I think that IRG experts had agreed that captions couldn't be the only source of the evidences for submitted ideographs in IRG meeting #62. So unless other evidences can be provided, the ideograph should be postponed.
The decision about using captions as evidences was clearly stated in the meeting so this kind of situation should not have happened.

It is not the matter of the defination of caption or lyrics. It is the matter of the quality of the evidences. It is ridiculous to dicuss the defination of caption here but focus on the quality of the evidences.
Evidence of GDM-00507 and GDM-00508 are from at least two different buildings, which stands in the real world. The buildings are not something easy to change or vanish. What's more, the two ideographs are used by many local people so they can be used in the plaques of the temples, which are sacred.
However, the evidence of this ideograph is from a vedio created by someone on the Internet and the vedio can be edited or deleted by the uploader at anytime he wants. The vedio, which is too weak for encoding, is not even from a published material.
https://www.bilibili.com/video/BV1Ki4y127Cm/
If this can be accepted as evidence, then we may be going to submit all this to IRG, there are even pronounciations and definations:
https://www.bilibili.com/video/BV198411s7Ft/


Moreover, I don't think IRG have to write every this kind of unstable thing, for example, captions, lyrics, articles, instructions, notes... in PnP, which is unnecessary and endless.

The two characters in comment #629 are nonce characters and as such not suitable for encoding based on such evidence. The character UK-30621 is not a nonce character and so the evidences are suitable for submissions. In the same way that the existent of some nonce characters in printed materials does not prove that all printed evidences are unreliable, the existent of some nonce characters in non-printed materials does not prove that all non-printed evidences are unreliable.

I think nonce or not should be proved by undoubted evidences. I'd like to help the submitter to find evidences meet IRG requirements but the current evidence can not prove that ⿰久闹 is differernt from ⿱因八 or ⿱中分 in "nonce or not".
It should be noted that our center proposed a document "Application for encoding some ideographs used in Chinese geographical names(IRGN2649)" to IRG before, which was pointed out by an expert that it is not suitable as the only evidence for encoding. Our center is a formal institution established by Sichuan International Studies University, which is belonging to The People's Government of Chongqing Municipality(重庆市人民政府). It will be very offensive and so unacceptable if videos on the internet are considered more trustable or suitable for encoding than an application with our seal on it.

I will post next week evidence that ⿰久闹 is not a nonce character, however it should be noted that even without more those familiar with both Chinese and Zhuang can see a difference. Requesting evidence to show that this character is not a nonce character is not a problem.

Thanks for that. I will check my books these days, too.

As far as we are concerned, actual evidences are more credible than expert's experiences. Expert's experiences are very helpful when qualified evidences are provided. But the experiences can also be harmful if they are over relied. Thus although we have many excellent experts here, qualified evidences are still needed for this character(UK-30621), ⿰大老(UK-30639) and ⿰丫要(UK-30620).

If there are other subbmitted ideographs whose evidences are only from online video, we think that they should be postponed too if no more qualified evidence can be provided.

From 《基于 W indowsIME 古壮文输入法编辑器的设计与实现》published June 2013 《现代计算机》

The new evidence still shows no running text but only a screenshot of a computer font.

What the new evidence shows is that the character is not a nonce character. The first evidence is from 2022 and the second from 2013. The question is not what does each evidence show on their own but rather what they show together.

Screenshots of computer fonts and vedios from the internet are absolutely not qualified evidence for IRG. If no other evidence can be provided, then this ideograph should be postponed.

The latest IRG PnP (version 17) explicitly allows for multimedia evidences and submissions to this working set include among other things internet evidences from Instagram, Twitter and Bilibili. It is inconsistent to object to evidences from Bilibili just because they are from Bilibili, but not to object to evidences from Instagram just because they are from Instagram, nor object to to evidences from Twitter just because they are from Twitter.

IRG PnP Version 17, page 11-12
Currently, IRG mainly accepts evidence from printed material if they are accepted as IRG sources.
In general, IRG DOES NOT accept multimedia material as IRG sources.
Note: the acceptance of the multimedia material, the popularity of the material, cultural influences, and other factors that warrants its acceptance.
We can't find a sentence in IRG PnP states that being posted on Instagram, Twitter or Bilibili once by any uploader will warrant the evidence's acceptance.

Furthermore, the screenshots of computer fonts prove nothing but the font producer has made the font. This cannot prove the shape is actually used in texts or even exists. As far as we know, the uploader of the vedio use ⿰久闹 just because he saw the font in a friend's computer without knowing the pronounciation or meaning.

I'd like to point out that using these as evidences is against UK's general requirements for the quality of evidences. I really don't think other experts will accept these two images as qualified evidences even if I were persuaded. So please find qualified evidences for the ideograph or postpone it.

You have made a mistake reading and quoting PnP v17 that changes the meaning. You have combined the start of footnote 13 at the end of page 11 with the second part of section 2.2.5b(7) at the start of page 12. Being a footnote the second part of footnote 13 is to be found at the bottom of page 12 not the top. Footnote 13 reads, "Currently, IRG mainly accepts evidence from printed material if they are accepted as IRG sources. In general, IRG does not accept multimedia material as IRG sources. Note: the acceptance of evidence from captions and subtitles may warrant the acceptance of some multimedial materials as IRG source in the future."







Footnote 13 when read correctly clearly says that the IRG may accept some multimedia sources, hence whether Bilibili evidence from the UK or Instagram and Twitter evidence from the UTC these conform to current PnP and may be accepted by the IRG. It is therefore incorrect to say that video evidence is not acceptable to the IRG.

A key to understanding evidence is to look at it carefully. In evidence 1 the information is in an interlinear format, the line below each character gives the pronunciation and the line below that gives the meaning in Chinese. Also it is colour code so the lyrics are in red and the pronunciation and meaning in black. Therefore it is clear that the uploader understands the meaning and pronunciation.

It should of course go almost without mention that the evidence conforms to the requirements of the UK.

Thank John for pointing that out and I am sorry that I made the mistake. But I still can't find a sentence in IRG PnP states that being posted on Instagram, Twitter or Bilibili once by any uploader will warrant the evidence's acceptance of IRG.
Comment #2304 says:"Also it is colour code so the lyrics are in red and the pronunciation and meaning in black. Therefore it is clear that the uploader understands the meaning and pronunciation."
I think it is obviously wrong. Logically, I can use 鹿 with pronunciation mǎ and meaning 马 in my vedio. It will be very ridiculous to say that 鹿 pronounciates mǎ and means 马 just based on my vedio. The paired pronunciation and meaning in the vedio proves nothing but only the uploader used ⿰久闹 with that pronunciation and meaning in the vedio. This fact warrants nothing.
I'd like to say that I am kind of sure that the uploader didn't know the pronunciation or meaning before using it. So please find qualified evidences for the ideograph or postpone it as experts will suggest in IRG meeings.

Comment #2304 also says:"It should of course go almost without mention that the evidence conforms to the requirements of the UK."
Comparing the evidences for this ideograph with the evidences for most of other ideographs, we still think that the evidences for this ideograph is against UK's general requirements for the quality of evidences. It would be very worrying if the quality of them were the same.

IRG PnP clearly states that they may accept some multimedia sources there is no need for the PnP to mention specific names. such as Instagram, Twitter or Bilibili nor would it be productive to do so. It is in many respects the content not the format that determines the significance of certain evidences.

When a person says what they think is the pronunciation and meaning of a character it gives information about much the person understands about a character. If a video had 鹿 is pronounced mǎ and means 马 it would not be suitable to use as evidence because either the producer of the video does not understand the character 鹿, or had mistyped the wrong character by mistake or was making a joke. By comparison saying the character ⿰久闹 has the pronunciation naus aka nauq and means 永远 shows the writer understands the character in the same way that saying 妈 has the pronunciation ma1 aka mā and means mother would show the writer understands the character 妈.

Evidence 1 has many strengths:
- it is a primary source of evidence
- it shows the character is used in running text
- it shows clearly the shape of the character
- it accurately gives the pronunciation and meaning of the character

The second evidence:
- confirms the shape of the character
- shows the pre existence of the character
- shows that multiple fonts contain the character (the font used for the video is not that shown in the computing article)

Furthermore since the up-loader of the video in 2022 was around 20 together they show the character is stable, that the character has already passed on to the newest generation, which is significant.

It should be noted that many more multimedia items where considered and that only those of good quality were used.

In comment #2798, it says "Furthermore since the up-loader of the video in 2022 was around 20 together". I think the submitter should provide the screenshot of them all to prove that this is true but not provide comments in texts only. Still, it is so clear that the quality of internet video uploaded by random uploaders is too weak for encoding.

Search result of 172画 huang in Bilibili
Should we encode huang? The number of the uploaders of huang is far bigger than 20.


Comment #2798 says:
Evidence 1 has many strengths:
- it is a primary source of evidence
- it shows the character is used in running text
- it shows clearly the shape of the character
- it accurately gives the pronunciation and meaning of the character
The second evidence:
- confirms the shape of the character
- shows the pre existence of the character
- shows that multiple fonts contain the character (the font used for the video is not that shown in the computing article)

However, even evidence 1 itself is suspicious, how can we assure the information in it is correct?
The second evidence is also too weak for encoding. In the process of making fonts for ideographs used in books, many errors can be found. Since both of the evidences are not qualified for encoding, these two evidences cannot be used to prove anything else.

As requested screenshots to confirm the video was posted in 2022 and at that time the poster was around 20.

Screenshot 1 showing the date of posting of the video as 19th February 2022



Screenshot 2 of video posted by the same person on their 21st birthday on 2nd June 2022 (face deleted) 【壮语】情感语录—致我第21个破蛋纪念日 https://www.bilibili.com/video/BV13Y4y1576E/?spm_id_from=333.999.0.0



The 41 second video with sound track in Zhuang and including Chinese subtitles for those who don't know Zhuang includes a number of photos of the poster as they grow up including photos from their 16th and 21st birthdays.

Here the question is what to do about ⿰久闹 and ws2024.

One way to confirm the ⿰久闹 character is by noting the sound comes from 闹 and the meaning from 久. These are obvious to anyone who knows the language.

The sum of the parts can be greater than the whole, the question is not what the evidences show separately but what they show together.

When you say, "evidence 1 itself is suspicious" I think you may well be in a minority of one, and that the vast majority of people would say the video is genuine.

Sorry about the request. I thought that there are 20 people who use ⿰久闹. If 20 is the video that the uploader uploaded, then it cannot prove ⿰久闹 is valid to any extend.
Anyway, it will be too ridiculous for me to believe that vast majority of IRG experts will support encoding ⿰久闹 in the current situation.
Although I am not angry about the personal attack in Comment #2880 at all, but I still hope that there won't be any more.

20 was the age in years of the person when they posted the video used for evidence 1.

To say that the video is genuine is to say that the video is what it purports to be.

Comment #2880 was not intended as a personal attack. I will endeavour to avoid making comments that could be viewed as a personal attack.

Clarifications:

In comment #2304 'the evidence' in the sentence "It should of course go almost without mention that the evidence conforms to the requirements of the UK" refers to Evidence 1 from 2022 bilibili video 【壮语歌曲】《如礼金万》uploaded submitted by the UK.

The second evidence was provided by an individual expert, a page from a 2013 article of the computing journal《现代计算机》about a Zhuang character IME. That the page contains a screenshot from the PUA part of a font should not be taken to imply that the individual expert thinks that characters only found in such fonts should be encoded.

From the Zhuang part of a multilingual version of "Only My Railgun"



1:25 minutes



1:31 minutes

(whilst the video https://www.bilibili.com/video/BV1ut421a7Eb/ has 60k views it is strictly for fans of the anime television series 'A Certain Scientific Railgun')

壮语歌曲《承诺》[Zhuang song "A Promise"] uploaded 2024-11-17 to https://www.bilibili.com/video/BV1vZU6YKEjR/

Same kind of evidence like before. The uploader is also the same as the existing evidence. More convincing evidence needed.

The UK still has not been able to provide evidence from historical document or other high-quality sources for this character. On the contrary, the UK has been providing evidence of the use of this character by one or two users on unstable carriers, and such evidence is not sufficient to support the encoding of this character. We have made it very clear that the evidence of this type provided by the UK is not sufficient to support the encoding of this character. However, the experts from the UK seem unable to understand what we mean and they are still insisting on providing evidence of the same type with insufficient quality, and also insisting on saying things that are obviously inconsistent with the facts. This kind of behavior is of no benefit to the review work and cannot make these characters be removed to the M-set.
At the last meeting, we have already suggested that if the experts from the UK believe that this character is not a newly self-created character and has the value for encoding, they can simply publish a paper that includes this character and vouch for these characters with their own reputation. Let me be more straightforward. It should not be a difficult task for the experts from the UK to publish a paper. If the experts from the UK are not willing to risk their own reputation, they should not insist on using evidence that obviously does not meet the requirements to request the encoding of this character in the IRG.

英国（UK）至今仍未能提供这个字的历史文献证据或其他高质量来源，相反，英国一直在提供由一两个使用者在不稳定的载体上使用该字的证据，这些证据并不足以支持对该字进行编码。我们已经说得很清楚，英国提供的这种类型的证据不足以支持编码该字，但英国的专家似乎无法理解我们表达的意思，仍在坚持提供相同类型、效力不足的证据，坚持说一些明显与事实不符的话。这种行为对审核工作没有任何益处，也无法让这些字从 D-set 返回 M-set。
在上次会议上，我们已经建议：如果英国专家认为这个字并非个人新造，且有编码的价值，完全可以自己发表一篇包含此字的论文，以自己的声誉给这些字作担保。让我把话说的再直接一点：发表一篇论文对英国专家来说应该不是什么难事，如果英国专家不愿意拿自己的声誉冒险，就不应该坚持在 IRG 以明显不符合要求的证据来要求对该字进行编码。

The evidence provided for UK-30621, ⿰久闹, conforms to the requirements of the current PnP which permits a wider range of evidences than the PnP for earlier working sets, this was done specifically to allow consideration of a small number of characters not having more 'conventional' evidences, it is therefore not sufficient to argue that a character should be kept out of M-set simply because it does not have the type of evidences used in earlier working sets. According to current PnP 'the popularity of the material, cultural influences, and other factors' may warrant the acceptance of multimedia material as sufficient evidence for encoding. That the evidence for a character is that it is used by ordinary people for ordinary things is sufficient.

It should be noted that there are differences in the number and types of evidence for the 3 characters and therefore they should be considered on a case by case basis.

The evidence for UK-30621 comes from multiple sources and from multiple authors. The evidence in #8709 comes from the same author as evidence 1 demonstrating that use in the earlier evidence was not made by mistake or out of ignorance. The other 2 evidences come from other people. All the evidences given are verifiable .

The character is culturally significant in that it is the only known character used exclusively for nauq [forever].

The evidence for ⿰久闹 is clear there is no doubt about the glyph shape, the character clearly preserves an ancient Zhuang word, nauq [forever], already lost in some places where it has been replaced by a loanword. The evidences show there is a need for us to encode the character.

China thinks the evidence is still not sufficient and the character should be kept in D-set until better evidednce is provided.

There are still a few issues that need to be discussed:
1.According to PnP regulations, when a character is submitted for encoding, it must first comply with the authority of evidence: "Original Source: The source of evidence must be considered authoritative by IRG, as validated by past literature and IRG experts. IRG has the right to reject characters from questionable sources”.
2. Obviously, so far, IRG has always believed that literature evidence is the most authoritative source of character encoding. If there is no original literature, important geographical indications, identity documents, etc. should be considered authoritative evidences.
3.Although the submitted evidence includes photos/images/pictures, the authority clearly varies greatly. The photos taken for historical buildings, whose plaque names have been place names for hundreds of years, inevitably have the authority of place name characters. However, screenshots of online videos cannot verify the historical heritage of their text, nor can they prove the actual source of their text, and do not have convincing authority.
4. The subtitles in the video can be added arbitrarily, and the glyph can be designed by the creator. How can we prove that the glyph is not a personal or small-scale design?
5. PnP claims that "the font used for encoding submissions should provide multiple sources of evidence as much as possible." Of course, multimedia evidence will not be rejected, but it cannot be considered that the authority of the evidence can be abandoned, and multimedia evidence that has not been widely disseminated and recognized should not be considered as evidence that meets the authoritative requirements.
6.This character has already coded in 《古壮字的字符码位表》in 2013, that means the documentary evidences do exist. It's better to submit some pictures of books to prove the actual usage in paper document.

China, as a member body of WG2, disagreed to let the character go back to M-set because the current evidence is not sufficient. It is astonishing that this should be neglected so easily in the last day's meeting.
It is also astonishing that the three characters were added back to the M-set even if the evidence shows that the characters (especially the other two, ⿰丫要 and ⿰大老) are used only by one or two ordinary people in unstable internet vedios, and no one have seen the characters were used in historical document. Are these vedios have been considered authoritative evidence by IRG? Is this the way to ensure the quality of the standard is great? I really cannot understand it.
I just want to say that if the three characters (i.e. ⿰久闹, ⿰丫要 and ⿰大老) are added back to M-set based on the current evidence, our center will draft official documents to the Guangxi University(广西大学) to verify the origin of ⿰久闹 in the font and if its experts' comments were right. What's more, we will also draft official documents to the Ministry of Industry and Information Technology of the People's Republic of China(中华人民共和国工信部) stating the situation here.

Thank you Tao Yang for your questions in comment #9321 here let me reply to points raised

Reply to 1.:

According to PnP 'original source' is one type of evidence that the IRG can accept but not the only type. There are 5 types of evidence that can be accepted: original source, multuple source, semantics, context and usage. The PnP says:-

'a character submission must be accompanied by evidence to satisfy at least one of the following
conditions:
a) Original Source ( 證 據 源 限 制 ): The source of evidence must be considered authoritative by IRG, as validated by past literature and IRG experts. IRG has the right to reject characters from questionable sources.
b) Multiple Sources ( 多 源 證 據 ): Supply character use evidence from multiple independence sources. IRG has the right to reject characters with evidence of use from only a single source, especially if the source is not considered authoritative by IRG.
c) Semantics (字理考證): Supply sufficient evidence on the meaning and phonetics. Supply of other information on its origin and evolution would be very helpful.
d) Context (上下文信息): Sufficient context in text to decipher the semantic meaning of the character. IRG has the right to reject characters that do not have sufficient evidence for IRG to decipher its semantics.
e) Usage (需求限制): The use of characters must be for justifiable public interest. Examples of public use include evidence of: governmental needs; scientific use; digitization projects for public use; and working systems of significance as accepted by IRG. IRG has the right to reject characters that do not have sufficient evidence for IRG of justifiable public interest.'

The evidences given are of the multiple source, semantics, context and usage type. Original source evidence is evidence for a character from a recognised authoritive source, a single source that is so important that we can say the character should be encoded because it is in that source. The evidence for this character is not of the 'original source' (an important authoritive source) type. However it is of three or four of the other types, it is: (1) from multiple sources (2) semantics as it clearly states the pronunciation and meaning (3) context sufficient for a native speaker of the language (4) usage - it is and can be used.

Reply to 2.

One change in the current PnP to previous versions is the permitting of a small number of characters with multimedia evidence. Literature is and will always be the main source of eidence for IRG but it does not need to be the only one. Multimedia evidence is new for the IRG and so there is need to talk about different different questions and build consensus.

Reply to 3

All the evidences given are verifiabled. Whilst there are many unreliable videos but this does not mean all videos are unreliable.

Reply to 4

Analysis of the sources used revealed that the characters used were neither arbitory nor designed by the creator. Also checked was that the character concerned was used by diffferent generations and dialects, and therefore can say is not small scale.

Reply to 5

This I think goes back to point one - the evidences are considered representative, or indicative. It is not suggested that the character is important because it is found in these evidences, but rather that because the character is impotant it is found hear.

Reply to 6

This character was even in the earlier 2006 version of the table.

Thank you John, very appreciate to your analysis.

We made overly complicated statements, but in reality, the problem is only about one dangerous point:

Submitting only video clip evidence will invalidate IRG's review and PnP.

If certain characters cannot be accepted by IRG, the submitter can create several videos and upload them online as new evidence to IRG, which is equivalent to using a very basic method to bypass our complex review mechanism.

I believe these characters must exist in the literature, but we cannot ignore the possibility of bypassing the review process. This is a very noteworthy issue.

Because it is too easy to use self-made combination characters in videos, it is difficult to prove their existence in literature or their widespread use in daily life through a few videos. There are indeed actual cases of use in the video, but we cannot prove that these characters propagated through the video have become characters that everyone can accept and encode.

The previous multimedia evidence submitted by Eiso Chan, came from authoritative institutions such as Shanghai Animation Studio and Xi'an Film Studio, which have been reviewed by the National Film Administration, so we can use it as evidence. However, such subtitles were produced and uploaded by individuals, and the authority of the multimedia evidence we received last time cannot be compared.