Misprint of 𣝼. 《漢語大字典》pp. 1404 quotes《說文》：𣝼，屋𣝼䏈也。

https://homeinmists.ilotus.org/hd/hydzd3.php?st=page_no&kw=1404

The followings are eight editions of 《集韻》 at the same entry. At the same position, some are the submitted one (red), some are ⿰木⿱鳥𠔿 (blue) and 𣝼 U+2377C (orange), which mean the same character. ⿰木⿱鳥𠔿 could be unified to this character.



▲ 《集韻校本》, 卷之一


▲ 方成珪: 《集韻考正》, 光緒刻本, 卷之一



▲ 《集韻》, 新興書局影四部備要本, 卷之一


▲ 《集韻》, 文淵閣四庫本, 卷一


▲ 《集韻》, 日本天保九年重刊顧廣圻補刻本, 卷一


▲ 《集韻》, 曹氏楝亭本, 卷一


▲ 《集韻》, 寧波明州述古堂影宋鈔本, 卷一


▲ 《集韻》, 潭州宋刻本, 卷一

Unify to 𣝼 U+2377C as per comment #9519.

Add new UCV ⿳自㓁𠔽 ~ ⿱鳥𠔿 ~ ⿱鳥囚.

The stable variants could be encoded.

Agree with Tao Yang. This is a stable variant and is not a misprint.

Agree unify to 𣝼 (U+2377C).
Evidences all point to Shuowen, can apply liding unification rule.

The proposed "liding" rule has yet to be discussed during a meeting, let alone accepted. A central question to be debated is whether or not it is possible to have a rule that just applies to shouwen, and if that was the case does it not mean that the decision to unify is ad hoc and therefore should not result in a new UCV since if the a character of exactly the same shape but unconnected to shouwen was submitted then it would not be unified.

1. “Misprint” should be carefully used.
That means the error originated from reasons such as printing machine malfunction, insufficient printing technology, and poor paper quality.
However, for ancient Chinese books, all characters are handwritten, engraved, and printed, and there are very few printing quality issues with the literature used as evidences, and there are no errors could be named as “misprint”.
2. "The rule of Liding" exceeds our authority.
Up to now, the study of ancient Chinese characters has been passed down for over 2000 years, and there has not yet been a unified set of Liding rules, making it impossible for us amateurs to provide professional advice.
Our responsibility is to ensure that the glyph of Kai style characters is stable, reliable, and can be encoded, and should not involve professional work in neighboring fields.

The liding unification rule does not involve any authoritative discourse, but only the identification of inter-character relationship, which is in the scope of IRG work.

Determining that two Kaishu actual shapes correspond to the same ancient form, as suggested by IRGN2612, limits the determination to a relatively small scope. This unification rule would not be implicated if it were not special document (e.g., evidence or repertoire declaring the source of ancient forms), or special content (e.g., evidences here which are discussing character cognition directly referring to Shuowen or pre-Qin materials).

Disagree with unification. Unification in this case would be inconsistent with established unification rules, clearly a case of different components therefore abstract shape is different. The abstract shape of a character must be determined by it's regular, aka kaishu, shape it can be determined by the shape of a character in another script.

Reply to #14385,

Sorry. “Clearly a case of different components therefore abstract shape is different” may not be clear itself.
In current UCV list, one may find many so-called “different components” that correspond to same abstract shape, such as |壬| |王| (#1), |天| |夭| (#8), |戌| |戍| (#49), |凡| |卂| (#55), |圣| |𢀖| (#87), |屮| |山| (#96), and so on. So only when “different component” means “different abstract component”, the sentence is correct.
In this case, |囚| and |𠔽| are indeed different actual shape, but according to evidence, they belong to the same abstract shape [𠔽] that is part of [𦤝].

Reply #14398 Yes, here saying "different abstract components" would be clearer than "different components". In the sentence above, that the components are abstract is implied by context . In this case it seems we both agree the abstract shape of the upper part right hand components have different abstract shape and therefore that the two characters have different abstract shapes by usual Annex S and UCS usage of the words. The "liding" rule uses different criteria. We can say that by Annex S conventions the two characters have different abstract shape therefore by that convention should not be unifie, but that the "liding" rule to say they can be unified.

1. It's not the proper way to add new UCV
If we want add a set of new UCV, the font should be commonly used, directly related, and with minimal variation. It's feasible to add ⿱鳥囚~⿱鳥𠔿~⿱鳥𠔽 as a new UCV. The right part ⿱鳥囚 is already totally different with ⿳自㓁𠔽, it does not comply with our previous principle of adding new combinations.
2. Representative variant characters should be independently encoded as unification targets
Other similar glyphs of ⿱鳥囚 are also collected during the project process. We chose one of them as the representative form and demonstrated its difference from 𣝼.
The encoding work is to find out valuable glyphs, rather than eliminate all variants. We should ensure the scale of unification to prevent excessive cultural loss.

3. This is a issure about Erroneous Transformation rather than Liding Unification Rule.
This character is generated during version transfer, not Liding from Small Seal Script directly. This is a question about variant characters, which has nothing to do with the rules of Liding.
4. IRGN2612 is not yet a rule that has been discussed and approved.
All the features of this character comply with the relevant regulations of PNP2.1, while IRGN2612 hasn't entered the discussion agenda, we don't need to take into account the imagination that has not yet become a reality.

As comment #14400 point 3 illustrates, if there was a 'liding' rule, or an IRGN2612 might be called, the result would be long debates about whether or a character was IRGN2612 enough to be applied to.

Oppose unification to 𣝼 (U+2377C) as there is no liding rule, and a new UCV for ⿳自㓁𠔽 ~ ⿱鳥𠔿 ~ ⿱鳥囚 is not reasonable. Sufficient evidence has been provided that this is a stable variant which should be encoded.

I still disagree to unify this character with 𣝼 (U+2377C). The current evidence is sufficient to encode it as a variant. Other editions used other forms, which does not affect the separate encoding of this form.