Commons:Stroke Order Project/Fonts
This page lists available fonts for regional-specific glyphs.
Fonts and Commons edit
Font licenses, Chinese characters, copyrights and Commons create together a complex issue.
Aside of Songti/Mingti fonts such as Google's NOTO series, there isn't CJK fonts under true Open License.
Nevertheless, lowly stylized CJK characters shapes such as Kaishu, Mingti, Songti, Clerical, Seal are very ancient shapes which cannot be copyrighted and are therefor within the Public Domain. The shape —aka the character— can be uploaded on Commons. But the font files and it's original vector data —the x/y numbers in the font and into the svg— are copyrighted and cannot be uploaded to Commons. Also :
- A raster image of a character, presenting the shape but not the font data, can be uploaded to Commons.
- A SVG image where svg nodes data are different but for which the overall shape stay nearly the same has no copyrighted data (but yours) and a PD shape, and therefor, can be uploaded to Commons.
For questions, please message on talk page and ping us on Commons talk:Stroke Order Project.
Simplified Chinese Hanzi edit
- HDZB_36.ttf - 汉鼎简楷体 - Hàndǐngjiǎnkǎitǐ - Simplified Glyphs
- HDZB_7.ttf - 漢鼎繁隸變 - Hàndǐngfánlìbiàn - Traditional Glyphs
- HDZB_74.ttf - 漢鼎繁中楷 - Hàndǐngfánzhōngkǎi - Some of both kinds
These are (supposedly; we need a primary source to verify that) free fonts that have been used so by Commons Stroke Order Project contributors. They can be obtained from http://www.certifiedchinesetranslation.com/fonts/Chinese.html.
Note, however, that these fonts only contain GB2312 mappings, which will be unusable by graphics software the requires a Unicode font. This includes the Windows version of "The GIMP"! User:Tauwasser took the effort to remap these to the Unicode character set and has made them available at Drop.io/sala6re.
Traditional Chinese Hanzi edit
- edukai.tff (edukai-4.0.zip) (CC-BY-ND-3.0—Taiwan Ministry of Education / 中華民國教育部 - english).
- kai-pc.ttf - 教育部標準楷書 - PC version
- kai-linux.ttf - 教育部標準楷書 - Linux version
- ct.sit - 教育部標準楷書 - Macintosh version
These are free typefaces standardized by the Ministry of Education, R.O.C. (Taiwan). They can be downloaded from the R.O.C. Ministry of Education's National Languages Committee's website.
Japanese Kanji edit
There are several nuanced differences between Kanji and traditional Chinese glyphs that makes it easy to mix up one's fonts. The grass-crown, for example, has three strokes in modern Japanese glyphs but four in traditional Chinese.
The Sazanami (now VL?) and Kochi-substitute fonts are free, but not nearly as nice as the Chinese fonts above.
- An alternative is to simply modify the Chinese fonts to fit with the Sazanami shape.
Already checked...
Korean Hanja edit
-- To be included --
Vietnamese Hantu and Chu Nom edit
The Vietnamese Hantu set is currently not encoded into Unicode! Therefore, we cannot say for sure that codepoints won't change! See w:Chữ_Nôm and w:Hán tự for further reading.
That said, there are plans for plane 2, the SIP (Supplementary Ideographic Plane), to contain these as the CJK Unified Ideographs Extension C. The original submission for Extension C has, however, been split into C1 and C2[1]. Part C1 is currently under review, the latest progress of which can be seen on the Documents for CJK Extension C Project website. There are currently numerous documents containing the unicode mapping for Extension C.[2]
The entries to Unicode should map according to the index in these lists.
Fonts available including the Han Nom characters (with future mappings) are
- Han Nom A.ttf - contains additional characters not implemented additional to Han Nom
- Han Nom B.ttf - contains characters in plane 2A7 (U+2A700 - U+2A7FF)
which can be obtained from the SourceForge VietUnicode Project. These, however, also lack mappings that were added in Unicode 5.0, so they may be a little outdated.
Basically, it hasn't been decided which glyphs will be adopted into Unicode. Current proposals may therefore become invalid and character codes might point to different characters alltogether. This would be an issue here on this project in terms of filenames referring to the correct image.