請還原Traditional Chinese的眞正Tradition寫法

[SyaoranHinata]

正體(繁體)中文所用的寫法，個人認為大有問題。從圖例顯示，寫法是向台灣敎育部靠攏。然而，使用正體中文的地區有許多，如香港、澳門和海外華僑等，其他地區並不以台灣寫法爲尚。此外，即使在台灣，過去舊日的書刊出版，乃至今天的報章，主要使用的也不是敎育部寫法。大家主要使用的，是過去傳統字書裏的正體寫法，有人稱作「舊字形」，日本朋友會叫「康熙字典體」——不是指某款遭濫用的字型，而是指參照同文書局原版的《康熙字典》每字字頭之寫法。這種寫法，一來有充份字理，二來在字型美學上也較美觀。至於台灣敎育部寫法，則以楷書寫法，來強行扭曲明體、黑體等印版字型，既缺乏字理，也不夠美觀，已有不少人詬病。在下由衷感謝　Adobe的貢獻，但極望　Adobe能把正體中文的字型，改回眞正正統的《康熙字典》寫法（即「舊字形」），而不是台灣以手寫楷書扭曲黑體的寫法。不勝銘感！

我不反對有台灣人想用台灣敎育部的寫法，但也應還其他Traditional Chinese使用者，使用眞正Tradition寫法的空間，分拆開「Taiwan」和「Traditional」兩體。而不是強迫其他正體使用者依從台灣那種以楷扭曲黑的寫法。

[RJHsiao]

Please Write down your description in English. The maintainer may not understand Chinese.

[be5invis]

He said that you should create another variant which follows the glyph shapes in Kangxi Dictionary (康熙字典), which is closer to Korean variant, instead of the Taiwan standard.

[kenlunde]

The glyphs that we included were constrained in a small number of ways. First, the representative glyphs as used in the national standards of each region are those that are preferred, and it is not really a matter of correctness (which can be subjective, and can change over time) but rather one of following current conventions. Second, due to the 64K glyph limit, we needed to limit the scope of the supported standards, which was also practical. For Simplified Chinese, GB 18030 is a requirement and means URO and Extension A support. We also found that with approximately 200 additional glyphs we could support China's latest list of 8,105 hanzi, so we supported that. For Traditional Chinese, the scope is Big Five and Hong Kong SCS. The CNS 11643 planes outside of Planes 1 and 2 are thus not supported, in terms of having appropriate glyphs for Traditional Chinese. In other words, if a glyph does not look appropriate for Traditional Chinese, it is best to first check whether it is outside the scope of Big Five (CNS 11643 Planes 1 and 2) or Hong Kong SCS.

[Artoria2e5]

Not really accurate translation:

Please make Traditional Chinese glyphs really traditional

In my opinion, the glyphs for Traditional Chinese has big problems. According to the examples, the glyphs are designed mostly according to the standards of Taiwan's Ministry of Education. However, there are a lot of areas where Traditional Chinese characters are used (e.g. HK, Macau (did I misspell it?) and among Chinese overseas). What's more, the MOE standard is not widely used in Taiwan publishing in the past and now. The glyph standard that most people use is still the one used in Kangxi Dictionary, which people call The old typeface or Kangxi Dictionary Glyphs, which represents fonts that uses the glyphs shapes in the Original Kangxi Dictionary published by 同文書局. This shape of strokes and characters is more reasonable, as well as more beautiful. On the other hand, MOE's standard follows the style in Kai(Handwritten)(Ref-1), which changes the styles of printing typefaces of Ming(Song) violently and unreasonablely. Lots of people are against this. I sincerely thank Adobe's contribution of making such a nice font Open Source, but I really hope that Adobe can use the KX Dict glyphs instead of the 'malformed' MOD glyph standard.

I am not against the Taiwanese who may want to use the MOE standard. But please at least make a really Traditional one, and move the existing one to Taiwan, instead of making everyone use the Taiwan MOE one.

---

Some random Googling...

Ref 1.0: Making printed glyphs look closer to those handwritten Kai is often called “宋体楷化”, literally "Kai-lizing Ming(Song)." Err, everyone knows this is not serif font, so let's say 'handwritizing printed typefaces'~

This is often discussed, especially on a Q&A SNS site called zhihu in China: www.zhihu.com/question/20770206

Ref 1.1: http://blog.justfont.com/2013/05/lets-talk-about-kanghsi/

Ref 1.2: http://www.douban.com/group/topic/38841615/#!/i!/ckDefault

新字型標準有甚麼問題？What's the problem of the new MOE standard?

• http://founder.acgvlyric.org/iu/doku.php/%E8%AA%AA%E6%96%87:%E9%8C%AF%E5%AD%97%E8%88%87%E5%82%B3%E7%B5%B1%E5%AD%97%E5%BD%A2
• http://founder.acgvlyric.org/iu/doku.php/%E8%AA%AA%E6%96%87:%E5%AE%8B%E9%AB%94%E6%BB%BE%E8%9B%8B_%E9%82%84%E6%88%91%E7%B4%B0%E6%98%8E
• http://founder.acgvlyric.org/iu/doku.php/%E8%AA%AA%E6%96%87:unicode%E6%91%A7%E6%AE%98%E6%AD%A3%E9%AB%94%E5%AD%97
• http://founder.acgvlyric.org/iu/doku.php/%E8%AA%AA%E6%96%87:%E4%B8%8D%E7%9F%A5%E4%B8%B9%E9%9D%92_%E6%9E%89%E8%AB%87%E6%BC%A2%E5%AD%97
• http://www.ideographer.com/articles/article.php?aid=67
• http://www.ideographer.com/articles/article.php?aid=68
• http://www.ideographer.com/articles/article.php?aid=69
• http://www.ideographer.com/articles/article.php?aid=70
• http://www.ideographer.com/articles/article.php?aid=17

Ref 1.3: zh.wikipedia.org/wiki/黑體-繁#字形變遷

[lianghai]

@kenlunde: The reporter is talking about glyph standard instead of character set coverage... However personally I totally understand why 國字標準字體 was employed as the glyph standard.

[zerng07]

I guess the original poster's idea is to request a "kangxi variant" of traditional Chinese besides "MOE Taiwan" variant which now Source Han Sans provides, because some people who use traditional Chinese in other regions (for example, Hong Kong, overseas Chinese...) do not follow Taiwan's standard.

So the main problem here is that is it possible to have a new "kangxi variant" for traditional Chinese? Will there be any chance for community to create such a variant via some help from Adobe or Google if it is not going to happen in official plan?

[Artoria2e5]

I think this would be some really hard work for the community, since a great number of structures will be changed to fit the kanghsi one.

Hmm, So it would be 思源黑体 从正/Source Han Sans Kanghsi?

Well, great news to those who has some knowledge on Chinese characters and their sources, that would be the real “思源”……

[ShikiSuen]

Please KEEP the Traditional Chinese Glyphs as current (Taiwan Gov. Standard, aka. CNS 11643 ). This glyph set is the same as what people likely to write in daily life (for people lives in Taiwan). (Thanks to LiangHai for his suggestions on this comment.)

Meanwhile, I don't think there's a referrable standard of what SyaoranHinata want... Maybe HongKong authorities? Whatever, HongKong standard could be a little bit considerable if treated as a branch standard of Traditional Chinese Glyphs.

As what LiangHai said in following comments: 「The reason Source Han Sans TWHK must follow it is: CNS 11643 is the national standard in Taiwan — although it's not mandatory.」

The current Traditional Noto Sans should only be treated as TAIWAN version.

======Chinese Version====== 請保留國字標準字體（CNS 11643，台灣當局現行標準）， 因為那更趨向於台灣人平時在寫的筆劃、且該標準動手寫的話也是迄今所有漢字標準當中最順手的。 同時，我不認為版大/PO主/樓主想要的那種字體方案有無固定規範可考…或許是香港標準也說不定。 如果可行的話，倒是可以給Noto Sans的正體中文字體單獨做一個香港繁體中文的分支。 我不認為舊字型被排擠是好事…但如果要讓舊字型徹底取代國字標準字體、讓大家沒有國字標準字體可用的話，我只能說髒話了。

（某種程度上而言，我認為華文社做出這種國字標準字體真的很良心， 這是我見過的最讚的國字標準黑體套裝，相信明天會更好。）

From @lianghai 's Twitter: 「不做开源项目的理由：免得 https://t.co/fkG3Tuv6NI 这种＊＊＊censored adjectives＊＊＊找上门。旧字形（或至少是非「国字标准字体」的繁体中文字形）很重要，但不需要你这种下三滥的革命开炮方式。Adobe 没有做旧字形，是因为根本就没有精确的标准可循、也没有地区强制要求它。」

[lianghai]

This glyph set is the same as what people writes in daily life.

No. 國字標準字體 is hardly what people write every day. The reason Source Han Sans TWHK must follow it is: 國字標準字體 is the national standard in Taiwan — although it's not mandatory.

[irvin]

You should report the problem to standardization bodies but this font, if you consider to following the standard has problem. Actually, it's really glad across the community that a long-waiting standardized glyphs Traditional Chinese open source font had been birth.

[hfhchan]

@irvin the standardization process of Chinese characters in Taiwan has been famously controversial. The Taiwanese government has long taken a very strong stance against the revivial of KangXi radical styles.

[hfhchan]

To make it clear, the writing form used by Hong Kong people closely resemble our 常用字字表 "List of Graphemes of Commonly-used Chinese Characters", and the fonts adhering in general to 香港電腦漢字字形參考指引 "Guidelines on Character Glyphs for Chinese Computer Systems", both produced by the Hong Kong Government.

As it is frequently misunderstood by people, and to make it clear again, the Taiwan MOE Standards and the Hong Kong Standards are VASTLY DISTINCT and depart to the extent that is similar to the PRC Standard and the Taiwanese Standard. In fact, the page http://zh.wikipedia.org/wiki/%E5%B8%B8%E7%94%A8%E5%AD%97%E5%AD%97%E5%BD%A2%E8%A1%A8 contains a pretty lengthy (but nowhere complete) comparison between the two standards.

To further clarify (complicate?) the issue, the glyph forms used normally in Taiwan are not the forms specified in the KangXi dictionary, but instead are a modern variant (read: 20th century). In traditional Chinese using communities, the term "KangXi forms" usually refers to these modern variations. However, in simplified Chinese communities (especially mainland China), the term "KangXi forms" refer explicitly to the original versions. The reason for this discrepancy is another political and cultural issue, however the point is: the distinction in definition should be noted for a proper and accurate discussion.

In Hong Kong, the KangXi forms the original forms are completely out of use in commercial and personal contexts; it is only when a classical feel is intended, then are the original KangXi forms are used. The modern variations, however are still used by small/middle business and organisations. The modern variations have not ever been standardized properly, and it is still debatable of whether the KangXi forms were actual forms used, or simply a standardization for the sake of standardization.

[ShikiSuen]

小林剣博士へ(to @kenlunde) : 内木さんの欲しい字体は香港の常用字字形表ではなく、康煕字典体でもなく，康煕字体の現代の支脈です，そして、適用される標準はありません。 What Uchiki Ichirou (@SyaoranHinata) want does not met Hongkong Standards, and is actually a modern branch of Kouki Glyph System which has never been standardized.

[lbrabbit]

Can ShikiSuen translates his reply to @SyaoranHinata in English? Why just translate the reply to @kenlunde?

[SyaoranHinata]

感謝大家的幫忙、翻譯和討論。在下的英文甚爲蹩足，請恕在下先暫用中文把意思寫出來。

香標有教學標準《常用字字形表》，不過限於楷字，編撰者說明只是給小學至初中的識字敎育作一個參考，並無硬性規範一切之意。在下成長時，敎科書內容若着重規範，就只使用楷體。明體、黑體字，長期已來沒有給限死。後來聞說是爲了方便把「香港字」（e.g.嘅、喺、鰂、𨋢、邨）申請作國際標準，「中諮會」搞了套宋體指引「香港電腦漢字字形參考指引」，但其字形由台灣標宋修改而成，有若干地方與《常用字字形表》不一致，現實上也沒有誰使用。所以沒有提出以它為據。但我看到issues/18對此已有討論。要是也能製作這種香港版本，我也贊成的。

因此我就着Traditional Chinese各使用地區的情況，提出「康熙字典寫法」這種最大公因數方案。大家討論到「康熙字典體」不標準，或者無現行政權承認。我曾在個人舊文裏指出，《康熙字典》裏個別字的寫法已out了，給民間眞正約定俗成的寫法取代，例如「在」、「壺」等字，但爲數不多，而且其取代局面有目皆見，根本就沒有爭議，update之便可。日本update它後，製成的《大漢和》舊字形標準，仍依習慣稱作「康熙字典體」。至於政權方面，哪個政權都不可能千秋萬代，都會更替。漢字卻是由古代一直傳承至未來的。因此個人會着眼於文字的實際使用情況。

還有，也許在下一向不擅表達，在開首的發言裏，可能另人誤會要取代台標、不許別人使用台標。這是在下的責任，非常抱歉。我的意思不是要取締他人，而是倡議大家可以選擇。當然，有技術限制，這點我明白，但理想的情況，是能有TW edu style、HK edu style、Traditional style等不同選擇，使大家各適其適。

[ukyoi]

There is no technical restrictions but a flood of work to do, I presume.

[Artoria2e5]

233，居然已经有人翻译好了，我干完了才看见…… 深深为我的英语水平担忧

Thanks for everyone's help, translation and discussion. I am not really good at English, so please let me write it in Chinese first.

HK has an standard meant for education, The glyph list of common characters(常用字字形表), but it is limited to Kai/Handwritten font and it is just used as a reference for education, not for use as a mandatory standard, according to the au When I was young, when the textbooks focus at the standards, they use the Kai fonts. There is no mandatory limitations for printed fonts like Ming(Serif,Song) and Hei(Gothic?). Some years later, I heard that to help make some 'HK Characters' (e.g.嘅、喺、鰂、𨋢、邨) into international standards, the HK Advisory Committee for Chinese Interface (ACCI, 中諮會) made a Ming(Song) font guideline called HK Reference Guidelines for Chinese Characters on Computers(香港電腦漢字字形參考指引), but it is actually a mod of the TW Standard Song, and it doesn't match the Common Characters really well. People hardly use the Reference Guidelines font, so I didn't suggest using it as the standard and the basis of Source Han Sans. But I have seen the discussion in #18 , of course I will also be happy if there is such a HK Ver.

So according to the actual situation of areas where Traditional Chinese are used, I came up with the idea of using KangHsi font, since it is close to all these variants and is their 'common ancestor'. When we discussed about this issue, someone pointed out that the KangHsi glyphs are not standardized, or no governments really agree with it. I used to point out in my personal blog that some characters/glyphs in KangHsi is already outdated in its shapes, and are replaced by the folk conventions, (e.g. 在、壺), but there's not a lot of them and there's no controversy about those replacements (Citation needed), we can just update them. The Morohashi dictionary, an updated version of KangHsi is such an example, and it is also commonly called KangHsi by common practice. (On the other hand, about the governments...), but Chinese characters always kept evolving from the ancient times to the present and the future. That's why I focus on the actual usage of characters.

Btw, I am not good at expressing (and Arthur is not good at translation, just kidding), someone may mistake my ideas as stopping others from using the TW MOE standard. This is my fault, sorry for that. I meant to give everyone a chance to choose what they really want from a bigger variety of writing styles. I know there may be some technical issues, but the most ideal situation is having many styles to choose from, like HK, TW, Traditional.

[ukyoi]

@Arthur200000 Your translation is better than mine so I deleted mine. 我那半吊子英文就不摆着丢人现眼了……

[extc]

It is easy to understand what is 'Old Shape' vs 'New Shape' standard (the terms refer to "印刷通用汉字字形表" of PRC). http://zh.wikipedia.org/wiki/%E5%8D%B0%E5%88%B7%E9%80%9A%E7%94%A8%E6%B1%89%E5%AD%97%E5%AD%97%E5%BD%A2%E8%A1%A8

http://zh.wikipedia.org/wiki/%E6%96%B0%E8%88%8A%E5%AD%97%E5%BD%A2%E5%88%97%E8%A1%A8

If Dr. Ken Lunde needs to follow some standard, just refer to the PDF of Adobe-CNS1-6 http://wwwimages.adobe.com/content/dam/Adobe/en/devnet/font/pdfs/5080.Adobe-CNS1-6.pdf

'Adobe 明體 Std L' and 'Adobe 繁黑體 Std B' just demonstrate good examples of de facto standard of Traditional Chinese forms. (Although some components like 爫, 文, 者, 示字部 of both font use modern form). Both fonts, with 華康中黑體 are old forms accepted by both Taiwan and Hong Kong people.

[mandel59]

As mentioned in https://github.com/adobe-fonts/source-han-sans/issues/6#issuecomment-49147331, you would be able to use the Korean variant, which has almost the old (‘Kangxi’) shape, though it has differences in punctuation marks. Korean variant with Chinese punctuation marks would be what you need.

[hfhchan]

With all due respect, I am afraid @extc you've had an over simplification (and slight muddling) of the issue. The pictures you've included, especially the last one, is incorrect in that it mixes various glyph reforms carried by both the PRC, Taiwan and Hong Kong into one picture.

To fill in the void, please allow me to attempt to be comprehensive and concise to re-explain the situation.

Due to the vast number of Chinese character users, Chinese characters have not often been written in the same way. In the Qing Dynasty, Kangxi Emperor (reportedly) ordered the Kangxi Dictionary to be made, which (reportedly) became the compulsory standard of writing Chinese characters for all official documents. The Kangxi Dictionary used a type of script now known as "Song" or "Ming", which loosely refers to "serif" in Western terms.

Evidence, however, has shown that commoner's writing have often deviated from this official standard. Furthermore, the script in use in normal contexts was "Kai" or "Regular", which was never standardized. These stroke differences were well tolerated. (compare: Would you care if someone crossed their t with a horizontal line, a horizontal slant upwards, or a horizontal slant downwards?) For most of Chinese history, more pressing was the issue of characters which have completely the same meaning but written in different ways. These words are called character variants 異體字.

In the early 20th century, there have been calls to drastically simplify the Chinese writing system due to the fact that it was seen by some scholars as archaic and outdated. Many character variants were created. By the 60s, the People's Republic of China was stuck between simplification of the Chinese writing system, or completely abolishing written Chinese. Reforms were carried out and the first reform in 1965 focused standardizing stroke differences. The abolished forms were called Old Shape and proposed ones called New Shape 新字形. The process involved reshaping the official Song standard to closely correlate with the strokes in the desired Regular script. This process is also known as 宋体楷化 or "Kai-fication of Song Script". Here is a selection of certain character reforms (which, ironically, are the one that are least controversial and been accepted into Taiwanese / Hong Kong standards to varying degrees): For the above picture, these usually concerned tweaking of certain strokes to correspond with the Kai script. There were also stroke reforms that involved accepting the removal of certain dots as official (e.g. 盗 and 盜). These other stroke reforms have not been incorporated into Taiwanese or Hong Kong standards.

A second reform was made to simplify words at the component level, e.g. from 該 to 该 (reportedly "borrowing components" from other scripts, e.g. Cursive Script), and at character level: selection of standardized form for semantic duplicates (e.g. [葱蔥] to 葱), and reduction of words by combining several words (including re-use of archaic words) to one simplified form (e.g. [臺台] to 台). Character reforms at the stroke level, and component and character level, are known as separate process in the People's Republic of China.

Across the strait, however, in the Republic of China (Taiwan), started their standardization of Chinese characters in 1973 and announced their results in 1982. Chinese character reforms were carried out in one go: one process took care of stroke level, component level and character levels reforms. Standards were set for the Kai script first, and then the Song script was also adjusted to be an "exact match" with the Song script. The standard heavily focused on Stroke reforms and selection of standardized form for semantic duplicates; component level and other character level reforms were of far subtle nature. However, the stroke level reforms imported various elements from Running script, and the hard-on approach led to widespread criticism as damaging the soul of Song script.

Hence the call for reverting to the character forms in the Kangxi Dictionary.

NOTE 1 The lines between stroke level and component level changes are quite blurred in People's Republic of China 盗 and 盜, 争 and 爭 are seen as stroke level variants, while 净 and 浄 / 凈 and 淨 are seen as component level variants.

NOTE 2 Due to the reform-in-one-go in Taiwan, and the counter-intuitive distinction used by the Peopls' Republic, in Western countries and in Traditional Chinese speaking areas, people tend to call stroke level character variations directly as character variants and name the exact orthographic chosen by the People's Republic to be "Simplified Characters". In essence, most of them only differ to the official "Traditional Characters" of Taiwan by minor stroke differences.

NOTE 3 People from People's Republic of China sometimes incorrectly assume Taiwan and Hong Kong use Old Style shapes and fail to recognize that Taiwan has itself carried out their own character reforms. The standardized form for semantic variants also differ among China, Hong Kong and Taiwan.

NOTE 4 Despite borrowing of Kai features into the Song script in the PRC, the first standardized form was for the Song script only. This is in stark contrast with Taiwan, whose Song script is an exact transposition of Kai strokes into Song. Hence in PRC, the word 能 still keeps the straight line downwards in the lower left component in the Song script, while in Taiwan the official form is a stroke outwards, damaging the squareness aesthetic found in traditional Song script.

Often left out of the picture is the character standardization (read: not reform) in Hong Kong. In light of the controversies and the "bullshit" surrounding arguably minor-but-hideous stroke level reforms, especially the artificial invention (or, officially by the two jurisdictions, transposition of elements from a different script) of new stroke forms by the PRC and Taiwan, the government sponsored a project to set out a reference for teachers and students to adhere. Popularity 普遍性 was the major consideration in setting up this reference style. Despite of suggestive nature, it later became the standardized form that all approved Chinese textbooks for primary and secondary schools had to adhere to. However, teachers are still encourage to be lenient and accept variations in strokes.

Unlike PRC and Taiwan, the standard does not apply to Song script, however there exists an IT Industry guideline on application of these Kai features to Song. Hong Kong's higher leniency to character variants means major commercial fonts have greater variations from the official standard than that occuring in the other two jurisdictions. However, it is still very important to note that Hong Kong's accepted stroke variations are still more similar to our Standard and Guideline than that of Taiwan, PRC. In some cases we write and print exactly like the PRC, in some cases we write and print similar to Taiwan. FYI, despite not carrying out any simplification work similar to the PRC, our standardized word form for semantic variants also depart greatly from the Taiwanese standard (it is very fortunate this is addressed perfectly by Unicode already.)

[hfhchan]

To further complicate the issue, certain stroke-level differences have been given different unicode points. However, the majority haven't. I'm sure the font creators know of this issue :)

In my honest opinion, I don't think the Kangxi forms are preferred over the ROC standardizations either. Both standardizations are attempts at (over) engineering a certain written form of language. Kangxi was an emperor but the ROC is a democracy and criticism is not suppressed. Anyhow, fonts that adhered to the Kangxi forms (e.g. HeiTi-TC on Mac) also drawed widespread criticism. Commercial fonts have always tried to find a balance between the KangXi forms and ROC MOE forms. This is want Hong Kong's character standard practically does and what the other commerical fonts (especially MonoType) has done.

I am not entirely convinced that the Kangxi forms should replace the MOE form in the TC font due to this. Though I am still in favor that this font that adheres to MOE form should be rightfully named as Source Han Sans Taiwan or Source Han Sans ROC, where Taiwan/ROC in this case refers to the jurisdiction which requires it. Again, this font deviates strongly from Hong Kong's written norms. Addressed in issue #18

[Artoria2e5]

请放过思源 黑 体……

[zerng07]

Humanlist Sans-Serif is a kind of font style. Source Han Sans following the calligraphy writing style is a feature of Humanlist Sans-Serif.

Furthermore, there is only ONE existing standard available for sans-serif style of traditional Chinese characters - Fanti standard by MOE in Taiwan. And the standard uses calligraphy style (Kai style) too. I don't see any reason for designers or vendors not to follow this standard when making a Humalist Sans-Serif font for traditional Chinese.

[ShikiSuen]

I totally agree with zerng07.

[hfhchan]

@zerng07 Designers and vendors for popular Chinese fonts in the Traditional Chinese areas (namely MonoType, Dynacomware) have often refused to follow the MOE versions for reasons that include language-re-engineering and the balance-upsetting hideous nature of the standardized MOE character glyphemes for Fangti. I understand it is easier off for big organisations to be politically correct and follow the standards. But it is only up to these organisations that have the manpower to create comprehensive high-quality open-source fonts.

A fork should definitely be considered, considering that people who prefer the Kangxi style would seriously outnumber some other less used languages. And still, it is disrespect to call this for TWHK when obviously it does not respect the Hong Kong writing norms.

[zerng07]

@hfhchan To make a sans-serif font follow calligraphy writing styles or not is just a design principle for designers or vendors to choose freely. Those examples you provide above mean nothing. You cannot say that "Oh, other Sans font products by major Chinese type foundries doesn't follow calligraphy writing styles, so you shouldn't consider this option." Right?

In my opinion, any words about "following calligraphy writing style really hurts my eyes" or "following calligraphy writing style breaks the beautiful balance which comes from Mingti style or Kangxi style (Serif font) for Heiti (Sans-serif font)" are just personal taste of aesthetic preference.

I have to say that again, the real problem here is that is it possible to have a new "kangxi variant" for traditional Chinese from Adobe or Google? And will there be any chance for community to create such a variant via some help from Adobe or Google if it is not going to happen in official plan?

By the way, I agree the idea that "TWHK" should be rename to "TW" too.

[SyaoranHinata]

"Humanlist Sans-Serif" is not the same as "calligraphy writing style". Like Calibri ( https://en.wikipedia.org/wiki/Calibri ), the shapes of letter "a" and "g" are not the same as handwriting. Even source-hans-sans is a humanlist sans-serif, it is still a 黑體"East Asian gothic style", so characteristics of 楷體"Kai style" are never really match 黑體. There're really different requirements between 楷體 and 明&黑體 in Chinese.

Besides, there's a HK educational reference document 常用字字形表 "List of Graphemes of Commonly-used Chinese Characters". And 康熙字典"Kangxi Dictionary" itself, or 大漢和”The Morohashi dictionary/The Dai Kan-Wa Jiten” itself, is also a standard. I don't think there is only ONE existing standard available. And now "Kangxi" or "Original style" (#6) and "Hong Kong educational style" (#18) are suggested.

One reason that I suggest "Kangxi" or "Original style" is we often use it in real life. I don't mean the people only use Kangxi style. Usually, people may choose Kangxi style (e.g. the original Window's MingLi, Dyanfont Ming ar Hei, Arphic font, Fangzheng old style series), the style that closely to "HK edu" comparatively (e.g. Monotype, Dynafont 儷"Li" series), and different educational standards... etc. (Still there're subtle differences inside a catalog of the rough classification) I don't mean to replace. I wish that people can choose.

Chinese characters are based on the system calls 字理"reason/logic of character". Neither POC or PRC standard have a good matching of it. This is another main reason that I've suggested "Kangxi" or "Original style", as it is a standard which reach the requirment of 字理 well.

[RJHsiao]

康熙字典"Kangxi Dictionary" itself, or 大漢和”The Morohashi dictionary/The Dai Kan-Wa Jiten” itself, is also a standard. I don't think there is only ONE existing standard available.

@SyaoranHinata I Have to say, 康熙字典"Kangxi Dictionary" and 大漢和”The Morohashi dictionary/The Dai Kan-Wa Jiten” are YOUR standard, NOT the REAL standard. In fact, no one said they are standard BUT YOU.

I think the MOST important point is "good to identification", should not just focus on the "reason/logic of character".

[ShikiSuen]

to @RJHsiao: That's not @SyaoranHinata 's standard, there's currently NO standard for what @SyaoranHinata want.

[RJHsiao]

@ShikiSuen But @SyaoranHinata said they are standard too! LOL

康熙字典"Kangxi Dictionary" itself, or 大漢和”The Morohashi dictionary/The Dai Kan-Wa Jiten” itself, is also a standard. I don't think there is only ONE existing standard available.

So I said that they are @SyaoranHinata's standard but not the real standard. Don't you agree me? I agree that you said there's currently NO standard for what @SyaoranHinata want. I think he just want someone can develop a font that following his wish and have opportunity to become standard, but why the developers in Adobe and Google have to work for his wish?

The best way to @SyaoranHinata is JUST FORK IT.

[ShikiSuen]

to @RJHsiao : Actually, @SyaoranHinata 's standard has not yet been started, they need lots of time to prepare.

[ShikiSuen]

to @RJHsiao (continue) Adobe and Google are not responsible for making a font based on an unstandardized glyph sets. That's what I supposed.

[RJHsiao]

to @ShikiSuen Totally agree. :100:

[SyaoranHinata]

目前這個討論是否有甚麼問題？！ 一部在學術界有權威性的字書之字形寫法標準，並不是標準？！ 那字書又不是我編的，它是「我的標準」？！ 是否有人連客觀事實，都可以如此歪曲？！

[ShikiSuen]

to @SyaoranHinata Chill out, Official standards are official standards.

[ShikiSuen]

(This is the Chinese version of my last reply which I think I should write it as a standalone reply)

to @SyaoranHinata : @RJHsiao 嗆你說那是你的標準， 自然是因為他看你這樣用驚嘆號脾氣暴躁、導致他不屑理解你那標準是三小。 你這樣打雞血的態度不會幫你解決任何問題，真的。

實際上，我想說的是：不要把「民間約定俗成」和「官方標準制定」混為一談。 集體無意識永遠都是混沌的， 你們不把你想遵守的那種約定俗成的東西變成明文規定的標準的話、就別怪谷歌和阿逗逼他們不依。 （哪怕你成立了一個民間的組織，以這個組織的身分頒佈明文規定的標準的話，也可以） （我還是想建議你動用海外全部粵人的力量，只有香港人撐腰的話沒用）

[RJHsiao]

@SyaoranHinata 客觀事實是：權威不等於標準，這是兩回事。你不能說我不同意你的說法就說我在歪曲一件客觀事實。 The fact is: "the authority in some region" is not equal to "the standard in some region", that's totally different.

你希望你認為的標準能真的成為標準，請先把那些東西標準化出個有實際的東西出來。 你希望大家能幫你，請先把你的構想、預計會進行那些事情、完成了哪些東西展現出來。 Linus Torvalds said: "Talk is cheap, show me the code"

[SyaoranHinata]

「標準」一詞，並沒有官方的意思。《國語辭典》：「衡量事物的準則。晉˙孫綽˙丞相王導碑：『信人倫之水鏡，道德之標準也。』唐˙韓愈˙伯夷頌：『彼獨非聖人而自是如此，夫聖人乃萬世之標準也。』」政權可以頒佈標準，但不一定只有政權頒佈的才是標準，這是客觀事實，世界上許多標準都不是政權頒佈的，學界權威人士發明、製訂或創造出的標準也多的是。

[ShikiSuen]

@SyaoranHinata 建議您暫時先休息一下、和Even Wu他們私底下商量一下對策，效果比你在這邊討論要好得多、且他們都會給你很有用的建議。乾著急的話傷身體（我先前還著急害怕國字標準字體被拋棄呢，現在我都看開了）。

//Have a good sleep, and then talk to Even Wu for suggestions. That should be more effective than wasting time here.

[RJHsiao]

@SyaoranHinata 我想我從頭到尾都沒有提到 "官方" 兩字，我是在提醒你，不該把權威跟標準混為一談。「標準」一詞的意思、含意，我多少是知道的，雖然我才疏學淺，無法像你這樣迅速的引經據典來解釋一個詞，至少許多標準為民間制定這件事我是知道的，而我也知道「權威」與「標準」就是不一樣。

I already know that the standards are not only can be defined by government, and a also know that "authority" and "standard" are totally different.

[hfhchan]

The point is, the ROC standards are not without widespread criticism. I'm not sure about Taiwan, but the rejection of ROC standards were (and still is) one of the main reason for the establishment of the Hong Kong Standards, and the Introduction of these Hong Kong Standards still criticize the ROC standard using words, arguably of subjective nature, such as "misleading" and "incorrect".

@RJHsiao If I am not wrong, the ROC has never attempted to dictate their standardized variants in use outside Taiwan. The Taiwan standardized version is not applicable to Hong Kong. IMHO the selection of ROC over Hong Kong standard is arguably simply for convenience or neglect to see that our written characters are more than just subtly different. What gives to make Taiwan MOE to be a standard, and the Hong Kong version as not a standard? Our word forms are rightfully Traditional Chinese characters too. The 《常用字字表》 may not have a name with the word Standard/標準 in it like in Taiwan (actually in Hong Kong, most standards are either laws or de-facto, rarely will you come across a document that has the words Standard/標準 in it), but the mere fact that these word forms are required by Primary and Secondary Textbooks, in use by local dictionaries, officially for forming the basis Chinese character studies, etc, suggest that it is in fact a standard. In comparison, the Taiwan MOE word forms are neither compulsory out of education contexts either. What gives to ignore Hong Kong standards?

If the Taiwan standards are under so much controversy, and there are also other standard/dictionary/whatever that has "generally consistent" character forms (Even ROC MOE refuse to align Plane 3 罕用字 with Plane 1 and 2), and is still in-use by a population that is very probably larger than many of the world's languages, I think that pretty much warrants the creation of that font?

I guess only Adobe and Google have the resources and manpower to develop such a font under a open-source license. Since they have said that this font is suitable for the Taiwan and Hong Kong region, I guess it is rightful to voice the opinion when they do not meet expectations. (whether or not it should be implemented is another issue)

[RJHsiao]

@hfhchan Your Right, the ROC has never attempted to dictate their standardized variants in use outside Taiwan. But I'm not try to ignore HK standard. I just arguing the point that I don't think 康熙字典"Kangxi Dictionary" is a standard like @SyaoranHinata said. He said it's standard because it's a authority glyph, but I don't agree that.

I don't think that is very important about which one between KangXi and MOE standard are the right in glyph shapes. I DON'T CARE. If someone think KangXi is regular and MOE standard is guilty, JUST FORK IT! Need help? Just show we the things you already done. If someone like your project, they will help you to keep going. Or ask Adobe or Google to create another new font to follow the KangXi way. Replace them to KangXi glyph is a TERRIBLE idea. Remember: They have NO RESPONSIBILITY. The most important thing for designing a font is friendly to see, and I think Adobe and Google are doing a great job. That's all.

And what about the HK standard... @hfhchan You already open #18 & #32 , don't you? I can just say: It's a very bad idea to ignore HK users.

[kenlunde]

My apologies for not replying to this rich discussion sooner. I have been without my computer since Friday, and won't have access to one until Tuesday. (I am using the GitHub app for Android on my Samsung Galaxy S4.) Anyway, please continue to discuss. This is helpful beyond words.

[irvin]

It seems that people are agree on "We're not going to REVERT the current glyphs" but "Suggest an zh-HK version of font", weather using Kanxi or other standard as reference, so maybe we can close this bug (resolved duplicate to #32 ?) and discuss on another threads to focusing on the topic.

[infinnie]

Well it might be better to free some CN glyphs (or merge them with the equivalent JP glyphs) for the HK glyphs...

[Shaun-Type]

I am an amateur type designer from Taiwan. Personally, I don’t think it’s necessary to follow the “standard” set by Ministry of Education of Taiwan. When Microsoft changed the strokes of MingLiU and introduced Microsoft JhengHei that followed MOE standard, a lot of Taiwanese people were not used to the “new” standard. Until today, I still hear my designer friends complain about the MOE standard. The MOE standard is okay when writing down with pen or pencil on paper, but when design them with “Ming” or “Gothic” styles, they often appear unbalanced or awkward.

Japan also has a standard similar to MOE Taiwan called ”学参書体” (Gakusan), mainly used for the school textbooks. A lot of type companies often publish TWO set of fonts, one set follows the usual industrial standard, and the other set follows Gakusan. And most books, newspapers, and magazines in Japan don’t follow Gakusan standard.

An interesting comparison between two standards: http://d.hatena.ne.jp/works014/files/%E5%AD%A6%E5%8F%82_%E4%B8%80%E8%88%AC%E6%AF%94%E8%BC%83.pdf

Anyway, I think there should be another traditional Chinese subset for Source Han Sans.

我是一個來自台灣的業餘字體設計師。我個人覺得在制定繁體中文字體時，不一定只能遵從台灣教育部的標準。還記得以前微軟改細明體、發佈正黑體也是依照教育部標準，很多台灣人其實不大習慣，直到今日我還有一些設計師朋友認為教育部「標準」很怪異，畢竟與我們從小習慣看到大的字不大一樣。 站在字體設計師立場而言，台灣教育部標準手寫時還可以，但一但設計成「明體」、「黑體」等傳統字體時，跟隨楷體的標準有時候讓字看起來不大平衡，有些筆畫甚是彆扭。 其實日本的文部科學省也有頒布類似台灣教育部的「学参書体」標準，主要用在教科書使用。許多日本字體廠商兩種標準都做，一種跟隨「学参」的標準，一種便是傳統的業界標準。兩種標準的字體也都會分開發行。「学参」標準的字，既使是「明體」、「黑體」，也會跟著楷書的樣子走，像是那「辶」字邊… 但除了教科書外，一般日本出版物，像是報章雜誌、書籍等，都還是用業界標準的字體。 上面的 PDF 就是比較日本「学参書体」與業界標準字體。 既然日本人都懂得分開對待不同的標準，沒道理不能有另一套「思源黑體」繁體中文字體，不是嗎？

[infinnie]

@Shaun-Type do the Japanese fonts whose names end with “N” follow Gakusan’s or the industry standards?

[ShikiSuen]

@infinnie The Japanese fonts whose names end with “N” does not follow Gakusan’s standards. That's what I know currently, and this may fit SyaoranHinata's appetite.

Reference (Simp. Chinese): 日本字体中教科书字体和学参字体的差异在哪里？ http://www.zhihu.com/question/23886050

You could check OS X's built-in Hiragino fonts, they have both "Pro" version and "ProN" version.

[hfhchan]

The JIS X 0213-2004 standards specifies glyphs in the jōyō kanji list ("common words list") should use shinjitai. Since some shinjitai and kyūjitai have been assigned the same codepoint (especially those with 辶 / 艹), it is unlikely that the original KangXi forms can be used.