@@ -5,0 +6 @@
+<re_pri>spec1</re_pri>

G n-grams:

ズバリ	1,106,891
ずばり	509,538
ズバリと	24,283
ずばりと	10,764

@@ -4,0 +5,3 @@
+<reb>ズバリ</reb>
+</r_ele>
+<r_ele>
@@ -7,3 +9,0 @@
-</r_ele>
-<r_ele>
-<reb>ズバリ</reb>

@@ -13,0 +14 @@
+<misc>&on-mim;</misc>
@@ -23,0 +25 @@
+<misc>&on-mim;</misc>

レベルの高い	323,757	99.2%	
レヴェルの高い	2,712	0.8%	
レベルアップ	939,105	99.7%		
レヴェルアップ	2,603	0.3%

Splitting out "rebel".
My refs don't mention floors or layers.
The source for the software versioning sense is Wikipedia, which says it's a term used by Fujitsu. I don't think it's worth including. Not seeing any evidence of more widespread use.
I think レヴェル can be hidden. Web results are mostly "Revel".

@@ -10,0 +11 @@
+<re_inf>&sk;</re_inf>
@@ -16,2 +16,0 @@
-<gloss>amount</gloss>
-<gloss>degree</gloss>
@@ -18,0 +18 @@
+<gloss>class</gloss>
@@ -20 +20,4 @@
-<gloss>class</gloss>
+</sense>
+<sense>
+<pos>&n;</pos>
+<gloss>level (amount or quantity)</gloss>
@@ -25,4 +27,0 @@
-<gloss>floor</gloss>
-<gloss>storey (story)</gloss>
-<gloss>layer</gloss>
-<gloss>stratum</gloss>
@@ -31,0 +31 @@
+<gloss>level (instrument)</gloss>
@@ -33,9 +33 @@
-<gloss>(surveyor's) level</gloss>
-</sense>
-<sense>
-<pos>&n;</pos>
-<gloss>minor release sequence number (software versioning)</gloss>
-</sense>
-<sense>
-<pos>&n;</pos>
-<gloss>rebel</gloss>
+<gloss>surveyor's level</gloss>

OK, but I'll propose レブル for jmnedict as it's the Honda "Rebel" motorbike, which is probably where its n-grams count originates.

daij, koj

I think レブル can go. It's not in the refs and most online hits are irrelevant.
The kokugos have a "水平面。水平線" sense but it's not adjectival so I've dropped sense 4 and made a new (broader) sense using some of the sense 1 glosses.

@@ -8,3 +7,0 @@
-</r_ele>
-<r_ele>
-<reb>レブル</reb>
@@ -21 +17,0 @@
-<gloss>gradation</gloss>
@@ -24,0 +21,4 @@
+</sense>
+<sense>
+<pos>&n;</pos>
+<gloss>level (plane)</gloss>
@@ -40,5 +39,0 @@
-<pos>&adj-no;</pos>
-<gloss>flat</gloss>
-<gloss>level</gloss>
-</sense>
-<sense>

レブル	6719
レベル	23353562
レヴェル	110535
http://www.rever.co.jp/

Merging 1146250. Much of the count for レヴェル seems to be coming from people/company names.

@@ -10,0 +11,3 @@
+</r_ele>
+<r_ele>
+<reb>レヴェル</reb>

Gloss to noun form.

@@ -31 +31 @@
-<gloss>just for show</gloss>
+<gloss>doing something just for show</gloss>

I am inclined to agree. I can't see evidence of ダテ being used that much.

would it make sense, then, to remove the katakana reading?

I'm not seeing why it is included, and the instances I find on the www lead me to believe the katakana version is only used for stylistic effect (that is, to make the word stand out while avoiding kanji in hiragana-heavy sentences) or possibly because the writers feel their meaning is slangy (plausible for gloss #2); nevertheless, the katakana version is quite a bit more rare than both the kanji and hiragana versions.

@@ -13,4 +13,0 @@
-</r_ele>
-<r_ele>
-<reb>ダテ</reb>
-<re_nokanji/>

as you can see above it is not the primary one.  it appears first on wwwjdic as a result of having both [uk] and [nokanji] tags.  that is ~usually~ a good solution to the problem of biological species (which is where the [nokanji][uk] combo normally appears), as it means that 鴨[かも;カモ[nokanji]] appears as カモ 《鴨》【かも】on wwwjdic.

As with だれも.

@@ -19,2 +19,2 @@
-<gloss>nowhere</gloss>
-<gloss>nothing (with neg. verb)</gloss>
+<gloss>nowhere (with neg. verb)</gloss>
+<gloss>nothing</gloss>

Similar to the entry for いつも.  As used in a phrase like どこも悪くないんだ -- "there's nothing wrong".

@@ -16,0 +16,6 @@
+<sense>
+<pos>&conj;</pos>
+<misc>&uk;</misc>
+<gloss>nowhere</gloss>
+<gloss>nothing (with neg. verb)</gloss>
+</sense>

Clearly intransitive, as it takes particle に instead of a direct object. Ja wiktionary as well as Japanese University instruction pages such as:
http://www3.u-toyama.ac.jp/niho/kanjien/iji_dokun_v/tuku_tukeru*.html
seem to concur.

@@ -14,0 +14,1 @@
+<pos>&vi;</pos>
@@ -19,0 +20,1 @@
+<pos>&vi;</pos>

Google N-gram Corpus Counts
╭─ーーーー─┬────────┬───────╮
│ 猫かぶり │ 25,591 │ 89.0% │
│ 猫被り　 │  2,601 │  9.0% │
│ ネコ被り │    442 │  1.5% │ add, sK
│ 猫被　　 │    112 │  0.4% │ add, sK (smk)
╰─ーーーー─┴────────┴───────╯

@@ -8,0 +9,8 @@
+</k_ele>
+<k_ele>
+<keb>ネコ被り</keb>
+<ke_inf>&sK;</ke_inf>
+</k_ele>
+<k_ele>
+<keb>猫被</keb>
+<ke_inf>&sK;</ke_inf>

meikyo, prog, gg5

Not vs.

@@ -15,5 +15,4 @@
-<pos>&vs;</pos>
-<gloss>feigned innocence or naivete</gloss>
-<gloss>fake niceness</gloss>
-<gloss>fake gentleness</gloss>
-<gloss>beguiling innocence</gloss>
+<gloss>feigned innocence</gloss>
+<gloss>feigned gentleness</gloss>
+<gloss>fake modesty</gloss>
+<gloss>put-on</gloss>
@@ -24,2 +23 @@
-<pos>&vs;</pos>
-<gloss>feigning ignorance</gloss>
+<gloss>feigned ignorance</gloss>

https://dictionary.goo.ne.jp/word/en/猫被り/#je-58293

@@ -16,0 +17,2 @@
+<gloss>fake niceness</gloss>
+<gloss>fake gentleness</gloss>
@@ -18,0 +21,6 @@
+</sense>
+<sense>
+<pos>&n;</pos>
+<pos>&vs;</pos>
+<gloss>feigning ignorance</gloss>
+<gloss>playing dumb</gloss>

Dunno where that "oK" came from.

GG5

Aligning, etc.

@@ -6,0 +6,3 @@
+</k_ele>
+<k_ele>
+<keb>撫でつけ</keb>
@@ -12,1 +15,3 @@
-<gloss>smooth or flowing hair</gloss>
+<xref type="see" seq="2791590">撫で付け髪</xref>
+<misc>&abbr;</misc>
+<gloss>hair that is smoothed down</gloss>

Google N-gram Corpus Counts
82,771	 96.4%	報いる
   846	  1.0%	酬いる
 2,258	  2.6%	むくいる

@@ -11,0 +12 @@
+<ke_inf>&rK;</ke_inf>

ok

vi vs. vt is not clear.

koj has as vt, meikyo has as vi, and shinmeikai has as both.  generally takes に not を (as in gg5 examples)

i recommend dropping vi/vt altogether

@@ -21 +20,0 @@
-<pos>&vt;</pos>
@@ -28 +26,0 @@
-<pos>&vt;</pos>

Don't see the need for these x-refs.

@@ -22 +21,0 @@
-<xref type="see" seq="2005410">報う</xref>
@@ -30 +28,0 @@
-<xref type="see" seq="1858910">一矢を報いる</xref>

I'm comfortable with the "io". 
Lack of okurigana standardization is odd and at times frustrating. I have trouble convincing some people in the lexicography community that multiple surface forms abound in Japanese and that there is often no accepted canonical form. To a German that can seem heretical.

[io]?  kind of unsightly and 引受人 is the only form listed in kokugos

@@ -9,0 +9,1 @@
+<ke_inf>&io;</ke_inf>
@@ -12,0 +13,1 @@
+<ke_inf>&io;</ke_inf>

GG5, 新和英中辞典

All part of the one sense, I think. 引き受け人 gets 40k hits; 引受け人 2k.

@@ -6,0 +6,6 @@
+</k_ele>
+<k_ele>
+<keb>引き受け人</keb>
+</k_ele>
+<k_ele>
+<keb>引受け人</keb>
@@ -12,0 +18,1 @@
+<xref type="see" seq="1613400">引き受け・ひきうけ</xref>
@@ -13,0 +20,2 @@
+<gloss>acceptor (e.g. promissory note)</gloss>
+<gloss>underwriter (e.g. stock issue)</gloss>

Meikyo vt and vi additions to n,vs entries
-*- via bulkupd.py -*-

@@ -12,0 +13,2 @@
+<pos>&vt;</pos>
+<pos>&vi;</pos>

@@ -13,1 +13,2 @@
-<gloss>setting a place or the table</gloss>
+<gloss>setting a place</gloss>
+<gloss>setting the table</gloss>

no

@@ -23 +23 @@
-<gloss>charred stake</gloss>
+<gloss>charred steak</gloss>

Daijr, GG5, 新和英中辞典, ルミナス (all only have やけぼっくい)
Koj for やけぼくい:  "「やけぼっくい」に同じ".

@@ -4,0 +4,3 @@
+<k_ele>
+<keb>焼けぼっくい</keb>
+</k_ele>
@@ -11,0 +14,3 @@
+<reb>やけぼっくい</reb>
+</r_ele>
+<r_ele>
@@ -12,0 +18,2 @@
+<re_restr>焼け木杭</re_restr>
+<re_restr>焼け棒杭</re_restr>
@@ -16,0 +24,1 @@
+<gloss>ember</gloss>

I agree.

焼けぼっくい	2308	49.6%
焼け木杭	594	12.8%
焼け棒杭	297	6.4%
やけぼっくい	1451	31.2%
やけぼくい	0	0.0%

Despite it being in Koj, I think we can make やけぼくい [sk] and drop the restrictions.

@@ -18,2 +18 @@
-<re_restr>焼け木杭</re_restr>
-<re_restr>焼け棒杭</re_restr>
+<re_inf>&sk;</re_inf>

Daijr, GG5, 新和英中辞典, ルミナス (all only have やけぼっくい)
Koj for やけぼくい:  "「やけぼっくい」に同じ".

@@ -4,0 +4,3 @@
+<k_ele>
+<keb>焼けぼっくい</keb>
+</k_ele>
@@ -11,0 +14,3 @@
+<reb>やけぼっくい</reb>
+</r_ele>
+<r_ele>
@@ -12,0 +18,2 @@
+<re_restr>焼け木杭</re_restr>
+<re_restr>焼け棒杭</re_restr>
@@ -16,0 +24,1 @@
+<gloss>ember</gloss>

Split seems fine.

removing はそう/はぞう as they clearly have a different meaning (which is the only meaning given in koj)

@@ -8,4 +8,0 @@
-<keb>匜</keb>
-<ke_inf>&oK;</ke_inf>
-</k_ele>
-<k_ele>
@@ -16,0 +12,5 @@
+<ke_inf>&oK;</ke_inf>
+</k_ele>
+<k_ele>
+<keb>匜</keb>
+<ke_inf>&oK;</ke_inf>
@@ -22,3 +23,3 @@
-</r_ele>
-<r_ele>
-<reb>はそう</reb>
+<re_restr>半挿</re_restr>
+<re_restr>𤭯</re_restr>
+<re_inf>&ok;</re_inf>
@@ -28,3 +29,3 @@
-</r_ele>
-<r_ele>
-<reb>はぞう</reb>
+<re_restr>半挿</re_restr>
+<re_restr>匜</re_restr>
+<re_inf>&ok;</re_inf>
@@ -34,1 +35,1 @@
-<gloss>tea pot like object made typically of lacquer ware and used to pour hot and cold liquids</gloss>
+<gloss>teapot-like object made typically of lacquerware and used to pour hot and cold liquids</gloss>

daijr daijs koj
http://dictionary.goo.ne.jp/leaf/jn2/181514/m0u/

* Add alt readings (saw as はぞう at museum (listed in 
daijs), and dicts list many variants)
* Add rare kanji 𤭯
* Link to 盥 for second meaning, as that’s the more usual 
term, and they’re used together (and it’s confusing; v.i.).
This is a bit confusing – たらい means the basin, while はんぞ
う means both the pot, but also, confusingly (in 
daijr/daijs), the basin. See koj for はんぞうたらい, of which 
this sense is probably an abbreviation.
Museum translated it as “ewer”, which seems too specific and 
misleading (it’s like a tea pot or 湯桶, not a wide-mouthed 
ewer).

@@ -14,0 +14,3 @@
+<k_ele>
+<keb>𤭯</keb>
+</k_ele>
@@ -16,0 +19,12 @@
+</r_ele>
+<r_ele>
+<reb>はにそう</reb>
+</r_ele>
+<r_ele>
+<reb>はそう</reb>
+</r_ele>
+<r_ele>
+<reb>はにぞう</reb>
+</r_ele>
+<r_ele>
+<reb>はぞう</reb>
@@ -23,0 +38,1 @@
+<xref type="see" seq="2096310">盥</xref>

Entry created

I agree with the xref, but I think a merge is stretching it a bit too far. That みそか, etc. merge is not too clean to start with.

* Add [see] for other reading of 晦
Indeed, should these entries be merged?
(It’s messy and would mean dropping the [arch], but passes 
2/3 with 晦 “last day of month”)

@@ -18,0 +18,1 @@
+<xref type="see" seq="1604510">晦・つごもり</xref>

Entry created

daijs
サステナブルな	4963
サステイナブルな	6351

形動 in daijs.

@@ -17 +17 @@
-<pos>&adj-f;</pos>
+<pos>&adj-na;</pos>

G n-grams: (+ raw Googits)
サスティナブル	14071  (391k)
サステナブル	13666  (945k)
サステイナブル	12228  (693k)
サステーナブル	539    (13k)

サスティナブル looks a bit odd. It was winning in 2007, but not 2004 when サステイナブル was clearly in front. ＫＯＤ追加語彙 has "サステ(イ)ナブル". I'm suggesting it go third as it's only marginally in front in the 2007 n-grams and is not supported by 2016 Googits.

@@ -10,0 +11,3 @@
+<reb>サスティナブル</reb>
+</r_ele>
+<r_ele>

http://tatoeba.org/eng/sentences/show/213534 : "私は「やまる」という読み方は初めて聞いたのですが..."

From the reactions of two of the native-speakers on Tatoeba to this reading of 止まる, I think an "obsc" tag is appropriate.

@@ -14,0 +14,1 @@
+<misc>&obsc;</misc>

(I meant sense 1 of とまる.) Yes, I agree. A native speaker has commented on Tatoeba that they are all とまる. I think I'll just close this off, and if anyone points out that sentence X is better as やまる, I'll change it.

i would guess that most example sentences are とまる.

Yes, at the time the sentences were linked there was only one 止まる entry. It is now split into 3. I have modified the indices; connecting them to either とまる or とどまる. I can't easily work out which should be とまる sense 2 or やまる sense 2, so at present やまる is missing out.
Any ideas?

daijs

@@ -6,0 +6,3 @@
+</k_ele>
+<k_ele>
+<keb>鉄漿</keb>
@@ -15,1 +18,0 @@
-<gloss>black teeth</gloss>

Daijr

koj

not sure it's an exp, esp. with the がみ reading

ルミナス, GG5

G n-grams:
ボカリーズ	698
ヴォカリーズ	13394
ボーカリーズ	332
ヴォーカリーズ	1827

@@ -5,0 +6,6 @@
+</r_ele>
+<r_ele>
+<reb>ヴォーカリーズ</reb>
+</r_ele>
+<r_ele>
+<reb>ボカリーズ</reb>

http://ja.wikipedia.org/wiki/ヴォカリーズ
Daijr, ＫＯＤ追加語彙

仕切価格  might be io

Koj, Eijiro, ルミナス, http://ejje.weblio.jp/content/仕切価格 
http://oshiete.goo.ne.jp/qa/2333818.html

If the katakana is the key form, we shouldn't merge.

koj, wiki, gg5

If the katakana is the key form, we shouldn't merge.

koj, wiki, gg5

possible merge on  鉄漿蜻蛉

@@ -26 +26 @@
-<xref type="see" seq="2182000">半挿・はんぞう・1</xref>
+<xref type="see" seq="2182000">半挿・1</xref>

@@ -21 +21,2 @@
-<gloss>wide-mouthed ceramic vessel having a small hole in its spherical base (into which bamboo was probably inserted to pour liquids)</gloss>
+<misc>&hist;</misc>
+<gloss>wide-mouthed ceramic vessel with a small hole in its spherical base (into which bamboo was probably inserted to pour liquids; from the Kofun period)</gloss>

I'll drop the note. It's a JIS X 0213 kanji.

@@ -21 +20,0 @@
-<s_inf>esp. 𤭯</s_inf>

the 𤭯 kanji in the note is stripped out in wwwjdic/edict2, makes the entry look funny

リーダーズ+プラス, Translators FB page (quoting an advertisement)

400k hits for シーズナル, 4k for シーゾナル (maybe "ik"?)

http://detail.chiebukuro.yahoo.co.jp/qa/question_detail/q1314713768

Seems 下膳 can also be read しものぜん with a quite different meaning.

@@ -15,0 +15,1 @@
+<xref type="see" seq="1689750">配膳・はいぜん</xref>
@@ -16,1 +17,1 @@
-<gloss>Clearing up leftover food and dishes.</gloss>
+<gloss>clearing up leftover food and dishes</gloss>

http://detail.chiebukuro.yahoo.co.jp/qa/question_detail/q1449
418629
http://solmaster.exblog.jp/1528048/

Seen on a sign at work. It was translated as "bussing", but I 
assume that was not intended to be a literal translation.

May as well.

moving to the ...と entry
(do we really nees both?)

@@ -18 +17,0 @@
-<gloss>to use a figure of speech</gloss>

GG5, 新和英中辞典

例えて言うと 7.8M hits, たとえて言うと 2.4M, たとえていうと 240k
I'm surprised (and sceptical).

@@ -4,0 +4,6 @@
+<k_ele>
+<keb>例えて言うと</keb>
+</k_ele>
+<k_ele>
+<keb>たとえて言うと</keb>
+</k_ele>
@@ -11,0 +17,2 @@
+<gloss>so to speak</gloss>
+<gloss>to use a figure of speech</gloss>

http://ejje.weblio.jp/content/たとえていうと  
http://tatoeba.org/eng/sentences/show/207779

http://en.wikipedia.org/wiki/Cyclocomputer

I've always called them "bicycle computer".

@@ -7,0 +7,3 @@
+<r_ele>
+<reb>サイクル・コンピューター</reb>
+</r_ele>
@@ -10,0 +13,2 @@
+<gloss>bicycle computer</gloss>
+<gloss>cycle computer</gloss>

http://ja.wikipedia.org/wiki/サイクルコンピュータ
.
Example:
こちらは、内蔵GPSを搭載した、自転車専用の端末です。サイクルコンピューターのような数値だけ表示するだけの機器とは違って、...