The glyph shown in the evidence is suspicious because no other encoded character has the component ⿱宷日. It is possibly an error for U+2AB3A 𪬺 (K5-01E2). Consider unifying to 𪬺 (U+2AB3A) if cognate.
The supplementary evidence provided by Eiso Chan shows the character to be a variant of 庵. It seems like 𤲅 is a common variant of 奄, and a possible candidate for a UCV.
Oppose unification as non-cognate. This character is used in China, and you have to apply two UCVs to change its shape to the G glyph form (⿱屮王), which is too dissimilar to ⿱山壬.
As 𡆮 (⿴囗士) is only listed in the "備考" section of KXZD, the KX glyph form cannot be considered to be authoritative, and KX should not be used as a source reference. The definitions "土入口也" and "沙土入口" clearly indicate that ⿴囗土 is the correct glyph form for this character. Given that GHZR (which supercedes the GHZ-10711.06 source reference) has corrected the glyph to ⿴囗土, the preferred solution is to correct the G glyph for U+211AE to ⿴囗土 and amend the source reference to GHZR-10766.06. If this solution is accepted then UK will agree to unification and withdraw UK-20835. Horizontal extension is only required if China is unwilling to change the glyph for U+211AE.
The character shown is very unclear because it is so small, but it is obviously intended to be U+704B 灋. If ⿰氵⿸廌土 is not attested in any other source then this can be considered to be a one-off error in this particular telegraph code book. It seems unnecessary to encode this particular error form for this single usage. Therefore suggest unification with 灋 (U+704B).
"If the technically correct (aka semantic) radical for an ideograph hampers its discoverability, or is region-dependent, the primary radical shall be assigned as though made by an ideograph expert who is neither a specialist in the history of the Han script nor familiar with ideograph etymology. The technically correct radical can be assigned as a second radical. Both are shown in the code charts, though the primary one serves as the basis for ordering within a CJK Unified Ideographs block."
As R31 囗 is not the obvious radical, and would hamper the discoverability of this character, the radical should be kept as R187 馬.
I'm not sure what IRG rules you are referring to. However, IRG PnP §2.2.1 d. (5) c) states:
"If the technically correct (aka semantic) radical for an ideograph hampers its discoverability, or is region-dependent, the primary radical shall be assigned as though made by an ideograph expert who is neither a specialist in the history of the Han script nor familiar with ideograph etymology. The technically correct radical can be assigned as a second radical. Both are shown in the code charts, though the primary one serves as the basis for ordering within a CJK Unified Ideographs block."
Therefore, in this case the primary radical should be R126 而.
Isn't there some question about the suitability of 汉字海 as a reliable source? I think it can be used as secondary evidence, but I would prefer to see additional evidence for ⿰木弯 as a place name.
A list of local vulgar simplifications does not seem sufficient evidence for encoding. At the very least I would expect to see evidence of the submitted character in actual textual use.
The evidence is insufficient. Firstly, please show the full page of the evidence. Secondly please show the original source that is being quoted so we can be sure that character shown in the modern source is not a mistake.
It is difficult to trust the modern edition of the text as experience shows that modern editions often introduce glyph errors or create imaginary characters. Therefore, please show an image of the original text, apparently from 《明憲宗純皇帝實錄卷之六十八》.
High resolution details of 順治河南通志 (top) and 雍正河南通志 (bottom):
Shunzhi version shows ⿰虫𦕅, but Yongzheng version shows ⿰虫聊. I suppose that ⿰虫聊 (liáo?) is the correct form as it likely rimes with 螬 cáo, and ⿰虫𦕅 is a contracted form due to the difficulty of writing ⿰虫聊 in the available space.
As 中华字海 and 雍正河南通志 both give ⿰虫聊 I suggest to accept the current glyph and IDS, and return GKJ-00464 to the M-set.
The full reference for the source should be provided (author, title, publisher, year), as well as the name of the author and title for the piece in which this character occurs.
Complete reference for the source (author, title, publisher, year) would be useful. It would also be interesting to see the complete page, and not just a tiny extract.
Seems probable that ⿰犭狂 is an error for 𢓯. In my opinion it is not appropriate to encode an error form on the evidence shown (rime tables are notoriously error-prone), and so suggest to postpone pending additional evidence.
The current evidence is insufficient. Please provide an image from an edition of 禮記 which shows this character. Otherwise it should be postponed pending additional evidence.
Please provide full page of evidence so we can understand the context. What are the characters shown in the evidence meant to represent? They look like a list of rime characters. If so, please provide an image of that shows the literary text where Li Yu uses the character ⿰犭斂.
Also, the character ⿰貝斂 is not encoded and not proposed for encoding. Why propose only ⿰犭斂 and not ⿰貝斂 immediately below it? It makes no sense.
Evidence 2 is suspect. What is the character supposed to mean here? A character with a 'dog' radical makes no obvious sense in this context, and I suspect that it is an error for some other character.
Therefore suggest to postpone pending additional evidence.
Babed on the additional evidence adduced by Huang Junliang, it seems probable that ⿰犭赤 is an error for 捇 in this one particular edition. Therefore suggest to postpone pending additional evidence for this form of the character.
汉字海 would not seem to be an authoritative source for bronze inscriptions. Given that that the evidence shown by Huang Junliang transcribes the character as 荆, additional evidence should be provided before encoding this character.
Additional evidence for ⿰犭麽 is given in WS2021-02373 GKJ-00551 Evidence 1. Evidence 2 shows the variant form ⿰犭摩 which is not encoded or proposed for encoding.
Hand-written evidence alone is not satisfactory, and in this case the submitted character appears to be an error or idiosyncratic variant of 灑. Therefore suggest to postpone pending additional evidence that ⿰氵⿱曲鹿 is a widely-used variant form.
The character is obviously a variant of U+6F09 漉, but the evidence is very insubstantial. It would be really nice to have additional evidence that this variant is not just used once in this one source.
The left side of the character does not entirely look like 牜, and if it is a variant of 犙 it is not clear to me what it means in this context. Additional evidence would be helpful.
Not exactly the same, but we can see a very similar reduction of the 節 component of 癤 in 《番漢合時掌中珠》 where the bamboo radical has been reduced to two strokes (it may even be a printing error for ⿸疒莭):
Based on the additonal evidence produced by Huang Junliang, it would seem that the submitted character is an error for 斸. Therefore suggest to postpone pending additional evidence.
The glyph in the original evidence is unclear, and the evidence produced by Conifer Tseng indicates that the character should be 敘. Therefore suggest to postpone pending additional evidence that this is not a one-off error for 敘.
Evidence image in #2964 is from 《東漢會要》(edition not specified). The text is a quotation from 《後漢書・應劭傳》 which reads "逆臣董卓，荡覆王室，典宪焚燎，靡有孑遗...". Therefore ⿰火秦 here should be a corruption of 燎.
I agree completely with Ken Lunde's comments on TCA-submitted ideographs with insufficient evidence (there is an implicit "agree" comment from me on all of Ken Lunde's repeats of this comment for other TCA characters).
Text in 山東通志 seems to be derived from 《水經注》 which has "又按《管子》:齊桓公二十年,征孤竹,未至卑耳之溪十里,闟然止,瞠然視...". It seems that ⿵門离 is an erro form of 闟, therefore suggest to postpone pending additional evidence that this is not a one-off error in this one particular edition of 山東通志.
Google search finds mostly "冬積柴水中為霖以取之", and Wikisource has "冬積柴水中為罧以取之", so it seems likely that ⿰氵⿱㓁林 is a corrupt form in this edition. Suggest to postpone pending additional evidence that ⿰氵⿱㓁林 is correct or is a stable error.
From context the character should be a variant or error for U+6548 效. Suggest to postpone pending additional evidence that this is not a one-off error in this one source.
This place name is given as 米{⿰米犀}橋 with the note 昔有富人積米{⿰米屖}建此故名. {⿰米犀} and {⿰米屖} should be the same character, and the way the note is written, it should be a common character that does not need an explanation. The two characters could be corruptions or variants of U+7CCF 糏 'rice grits left after hulling' which would fit the sense here. Regardless, additional evidence is needed to determine which of {⿰米犀} and {⿰米屖} is the correct glyph, so suggest to postpone for additional evidence.
Based on the comment by Huang Junliang, it seems possible that the submitted character is an error form for 㶖. It may be prudent to postpone pending additional evidence that this is not a one-off error in this single source.
It would be helpful to see additional evidence. Google Books shows that this text is recorded in 《地方志人物傳記資料叢刊: 華北卷》 (2002) vol. 27 p. 911 and 《中國地方志集成: 河北府縣志輯》 (2006) vol. 45 p. 557, but I do not have access to these.
《武強縣新志》 appears to have 漁, although I cannot confirm this.
Evidence
Based on #7499, ⿰氵𠅤 is probably an error for U+6F01 漁, therefore suggest to postpone pending additional evidence that this is not an error form in this one source.
The glyph is not very clear in the provided evidence, and I suspect that it may be a corrupt form of another character. Is there any additional evidence to support the exustence of ⿰火規 ?
The character ⿱暗灬 seems unlikely in the context, and is probably a mistake for some other character. Google books finds "只為一人長期翠裙羅费似烟" in 《天津圖書館孤本祕籍叢書: 史部》 but does not show me the actual text, and "费" does not seem plausible either. However, it does indicate that there is a source for the text shown in the evidence, so I think we need to find other examples of the text in order to confirm what the actual character is. Therefore suggest to postpone pending addditional evidence.
New evidence
啽囈集 confirms that ⿱暗灬 in the first evidence is indeed a variant of 黯:
This and the additional evidence provided by Eiso Chan indicates that it is OK to keep UK-20132 in the M-set.
It should be possible to find additional evidence for this person's name. Google Books suggests that he is mentioned in 《地方志人物傳記資料叢刊》 (華北卷) Vol. 21 p. 747, but I cannot access it.
This is a list of Song dynasty officials in a Qing dynasty book. There is a high probability that the character ⿱直木 is a mistake. 《景定建康志》卷二十四 gives the name as 趙汝末. Suggest to postpone pending additional evidence for the correct form of this person's name.
It is not clear to me what ⿰末攵 should mean here, and it seems quite possible that it is a one-off error for some other character (maybe 救). It would seem prudent to postpone pending additional evidence to confirm that this is not an error character.
The evidence seems insufficient. The character is quite possibly a mistake for some other encoded character, and the single occurence of this name in a note is not sufficient to determine whether this is a genuine character or an error character. Suggest to postpone pending additional evidence.
Other versions of this text give "急蘭亦䚟" (e.g. 《元史》卷二百一十), so ⿰月㝵 is likely an error for 䚟. Therefore suggest to postpone pending additional evidence that this is not a one-off error in this edition.
Other editions have "其妙聲，則清靜厭瘱", so the character here (which is not very clear anyway) should be an error/corruption for 瘱. Therefore suggest to postpone pending additional evidence.
Evidence
The additional evidence shown by Huang Junliang is much clearer. If it is an error then it can be considered a stable error.
I agree that ⿰豊豊 corresponds to 艷體 in the form of the poem quoted in #7485. However, here we have "{⿰豊豊}飡瓊瑤華" which preserves the five-character poetic metre, so ⿰豊豊 cannot be an accidental mistranscription (I think it is impossible to accidentally mistranscribe 艷體 as ⿰豊豊). It is possible that "{⿰豊豊}飡瓊瑤華" is the original form of the poem, and "艷體餐瑤華" is a revised version which expands the unusual character ⿰豊豊. But even if "艷體餐瑤華" is the original version, I believe that "{⿰豊豊}飡瓊瑤華" must have been a deliberate revision. Therefore, I think the evidence for ⿰豊豊 is satisfactory, and the character should be kept in the M-set.
Agree that the evidence shown is insufficient. I have been unable to find other examples of this character, therefore suggest to postpone pending additional evidence.
Incorporating a non-CJKUI element such as ☷ is not a valid reason not to encode this character. There are other CJK unified ideographs which include elements from other scripts; in WS2021 we have Hanja-Hangul hybrid characters with ᆨ such as KC-00811, KC-01727, KC-02286, KC-03983, KC-03520; and characters derived from Latin script symbols such as UTC-03225 (derived from ℔ pound sign), UTC-03226 (氵+ ʒ = fluidram), and UTC-03227 (氵+ ℥ = fluidounce).
I agree with the comment by Wang Yifan. Based on the reading and definition given in the evidence, this should be an error for U+9713 霓. Therefore we WITHDRAW this character.
The correct name of this Song dynasty person is 林瞪 (1003-1059), therefore ⿰日登 is certainly an error for 瞪. If this is a one-off error in this source then the character should not be encoded. Therefore suggest to postpone pending additional evidence.
Confirmation that the correct form of the character is ⿰口⿱刀臼 (unifiable with U+5557 啗).
Source: Bernard F. Meyer and Theodore F. Wempe, "The Student's Cantonese-English Dictionary" (3rd ed., Hong Kong, 1947) p. 626 (taâm)
It looks like someone mistakenly analysed 蒙 as the simplified form of ⿱𤇾⿱一豕, analagous to 萦~縈 etc. Is the ⿱𤇾⿱一豕 form used in other sources? It would be prudent to see actual textual examples of this character before encoding it, especially as the evidence page also shows this form of 蒙 with 月, 氵, 虫, and 木 radicals which I suppose would be submitted for a future working set.
The glyph does not conform to PRC conventions for 毛 on the left side (final stroke should be hooked up). Suggest to modify the glyph to conform to PRC conventions.
The 2.0 glyph is not an improvement. The top of the vertical stroke of 巾 does not conform to PRC conventions, and the left shoulder of 巾 should not protrude. Cf. the G-source glyph for U+4F48:
The 2.0 glyph is not an improvement, as it does not match the evidence or conform to PRC conventions for 礻 (should not show the final stroke protruding on the left). Therefore suggest reverting to the 1.0 glyph.
We could consider changing the Glyph and IDS to ⿱⿸尸皮肉, but I think that based on the first example shown in the evidence that the current glyph and IDS are acceptable.
The UTC glyph uses the G form of 跋. The H form of 跋 is as shown in the evidence. Therefore suggest changing the glyph to match the evidence (i.e. use expected H glyph form).
The form of 灵 used for almost all V-source characters is the same as the G-form (i.e. no protruding horizontal stroke). Only U+306FB (Ext. G) and U+31ADF (Ext. H) use the same form of 灵 as shown for this character (VN-F1A08). It would be nice if Vietnam could use a consistent form of 灵 for all characters.
Harmonization of radicals is a common phenomenon, so maybe it is incorrect to call this a "misprint". In any case, I support keeping this character based on the new evidence.
GKJ-00586 should not be "postponed for evidence" but rejected or withdrawn. Any character submitted without evidence at the time of submission (unless due to a technical error) should be rejected as not meeting the required IRG standards for submission.
⿰犭⿱𠂉奇 is not encoded or proposed for encoding, but it is recorded as a family name in 《中华姓氏源流大辞典》 and 《中华千家姓氏录》 so it could be a candidate for inclusion in a future submission.
Evidence 1 is likely to show a PUA font error where a PUA character for U+2B9EF 𫧯 was intended, but a different font was used to print the book which had ⿰饣善 for that PUA code point.
The new evidence for ⿰饣善 as a derived simplification for 饍 is sufficient, although additional evidence from China would be welcome.
Probably the same as 苦聪, but 夷人圖說目錄 has separate descriptions for "古宗" people (no. 7) and "苦葱" people (no. 30). The former is written as "狜猔" in 《滇省迤西迤南夷人圖説》.
Evidence 2 shows ⿰卩⿱⿳⿰一一⿰一一⿰一一皿 = 隘 on the same page (also written as ⿰卩⿱⿳一一一皿 in the small character text), so I think we need additional evidence to demonstrate whether the middle horizontal stroke should be broken or not.
No alternative encoding model has been proposed, and in my opinion there is no good reason not to encode this single character as a CJK unified ideograph. Therefore suggest moving it back to the M-set.
The following character (⿰氵𥬡) is WS2021-02079 UK-20208. The fact that both characters in his retirement sobriquet are unencoded suggests that he deliberately created two novel characters.
The preceding character (⿱興同) is WS2021-00745 UK-20207. The fact that both characters in his retirement sobriquet are unencoded suggests that he deliberately created two novel characters.
See this SCMP article for additional background discussion of the "X也" gender-neutral 3rd-person pronoun. Refusing to encode this character would be doing a great disservice to the user community who want to use this gender-neutral pronoun form.
Comment
Additional discussion on twitter showing a glyph form similar to that proposed by the UK with 㐅 rather than X:
局 is first attested with the meaning of 'to steam' or 'to bake' in Robert Morrison's 1828 《廣東省土話字彙》; and first attested as 焗 in 《粵語全書》 (1905). UK-20862 certainly seems to be the same meaning and Cantonese pronunciation as 局 in this sense. The ⿸尸可 form shown in the evidence may be a mistake for 局, but potentially it is a specific variant of the character used only for the culinary meaning, in which case it should not be unified. Therefore suggest to postpone pending additional evidence.
U+311A7 𱆧
Evidence provided for G_Z3862201 indicates that U+311A7 is a Zhuang word meaning 'pond loach' (泥鰍). UK-20804 also means 'pond loach', but in the Minnan language, which is a strange coincidence as Zhuang and Minnan are unrelated languages.
IRG Working Set 2021v3.0
Source: Andrew WEST
Date: Generated on 2026-01-19
Unification
Appears to be an error for U+25855 𥡕 in a single edition. Suggest postponing for additional evidence.
The glyph shown in the evidence is suspicious because no other encoded character has the component ⿱宷日. It is possibly an error for U+2AB3A 𪬺 (K5-01E2). Consider unifying to 𪬺 (U+2AB3A) if cognate.
As 𡆮 (⿴囗士) is only listed in the "備考" section of KXZD, the KX glyph form cannot be considered to be authoritative, and KX should not be used as a source reference. The definitions "土入口也" and "沙土入口" clearly indicate that ⿴囗土 is the correct glyph form for this character. Given that GHZR (which supercedes the GHZ-10711.06 source reference) has corrected the glyph to ⿴囗土, the preferred solution is to correct the G glyph for U+211AE to ⿴囗土 and amend the source reference to GHZR-10766.06. If this solution is accepted then UK will agree to unification and withdraw UK-20835. Horizontal extension is only required if China is unwilling to change the glyph for U+211AE.
Agree to unify to 𬝋 (U+2C74B).
The character shown is very unclear because it is so small, but it is obviously intended to be U+704B 灋. If ⿰氵⿸廌土 is not attested in any other source then this can be considered to be a one-off error in this particular telegraph code book. It seems unnecessary to encode this particular error form for this single usage. Therefore suggest unification with 灋 (U+704B).
Attributes
"If the technically correct (aka semantic) radical for an ideograph hampers its discoverability, or is region-dependent, the primary radical shall be assigned as though made by an ideograph expert who is neither a specialist in the history of the Han script nor familiar with ideograph etymology. The technically correct radical can be assigned as a second radical. Both are shown in the code charts, though the primary one serves as the basis for ordering within a CJK Unified Ideographs block."
As R31 囗 is not the obvious radical, and would hamper the discoverability of this character, the radical should be kept as R187 馬.
"If the technically correct (aka semantic) radical for an ideograph hampers its discoverability, or is region-dependent, the primary radical shall be assigned as though made by an ideograph expert who is neither a specialist in the history of the Han script nor familiar with ideograph etymology. The technically correct radical can be assigned as a second radical. Both are shown in the code charts, though the primary one serves as the basis for ordering within a CJK Unified Ideographs block."
Therefore, in this case the primary radical should be R126 而.
Evidence
Shunzhi version shows ⿰虫𦕅, but Yongzheng version shows ⿰虫聊. I suppose that ⿰虫聊 (liáo?) is the correct form as it likely rimes with 螬 cáo, and ⿰虫𦕅 is a contracted form due to the difficulty of writing ⿰虫聊 in the available space.
As 中华字海 and 雍正河南通志 both give ⿰虫聊 I suggest to accept the current glyph and IDS, and return GKJ-00464 to the M-set.
GKJ-00538 is clearly an error for 狂, so suggest China withdraw it.
Also, the character ⿰貝斂 is not encoded and not proposed for encoding. Why propose only ⿰犭斂 and not ⿰貝斂 immediately below it? It makes no sense.
Evidence 2 is suspect. What is the character supposed to mean here? A character with a 'dog' radical makes no obvious sense in this context, and I suspect that it is an error for some other character.
Therefore suggest to postpone pending additional evidence.
永濟縣志（清光緒刊本）卷3 folio 42a
This shows the correct character to be U+2402B 𤀫. Therefore suggest to postpone pending additional evidence that ⿰氵辦 is correct or is a stable error.
This and the additional evidence provided by Eiso Chan indicates that it is OK to keep UK-20132 in the M-set.
Therefore ⿰氵追 is a variant of or error for U+6425 搥.
Source: Bernard F. Meyer and Theodore F. Wempe, "The Student's Cantonese-English Dictionary" (3rd ed., Hong Kong, 1947) p. 626 (taâm)
Cf. WS2021-01385 which writes the same word as {⿰忄盟}𢛵.
I suggest to postpone pending additional evidence.
Glyph Design & Normalization
Other
⿰犭⿱𠂉奇 is not encoded or proposed for encoding, but it is recorded as a family name in 《中华姓氏源流大辞典》 and 《中华千家姓氏录》 so it could be a candidate for inclusion in a future submission.
The new evidence for ⿰饣善 as a derived simplification for 饍 is sufficient, although additional evidence from China would be welcome.
Data for Unihan
Evidence provided for G_Z3862201 indicates that U+311A7 is a Zhuang word meaning 'pond loach' (泥鰍). UK-20804 also means 'pond loach', but in the Minnan language, which is a strange coincidence as Zhuang and Minnan are unrelated languages.