JWN

@@ -6,0 +6,3 @@
+</k_ele>
+<k_ele>
+<keb>寒ぞら</keb>

@@ -22,0 +23 @@
+<gloss>shortly</gloss>

Google n-grams has the kana form leading.

@@ -20,0 +21 @@
+<misc>&uk;</misc>
@@ -26,0 +28 @@
+<misc>&uk;</misc>

間もなく	53214
間も無く	2506
まもなく	45414

roughly 1:1 on the n-grams but only 38/243 of the tanaka corpus sentences have kanji. same situation as たいてい?
-
perhaps a manual addition to the text-glossing kana dictionary, as was done with 1587580?

@@ -27 +27,2 @@
-<gloss>(after verb) lacking time to ...</gloss>
+<s_inf>after verb</s_inf>
+<gloss>lacking time to ...</gloss>

I think the best approach is to use a combination of the proportion of forms and the absolute numbers. Having a form out there like オンナノコ with over 100k usages in WWW pages and not having any mention of it in the entry would be a mistake IMNSHO. Having a rule-set such as "either 2% or more of surface forms, or 10k or more in the n-gram counts" would be better.
WWWJDIC will  catch オンナノコ for regular lookups because for kana-only cases it maps おんなのこ and オンナノコ together. It won't catch it for text-glossing because it doesn't fold the kana versions together there (for good reason). I'd certainly like to be able to handle text containing オンナノコ correctly without having to do some quite messy reprogramming.

Yes, I agree with Robin. I've said this plenty of times before - it doesn't make sense to just look at the raw numbers, or all our P-tagged entries should be horrible messes with lots of different and rare versions. There's a balance we need to strike between presenting easy-to-read entries, and trying to include absolutely everything. 

It's also worth noting that Yomichan for example has no problem whatsoever displaying the correct entry when presented with オンナノコ, even if that reading isn't included in the entry. (jisho falters, but that's something I think Kim could easily fix if it was brought to his attention)

I don't think it should be included. We generally don't include katakana forms unless they account for a significant proportion of the total hits. With less than 1% of the hits for 女の子, オンナノコ doesn't pass that test. I don't think it's helpful to have obscure katakana forms.

100k is a lot of hits. Worth having.

@@ -17,0 +18,4 @@
+</r_ele>
+<r_ele>
+<reb>オンナノコ</reb>
+<re_nokanji/>

女の子   	14136830
女のコ   	263573
女の児   	1935
おんなのこ	124421
おんなのコ	3647
オンナノコ	102435

even 女のコ is barely needed, gets less than 2% of the total hits, on the level of what I've been applying [rK] too. オンナノコ gets less than 1%, so I'm not really seeing a strong case for including it.

@@ -13,3 +12,0 @@
-<k_ele>
-<keb>女の児</keb>
-</k_ele>
@@ -18,2 +14,0 @@
-<re_restr>女の子</re_restr>
-<re_restr>女の児</re_restr>
@@ -24,7 +18,0 @@
-<r_ele>
-<reb>おんなのコ</reb>
-<re_restr>女のコ</re_restr>
-</r_ele>
-<r_ele>
-<reb>オンナノコ</reb>
-</r_ele>
@@ -34,2 +22,2 @@
-<xref type="see" seq="1420010">男の子・おとこのこ・1</xref>
-<xref type="see" seq="1420010">男の子・おとこのこ・1</xref>
+<xref type="see" seq="1420010">男の子・1</xref>
+<xref type="see" seq="1420010">男の子・1</xref>
@@ -43,2 +31,2 @@
-<xref type="see" seq="1420010">男の子・おとこのこ・2</xref>
-<xref type="see" seq="1420010">男の子・おとこのこ・2</xref>
+<xref type="see" seq="1420010">男の子・2</xref>
+<xref type="see" seq="1420010">男の子・2</xref>

@@ -26,0 +27,3 @@
+</r_ele>
+<r_ele>
+<reb>オンナノコ</reb>

女の子	14136830
女のコ	263573
女の児	1935 ← do we need this?

@@ -12 +11,0 @@
-<ke_pri>spec1</ke_pri>
@@ -28 +26,0 @@
-<re_pri>spec1</re_pri>

Bizarre. How did that make it in, and 
stay undetected for this long? Thanks 
for noticing it.

What is the source for the おんにゃのこ reading? It sounds like it is how this word is pronounced when someone is mimicking a cat 
speaking. 
I think it should be removed. It doesn't appear in the Kotobank entry.

@@ -30,4 +29,0 @@
-<r_ele>
-<reb>おんにゃのこ</reb>
-<re_nokanji/>
-</r_ele>

@@ -36,0 +37 @@
+<xref type="see" seq="1420010">男の子・おとこのこ・1</xref>
@@ -43,0 +45 @@
+<xref type="see" seq="1420010">男の子・おとこのこ・2</xref>

Daijirin: (1) 女である子供。女児。(2) 俗に、若い女性。
GG5: (1)〔少女〕 a girl; 〔娘〕 a daughter; 〔女の赤ちゃん〕 a baby girl. (2)〔若い女性〕 a young woman; a girl.
中辞典: 〈娘〉 a daughter; 〈若い女〉 a girl; 〈子供〉 a (little) girl; 〈赤ん坊〉 a baby girl

I don't think it's a separate sense - it's more an indication of the breadth of the main sense. I'd prefer to follow the references on this.

@@ -38,4 +37,0 @@
-</sense>
-<sense>
-<pos>&exp;</pos>
-<pos>&n;</pos>
@@ -42,0 +39 @@
+<gloss>baby girl</gloss>

I feel like having "girl" and "daughter" together makes it look like it mainly refers to someones daughter, rather than any young female in any context. 
Feels like two separate senses to me.

@@ -37,0 +38,4 @@
+</sense>
+<sense>
+<pos>&exp;</pos>
+<pos>&n;</pos>

Daijr, GG5

@@ -37,0 +38,6 @@
+<gloss>daughter</gloss>
+</sense>
+<sense>
+<pos>&exp;</pos>
+<pos>&n;</pos>
+<gloss>young woman</gloss>

@@ -34,0 +35 @@
+<pos>&exp;</pos>

gg5, daijr

I don't think "carrying someone on shoulders" is right.

@@ -22,3 +22 @@
-<gloss>lifting a person high (in celebration)</gloss>
-<gloss>tossing someone into the air</gloss>
-<gloss>carrying someone on shoulders</gloss>
+<gloss>tossing (someone) into the air (in celebration)</gloss>

I think "give up on" is less ambiguous

@@ -19 +19 @@
-<gloss>to give up</gloss>
+<gloss>to give up on</gloss>

I think the original edit was caused by someone reading this rather confused thread, for what it's worth:
https://old.reddit.com/r/LearnJapanese/comments/gfo5yv/do_i_understand_this_sentence/

The entry IMO should mention that it's disparaging in use.

Daijr, GG5

I don't think a second sense is needed.

@@ -15 +15 @@
-<gloss>full of human traits</gloss>
+<gloss>human-like</gloss>
@@ -17,5 +17 @@
-</sense>
-<sense>
-<pos>&adj-i;</pos>
-<gloss>smelling like a human</gloss>
-<gloss>reeking of human</gloss>
+<gloss g_type="lit">smelling like a human</gloss>

"full of human traits" is a bad gloss imo

https://kotobank.jp/word/人間臭い-593703
１ 人間の体臭のようなにおいがする。「―・いにおいがするぞ」

@@ -17,0 +18,5 @@
+<sense>
+<pos>&adj-i;</pos>
+<gloss>smelling like a human</gloss>
+<gloss>reeking of human</gloss>
+</sense>

@@ -14 +14 @@
-<pos>&n;</pos>
+<pos>&adj-i;</pos>

1) Text at hand - the meaning did not resolve when using "Text Glossing", since 
only the kanji form is recognized at present.
2) Found 684 unique references in Google for the form 人間くさい
3) EIJIRO

@@ -6,0 +6,3 @@
+</k_ele>
+<k_ele>
+<keb>人間くさい</keb>

OK, let's close it for now.

So to clarify my stance, I don't disagree with Robin that 
it's kind of dubious how helpful these glosses really are, 
but I prefer having them to moving what I think should 
obviously be usage notes to expl. I think we should try and 
be as consistent with our expl/note usage as possible.

Reopening.
In a forked edit Marcus commented:
I don't see how the [expl] glosses would be different from
the notes, though...? I mean, if they are notes on usage,
they should be in the notes. I've said it before on the
mailing list though; I think we should let through senses
which have notes but no gloss into the database.

OK. I'll approve to lock in this order and fix examples, and then reopen. I've suggested some extra glosses for 4 which match what I'm seeing.

@@ -28 +28,4 @@
-<gloss>now</gloss>
+<gloss>now, ...</gloss>
+<gloss>well, ...</gloss>
+<gloss>I tell you!</gloss>
+<gloss>you know</gloss>
@@ -34 +37 @@
-<gloss>o</gloss>
+<gloss>ooh</gloss>

I just don't think it's helpful to translate a particle like this (perhaps with the exception of sense 1).
I don't associate "do" with imperative sentences and I think it's impossible to come up with translations for senses 4 and 5. I can't see how "now", "wow" and "o" would aid with translation or understanding.
Since we can't just have notes, I'd make [expl] glosses out of the notes.

@@ -19 +19 @@
-<pos>&prt;</pos>
+<pos>&int;</pos>

But sense 1 was previously literally just 
"emotionality/emphasis". "emphasis" is in 
sense 4 and "emotionality" is in sense 5.

Having trouble seeing it. There are 200+ sentences for the former sense 1, e.g. 
「相変わらず短気だな」麗華の拳をいなしながら言う。"Short tempered as ever." He said while dodging Reika's fist. 
「金なら貸さないからな」、俺はにべもなく言った。"If it's money then I'm not lending any" I said point blank. 
I can't see then fitting senses 4 or 5, but they're fine for the old sense 1.
(Something odd going on with xrefs.)

@@ -20,2 +20 @@
-<xref type="see" seq="2100620">なあ・1</xref>
-<xref type="see" seq="2100620">なあ・1</xref>
+<xref type="see" seq="2100620">なあ</xref>
@@ -33,2 +31,0 @@
-<xref type="see" seq="2100620">なあ・1</xref>
-<xref type="see" seq="2100620">なあ・1</xref>

Senses 4 and 5.

@@ -20,2 +20,2 @@
-<xref type="see" seq="2100620">なあ</xref>
-<xref type="see">なあ</xref>
+<xref type="see" seq="2100620">なあ・1</xref>
+<xref type="see" seq="2100620">なあ・1</xref>
@@ -33,2 +33,2 @@
-<xref type="see" seq="2100620">なあ</xref>
-<xref type="see">なあ</xref>
+<xref type="see" seq="2100620">なあ・1</xref>
+<xref type="see" seq="2100620">なあ・1</xref>

What's happened to the former sense 1? It has an example sentence and there's an xref pointing to it.

@@ -21 +21 @@
-<xref type="see" seq="2100620">なあ</xref>
+<xref type="see">なあ</xref>
@@ -34 +34 @@
-<xref type="see" seq="2100620">なあ</xref>
+<xref type="see">なあ</xref>

chujiten

A first attempt to improve this entry. It probably needs 
some work, still, not least to add the various meanings it 
can have in ksb...

@@ -10 +10,2 @@
-<gloss>(sentence end, mainly masc.) indicates emotion or emphasis</gloss>
+<s_inf>prohibitive; used with dictionary form verb</s_inf>
+<gloss>don't</gloss>
@@ -14,3 +15,2 @@
-<misc>&abbr;</misc>
-<s_inf>from なさい</s_inf>
-<gloss>(used with masu stem verb) command</gloss>
+<s_inf>imperative; used with masu stem verb</s_inf>
+<gloss>do</gloss>
@@ -20 +20,18 @@
-<gloss>(used with dictionary form verb) prohibition</gloss>
+<xref type="see" seq="2100620">なあ</xref>
+<xref type="see" seq="2100620">なあ</xref>
+<gloss>hey</gloss>
+<gloss>listen</gloss>
+<gloss>you</gloss>
+</sense>
+<sense>
+<pos>&prt;</pos>
+<s_inf>when seeking confirmation, for emphasis, etc.; used at sentence end</s_inf>
+<gloss>now</gloss>
+</sense>
+<sense>
+<pos>&prt;</pos>
+<xref type="see" seq="2100620">なあ</xref>
+<xref type="see" seq="2100620">なあ</xref>
+<s_inf>used to express admiration, emotionality, etc.; used at sentence end</s_inf>
+<gloss>wow</gloss>
+<gloss>o</gloss>

Entry created

@@ -15 +15 @@
-<s_inf>imperative (from なさい); used with masu stem verb</s_inf>
+<s_inf>imperative (from なさい); used with -masu stem of verb</s_inf>

Nicolas: No worries.
Jim: looks better.

Just to be clear, I don't do it on purpose.

Nicolas (forked comment): "I'm ok with Marcus's suggestion."
Please note that comments and amendments MUST be attached to the most recent addition. Attaching a comment to an earlier one creates a "fork" in the flow of amendments, which can only be resolved by rejecting a branch.

Well, I've implemented my suggestion from last month. OK?

@@ -15,2 +15 @@
-<xref type="see" seq="2192950">なさい</xref>
-<s_inf>imperative; used with masu stem verb</s_inf>
+<s_inf>imperative (from なさい); used with masu stem verb</s_inf>

I'd missed the addition of "abbr"; I agree it's not. I'd prefer a note about なさい, as there 
was earlier. Reinstate and remove xref?

@@ -16 +15,0 @@
-<misc>&abbr;</misc>

I don't think it should be tagged as abbr (and I don't think the xref to なさい is helpful).
daijr for example says 〔補助動詞「なさる」の命令の言い方「なさい」を省略したものから
なさい is the etymology, it's not right to say it's an actual abbreviation of it today.

Until 2018 there was a "from なさい" note,

https://japanese.stackexchange.com/questions/2707/using-な-in-positive-instead-of-negative-imperative-e-g-行きな

@@ -14,0 +15,2 @@
+<xref type="see" seq="2192950">なさい</xref>
+<misc>&abbr;</misc>

OK, let's close it for now.

So to clarify my stance, I don't disagree with Robin that 
it's kind of dubious how helpful these glosses really are, 
but I prefer having them to moving what I think should 
obviously be usage notes to expl. I think we should try and 
be as consistent with our expl/note usage as possible.

Reopening.
In a forked edit Marcus commented:
I don't see how the [expl] glosses would be different from
the notes, though...? I mean, if they are notes on usage,
they should be in the notes. I've said it before on the
mailing list though; I think we should let through senses
which have notes but no gloss into the database.

OK. I'll approve to lock in this order and fix examples, and then reopen. I've suggested some extra glosses for 4 which match what I'm seeing.

@@ -28 +28,4 @@
-<gloss>now</gloss>
+<gloss>now, ...</gloss>
+<gloss>well, ...</gloss>
+<gloss>I tell you!</gloss>
+<gloss>you know</gloss>
@@ -34 +37 @@
-<gloss>o</gloss>
+<gloss>ooh</gloss>

I just don't think it's helpful to translate a particle like this (perhaps with the exception of sense 1).
I don't associate "do" with imperative sentences and I think it's impossible to come up with translations for senses 4 and 5. I can't see how "now", "wow" and "o" would aid with translation or understanding.
Since we can't just have notes, I'd make [expl] glosses out of the notes.

@@ -19 +19 @@
-<pos>&prt;</pos>
+<pos>&int;</pos>

But sense 1 was previously literally just 
"emotionality/emphasis". "emphasis" is in 
sense 4 and "emotionality" is in sense 5.

Having trouble seeing it. There are 200+ sentences for the former sense 1, e.g. 
「相変わらず短気だな」麗華の拳をいなしながら言う。"Short tempered as ever." He said while dodging Reika's fist. 
「金なら貸さないからな」、俺はにべもなく言った。"If it's money then I'm not lending any" I said point blank. 
I can't see then fitting senses 4 or 5, but they're fine for the old sense 1.
(Something odd going on with xrefs.)

@@ -20,2 +20 @@
-<xref type="see" seq="2100620">なあ・1</xref>
-<xref type="see" seq="2100620">なあ・1</xref>
+<xref type="see" seq="2100620">なあ</xref>
@@ -33,2 +31,0 @@
-<xref type="see" seq="2100620">なあ・1</xref>
-<xref type="see" seq="2100620">なあ・1</xref>

Senses 4 and 5.

@@ -20,2 +20,2 @@
-<xref type="see" seq="2100620">なあ</xref>
-<xref type="see">なあ</xref>
+<xref type="see" seq="2100620">なあ・1</xref>
+<xref type="see" seq="2100620">なあ・1</xref>
@@ -33,2 +33,2 @@
-<xref type="see" seq="2100620">なあ</xref>
-<xref type="see">なあ</xref>
+<xref type="see" seq="2100620">なあ・1</xref>
+<xref type="see" seq="2100620">なあ・1</xref>

What's happened to the former sense 1? It has an example sentence and there's an xref pointing to it.

@@ -21 +21 @@
-<xref type="see" seq="2100620">なあ</xref>
+<xref type="see">なあ</xref>
@@ -34 +34 @@
-<xref type="see" seq="2100620">なあ</xref>
+<xref type="see">なあ</xref>

chujiten

A first attempt to improve this entry. It probably needs 
some work, still, not least to add the various meanings it 
can have in ksb...

@@ -10 +10,2 @@
-<gloss>(sentence end, mainly masc.) indicates emotion or emphasis</gloss>
+<s_inf>prohibitive; used with dictionary form verb</s_inf>
+<gloss>don't</gloss>
@@ -14,3 +15,2 @@
-<misc>&abbr;</misc>
-<s_inf>from なさい</s_inf>
-<gloss>(used with masu stem verb) command</gloss>
+<s_inf>imperative; used with masu stem verb</s_inf>
+<gloss>do</gloss>
@@ -20 +20,18 @@
-<gloss>(used with dictionary form verb) prohibition</gloss>
+<xref type="see" seq="2100620">なあ</xref>
+<xref type="see" seq="2100620">なあ</xref>
+<gloss>hey</gloss>
+<gloss>listen</gloss>
+<gloss>you</gloss>
+</sense>
+<sense>
+<pos>&prt;</pos>
+<s_inf>when seeking confirmation, for emphasis, etc.; used at sentence end</s_inf>
+<gloss>now</gloss>
+</sense>
+<sense>
+<pos>&prt;</pos>
+<xref type="see" seq="2100620">なあ</xref>
+<xref type="see" seq="2100620">なあ</xref>
+<s_inf>used to express admiration, emotionality, etc.; used at sentence end</s_inf>
+<gloss>wow</gloss>
+<gloss>o</gloss>

Entry created

not really [exp] either, but that's how じゃない and related entries are. closer to the copula if you ask me.

@@ -8,2 +8,2 @@
-<pos>&aux-v;</pos>
-<misc>&sl;</misc>
+<pos>&exp;</pos>
+<misc>&col;</misc>

Entry created

ｐ２６５日本語表現文型辞典 http://jlptdictionary.com/viewentry.php?intEntryID=17819

Not a mere denial of わけにはいかない

I think using A and B makes it clearer.

I feel that's very opaque. Also to be fair 
you don't have one single style - it 
changes wildly from entry to entry.

That's not our usual style.

@@ -13 +13 @@
-<s_inf>as Aといい、Bといい</s_inf>
+<s_inf>as ...といい...といい</s_inf>

@@ -13 +13 @@
-<s_inf>as ...といい...といい</s_inf>
+<s_inf>as Aといい、Bといい</s_inf>

daijr, DAJG

I don't think we need a specific example.

@@ -13,3 +13,4 @@
-<s_inf>as in 色といいデザインといい</s_inf>
-<gloss>both... and...</gloss>
-<gloss>... as well as ...</gloss>
+<s_inf>as ...といい...といい</s_inf>
+<gloss>in terms of both ... and ...</gloss>
+<gloss>from the perspective of ... as well as ...</gloss>
+<gloss>whether it be ... or ....</gloss>

Daijr, GG5, 中辞典

Two different "いい". The first is いい/よい and the second 言い/いい. I'll reduce this just to the second sense. Closing and reopening.

@@ -3,0 +4,3 @@
+<k_ele>
+<keb>と言い</keb>
+</k_ele>
@@ -9,5 +12,2 @@
-<gloss>it would be nice if ...</gloss>
-<gloss>I hope ...</gloss>
-</sense>
-<sense>
-<pos>&exp;</pos>
+<misc>&uk;</misc>
+<s_inf>as in 色といいデザインといい</s_inf>
@@ -14,0 +15 @@
+<gloss>... as well as ...</gloss>

http://jlptdictionary.com/viewentry.php?intEntryID=17433
https://www.youtube.com/watch?v=7XZdhadlS68&list=PLINFE8v4DOhszv2Z5hTknh41IIN3ejuO4
http://jlptdictionary.com/viewentry.php?intEntryID=17395

My primary reason for including this is sense 2. While rather obvious, I added sense 1 too because having only #2 would confuse dictionary users, and since #1 is more common, I'm leading with that one.

Also I don't think it makes sense to say that the bakumatsu/Meiji cutoff is "vague" and instead suggesting replacing it with something like 
"old word that appears to have been somewhat common in the past but no longer". Which is the vaguer definition of "before 1868" and "in the 
past"?

Nikk looks for the oldest examples they can find, I believe. So the 1873‐74 usage will have been the earliest, not the latest, recorded use 
of it they could find. I agree how we use our arch-obs-dated etc. tags need to be looked over, but having some kind of a cutoff between 
Bakumatsu and Meiji makes sense. Lots and lots of books written in the Meiji period are still being widely read, while the same isn't true 
for Edo period literature.

to be fair, that bakumatsu cutoff is vague and doesn't really make a lot of sense.

what does the cutoff actually mean?  a single recorded usage of a word after the meiji restoration means that it no longer qualifies as being "archaic"?  we don't really have any way of gauging how common the usage of a particular word was throughout history.  our current distinction between "arch" and "obs" doesn't seem like the kind of thing a bunch of non-native speakers can really judge

"arch" should mean "old word that appears to have been somewhat common in the past but no longer"
"dated" should mean "old word that is still used but sounds old fashioned (and possibly inappropriate)"
"obs" should mean "old word with a very specific meaning that has been supplanted by another word"
"hist" should mean "current word that refers to a historical concept"
"obsc" should mean "a word that is rare for whatever reason (possibly because it is archaic but we don't have enough evidence to really say that)"

https://www.youtube.com/watch?v=xWxI-wnOJuc&t=4693s

See around the 1:08:50 mark

Nicolas, what was the context, exactly? It could be [hist], I suppose. 
But even if it's not [hist], the nikk example is from 1873‐74. Our threshold for arch is still bakumatsu or earlier, I believe? So arch is 
probably not right.

gg5
掃除屋　a cleaner; a housecleaning agency.

sense 1 is not.  sense 2 is probably obs, arch, or hist (and appears to be from the ref).  there aren't many people in japan gathering shite at night anymore

@@ -12 +12 @@
-<gloss>cleaner (person)</gloss>
+<gloss>cleaner</gloss>
@@ -17,2 +17,2 @@
-<gloss>night soil man</gloss>
-<gloss>night soil woman</gloss>
+<misc>&arch;</misc>
+<gloss>night soil gatherer</gloss>

https://kotobank.jp/word/掃除屋-2056533

Just saw this in a JLPT question, so not arch or obsc