Personally I'm not a fan of using [sk] for readings that aren't also used as surface forms. I wouldn't go as far as to say they're pointless; it's totally possible for people to search the database for readings, for example. I agree it's probably a very marginal use-case. But I think they're also mostly harmless.

At any rate, it's not unprecedented (see entries for 中国, 三日月, 占い者, etc). We'd like to have these sorts of policy discussion on github (https://github.com/JMdictProject/JMdictIssues/discussions) or the mailing list (https://groups.google.com/g/edict-jmdict). Feel free to raise the topic in those places.

>It's 99% pointless to hide readings that aren't used as surface forms

I totally get that, but e.g. I'm someone that often searches for words in kana, not with their respective kanji. にっぽん is a valid way of reading にほん. So even if it's just that tiny tiny 1%, isn't this also what hidden forms are for? They don't bother anyone. But then again, with that approach it's very difficult to draw a line, it would be very arbitrary.

>This has to my memory been brought up several times but the only instance I can find is Stephen Kraus on 碁会所

I remember that but about [rk], which I try to remember now but still sometimes misconstrue. I can't recall seeing that for [sk] (apart from the example you just gave), but maybe that was before I started contributing


In any case I don't plan to fight tooth and nail for this [sk] form, I just wish to know clear grounds if that really doesn't fall under how it can or should be used.

This has to my memory been brought up several times but the only instance I can find is Stephen Kraus on 碁会所:
"I'm not sure if it makes sense to use [sk] for readings that aren't also used as surface forms."

sk and sK are essentially for hiding "surface forms" that people might come across written out. It's 99% pointless to hide readings that aren't used as surface forms.

Why