Why do some meanings accept their respective readings?

For example, the vocab word アメリカ人 accepts “amerikajin” as a typo.

The word 少し accepts “sukoshi”.

The first one’s just fine, I can see this being a typo, but the second one - look at the screenshot above, its meaning is “A Few”, nothing like “sukoshi”, and there are no user synonyms. It might be accepting “skosh”, but that would be quite weird, as that is part of the reading mnemonic, and also it doesn’t show as one of the meanings. Why does this happen, “sukoshi” gets accepted as the meaning?

1 Like

The English colloquial word “skosh” originally came from the Japanese 少し, so perhaps it’s there as an invisible synonym.


Thank you, but what are invisible synonyms? And why are they invisible, is there a limit or did トフグ just want some to be invisible?

They are words added to an inclusion or exclusion list for meanings, they aren’t the real meanings so they can’t be shown as one, but they will get accepted if you put them.
For example 渡す is “to hand over” but it accepts “to hand out” even though this meaning isn’t shown as an “official” meaning.
Those lists are made to be lenient because what you wrote was close enough to the meaning to not be counted as a wrong answer but it isn’t the meaning either so they can’t teach it to you as a correct one.
As for the exclusions list I think it’s for situations where your answer might be close to it when speaking in English but the Japanese word really wouldn’t translate to that in the context of the kanji.


For example, “north” is near enough to “south” that the “close enough” checker would accept either as an answer to 南, so “north” was added to the block list so that that wouldn’t happen.


Skosh was added as an acceptable reading a few weeks ago actually, like last week or so. With amerikajin, if a word is long enough, you can tack quite a few extraneous characters onto the end without tripping the typo detection.


And that would also mean that 北 would be allowed to be “south” if it weren’t for the block list?


The checker is probably actually checking if your answer comes within a few English letters of a known correct translation, rather than checking against meaning.

I suspect this for a few reasons:

  • “south” is only a couple letters off from “north”

  • Before a recent update for 点く, I had accidentally typed “to turn off” and it was marked correct, though it’s clearly wrong. In the update they added “to turn on” to the block list, which takes care of both this and the fact that it had the wrong transitivity.

  • I’m a programmer, and I wouldn’t be surprised if the check is a fuzzy check against the letters typed. It would make sense to me.

That said, I can’t say this is the case for sure, though, as I didn’t make the site.

1 Like

Yes there is. Item Inspector reveals it in the allow list.



@proleau How do you view an Item Inspector? Is it for devs/mods only?

1 Like

It is a userscript. See here.

To see the allow list you need to configure a table element to see it.

1 Like

Thank you, @proleau.

(Sent through Gmail)

1 Like

Just a small correction, It is @prouleau with a u after the o.

1 Like

Specifically, it’s checking if the Levenshtein distance is at most 3 (I think) from an accepted answer.


Not in the allow list, sadly.


This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.