentences, like so: ゴミの日と丸つけられた。誕生日。
In brute-force English: The day that’s called “Garbage day” was circled. [It’s my] birthday.
Even more nuances to be discovered with these punctuation mark-free sentences
I originally thought an 同じ or something similar was left out at the end, so I read it as one sentence meaning something like “My birthday that is circled is the same as garbage day”.
Reading it again, it could also be something like “it was my birthday that was circled as garbage day”. Although that’s pretty much just @NicoleIsEnough’s translation but stuffed into one sentence
If it were me, I’d prefer for the item to reset its status unlearned, with a timer that would delay its return to the lesson queue by a few months. (Again, pie in the sky.)
If someone could make this (or point me to an existing extension) I’s be so happy. Often I’ll answer a vocab meaning question correct because I have the link between the sound and the meaning, but not the kanji and the meaning. So the answer is ‘correct’ but I’m not strengthening the kanji meanings at the same time.
What I’m looking for is to place the kanji meanings above the kanji when the questions is answered correct. They’re already shown on the page below the hidden information so hopefully this isn’t too much of an ask. It would be super helpful for me. よろしくお願いします！
).