source-han-sans
source-han-sans copied to clipboard
adobe-fonts
Consolidation of Post-V2 Glyph Redesign Suggestions
The following table shows the glyphs that will be redesigned in the next update, and unless otherwise noted, the corrections are from my own notes:
| Character | Glyph Name | Description |
|---|---|---|
| U+3191 ㆑ | uni3191 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3192 ㆒ | uni3192 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3193 ㆓ | uni3193 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3194 ㆔ | uni3194 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3195 ㆕ | uni3195 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3196 ㆖ | uni3196 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3197 ㆗ | uni3197 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3198 ㆘ | uni3198 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+3199 ㆙ | uni3199 | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+319A ㆚ | uni319A | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+319B ㆛ | uni319B | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+319C ㆜ | uni319C | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+319D ㆝ | uni319D | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+319E ㆞ | uni319E | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+319F ㆟ | uni319F | Make the glyph generic in terms of weight (see Noto CJK Issue #159) |
| U+4E13 专 | uni4E13-CN | Adjust the top of the top-most diagonal stroke to be slanted |
| U+556D 啭 | uni556D-CN | Adjust the top of the top-most diagonal stroke of the 专 component to be slanted |
| U+6E96 準 | uni6E96-CN | Adjust the heavy master so that the horizontal strokes of the 隹 component are evenly spaced. |
The following table shows the glyphs that were redesigned as part of the Version 2.001 update:
| Character | Glyph Name | Description |
|---|---|---|
| U+375A 㝚 | uni375A-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+389F 㢟 | uni389F-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+38A0 㢠 | uni38A0-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+3A29 㨩 | uni3A29-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+3BEC 㯬 | uni3BEC-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+3E76 㹶 | uni3E76-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+403D 䀽 | uni403D-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+414D 䅍 | uni414D-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4295 䊕 | uni4295-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+45F4 䗴 | uni45F4-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+45FA 䗺 | uni45FA-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4630 䘰 | uni4630-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4725 䜥 | uni4725-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4A60 䩠 | uni4A60-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4A65 䩥 | uni4A65-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4B48 䭈 | uni4B48-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4BD5 䯕 | uni4BD5-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4C53 䱓 | uni4C53-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+4FB9 侹 | uni4FB9-CN & uni4FB9-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5065 健 | uni5065-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+550C 唌 | uni550C-CN & uni550C-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5572 啲 | uni5572-CN | Lower the position of the 口 component |
| U+57CF 埏 | uni57CF-CN & uni57CF-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5A17 娗 | uni5A17-CN & uni5A17-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5A2B 娫 | uni5A2B-CN & uni5A2B-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5B1D 嬝 | uni5B1D-CN | Adjust the horizontal stroke on the right side so that it is centered between the horizontal strokes above and below (ExtraLight master only) |
| U+5B1D 嬝 | uni5B1D-TW | Adjust the horizontal stroke on the right side so that it is centered between the horizontal strokes above and below (ExtraLight master only); adjust the balance of the components that make up the 裊 component, using the balance of the JP glyph as a reference |
| U+5EAD 庭 | uni5EAD-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EF4 廴 | uni5EF4-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EF5 廵 | uni5EF5-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EF6 延 | uni5EF6-CN & uni5EF6-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EF7 廷 | uni5EF7-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EF8 廸 | uni5EF8-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EF9 廹 | uni5EF9-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EFA 建 | uni5EFA-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EFB 廻 | uni5EFB-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5EFC 廼 | uni5EFC-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+5FA4 徤 | uni5FA4-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+633A 挺 | uni633A-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+633B 挻 | uni633B-CN & uni633B-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+63F5 揵 | uni63F5-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+65D4 旔 | uni65D4-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+6883 梃 | uni6883-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+68B4 梴 | uni68B4-CN & uni68B4-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+6957 楗 | uni6957-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+69B3 榳 | uni69B3-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+6BFD 毽 | uni6BFD-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+6D8E 涎 | uni6D8E-CN & uni6D8E-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+6D8F 涏 | uni6D8F-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+6E55 湕 | uni6E55-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+70F6 烶 | uni70F6-CN & uni70F6-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+70FB 烻 | uni70FB-CN & uni70FB-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+7161 煡 | uni7161-CN & uni7161-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+728D 犍 | uni728D-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+72FF 狿 | uni72FF-CN & uni72FF-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+73FD 珽 | uni73FD-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+7777 睷 | uni7777-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+785F 硟 | uni785F-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+7B73 筳 | uni7B73-CN & uni7B73-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+7B75 筵 | uni7B75-CN & uni7B75-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+7D8E 綎 | uni7D8E-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+7D96 綖 | uni7D96-CN & uni7D96-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8120 脠 | uni8120-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8121 脡 | uni8121-CN & uni8121-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8171 腱 | uni8171-CN & uni8171-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8247 艇 | uni8247-CN & uni8247-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+839A 莚 | uni839A-CN & uni839A-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+839B 莛 | uni839B-CN & uni839B-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8711 蜑 | uni8711-CN & uni8711-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8712 蜒 | uni8712-CN & uni8712-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8713 蜓 | uni8713-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8A94 誔 | uni8A94-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8A95 誕 | uni8A95-CN & uni8A95-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8BDE 诞 | uni8BDE-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+8E3A 踺 | uni8E3A-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+90D4 郔 | uni90D4-CN & uni90D4-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+92CB 鋋 | uni92CB-CN & uni92CB-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+92CC 鋌 | uni92CC-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9375 鍵 | uni9375-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+94E4 铤 | uni94E4-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+952E 键 | uni952E-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+95AE 閮 | uni95AE-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9706 霆 | uni9706-CN & uni9706-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+97AC 鞬 | uni97AC-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9832 頲 | uni9832-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+988B 颋 | uni988B-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+99F3 駳 | uni99F3-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9A1D 騝 | uni9A1D-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9BC5 鯅 | uni9BC5-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9C0E 鰎 | uni9C0E-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+9F2E 鼮 | uni9F2E-CN & uni9F2E-TW | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+22321 𢌡 | u22321-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+22325 𢌥 | u22325-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+23DBB 𣶻 | u23DBB-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+2497B 𤥻 | u2497B-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+249E3 𤧣 | u249E3-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+249FB 𤧻 | u249FB-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+27A59 𧩙 | u27A59-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+289DE 𨧞 | u289DE-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+28AC6 𨫆 | u28AC6-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+28AE1 𨫡 | u28AE1-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+290C0 𩃀 | u290C0-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+2A2B6 𪊶 | u2A2B6-HK | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+2B127 𫄧 | u2B127-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
| U+2C629 𬘩 | u2C629-CN | Redesign the CN/TW/HK 廴 component to better match the TW/HK version of the 辶 component |
Firstly, thank you and the the design team of the new bopomofo in Source Han Sans 2.000. It looks way better than the 1.00X version. However, I have some minor concerns regarding the new bopomofo glyphs. Here they are:
-
The third tone mark / 上聲 / caron (U+02C7): The design doesn't look consistent with Source Han Sans style. Specifially, why is there a sharp lower joining? And the end of the strokes look strange, too. Overall the glyph looks more like a graphic symbol than a letter glyph. (Compare it with the second tone mark in bopomofo and the caron in Latin and Pinyin which are both great.)
-
The letter ㄞ (AI, U+311E) The apex/decoration of is a bit too significant comparing to letters like ㄘ (C, U+3118) and ㄣ (EN, U+3123). In a smaller font display (see picture below), the decoration may still be visible and become slightly distracting:
(I tested this on Windows 7, with TW subset OTF)
3. First strokes of letters ㄖ (R, U+3116) and ㄩ (Ü, U+3129)
The strokes are literally 豎折 (and upright one, not slanted 斜豎 as in ㄘ), but they are decorated as if they were 豎提. This decoration might look fine if the letters are designed for regular script style, but it looks awkward for a sans-serif typeface. I expect they should have stroke design similar to the first stroke of ㄓ or the 凵 radical.
4. The right ends of ㄆ (P, U+3106) and ㄡ (OU, U+3121) look closely aligned. As the final 點 strokes were evolved from final 捺 as in 又, I expect the 點 ends more to the right of 橫撇 end. I show ㄨ letter and the katakana ス for design that are better done. (The bopomofo letter ㄡ is not like katakana ヌ where the 點 stroke can be written short. Again, this stroke ending problem is not easy to see for regular script style, but turns obvious when re-rendered in sans-serif.)
@Explorer09 While I appreciate feedback on the typeface design, what you wrote above is borderline nitpicking. Unless others share your concerns, our typeface designer is unlikely to take any action on such feedback.
The new design of radical 162 辶 is of no doubt a facelift to the previous version. Thank you and your team for the great work! Will you consider also adjusting radical 54 廴 so that it follows the new design? The "𠄎" component in 廴 and 辶 is usually treated the same way, also 廴 and 辶 are interchangable in certain variant forms like 迪廸, 迫廹, 迴廻 and 巡廵. The second left one is my mockup glyph for your reference.
@tamcy I will need to determine the scope of adjusting the 廴 component in CN, TW, and HK glyphs, in terms of how many glyphs would need to be touched, in order to decide whether this can be considered as part of the next update, which is likely to be in the May 2019 timeframe (to convert the blank placeholder glyphs for U+32FF to actual glyphs that correspond to the two-kanji square ligature form of the not-yet-announced and forthcoming new era name in Japan). If the number of glyphs is relatively low, it is possible.
I just performed a check of the URO, and in terms of CN, TW, and HK glyphs that use the 廴 component, there are 41 ones that cover the following 32 ideographs: U+4FB9 侹 (2), U+5065 健, U+5EF4 廴, U+5EF5 廵, U+5EF6 延, U+5EF7 廷, U+5EF8 廸, U+5EF9 廹, U+5EFA 建, U+5EFB 廻, U+5EFC 廼, U+5FA4 徤 (2), U+65D4 旔, U+6957 楗, U+6BFD 毽, U+6D8F 涏, U+6E55 湕, U+7161 煡 (2), U+728D 犍, U+7777 睷, U+7B73 筳 (2), U+7B75 筵 (2), U+8171 腱 (2), U+839B 莛 (2), U+8711 蜑 (2), U+8E3A 踺, U+9375 鍵, U+952E 键, U+9706 霆 (2), U+97AC 鞬, U+9A1D 騝 & U+9C0E 鰎. Extension A has six: U+389F 㢟, U+38A0 㢠, U+3BEC 㯬, U+4295 䊕, U+45FA 䗺 & U+4B48 䭈. Extension B has seven: U+22321 𢌡, U+22325 𢌥, U+23DBB 𣶻, U+249E3 𤧣, U+289DE 𨧞, U+28AE1 𨫡 & U+290C0 𩃀. Someone should check my work above. If my work is accurate, it means 54 glyphs would need to be modified.
@kenlunde I had produced my list of affected glyphs using a database creating with SHS's cmap files, region map file and the IDS database from the CHISE website. Here are all CN/TW/HK characters with the 廴 component.
----------- ------ ---------- ---------- ---------- Codepoint Word CID (CN) CID (TW) CID (HK) ----------- ------ ---------- ---------- ---------- U+375A 㝚 3384 U+389F 㢟 3730 U+38A0 㢠 3731 U+3A29 㨩 4141 U+3BEC 㯬 4619 U+3E76 㹶 5321 U+403D 䀽 5809 U+414D 䅍 6096 U+4295 䊕 6443 U+45F4 䗴 7405 U+45FA 䗺 7411 U+4630 䘰 7466 U+4725 䜥 7723 U+4A60 䩠 8585 U+4A65 䩥 8590 U+4B48 䭈 8830 U+4BD5 䯕 8980 U+4C53 䱓 9112 U+4FB9 侹 10125 10126 U+5065 健 10425 U+550C 唌 12331 12332 U+57CF 埏 13488 13489 U+5A17 娗 14602 14603 U+5A2B 娫 14654 14655 U+5EAD 庭 16961 U+5EF4 廴 17115 U+5EF5 廵 17116 U+5EF6 延 17118 17119 U+5EF7 廷 17121 U+5EF8 廸 17123 U+5EF9 廹 17125 U+5EFA 建 17127 U+5EFB 廻 17129 U+5EFC 廼 17131 U+5FA4 徤 17431 U+633A 挺 18982 U+633B 挻 18984 18985 U+63F5 揵 19302 U+65D4 旔 20184 U+6883 梃 21280 U+68B4 梴 21371 21372 U+6957 楗 21659 U+69B3 榳 21831 U+6BFD 毽 22915 U+6D8E 涎 23521 23522 U+6D8F 涏 23523 U+6E55 湕 23862 U+70F6 烶 25093 25094 U+70FB 烻 25104 25105 U+7161 煡 25306 25307 U+728D 犍 25904 U+72FF 狿 26072 26073 U+73FD 珽 26478 U+7777 睷 28077 U+785F 硟 28448 U+7B73 筳 29943 29944 U+7B75 筵 29948 29949 U+7D8E 綎 31089 U+7D96 綖 31103 31104 U+8120 脠 32842 U+8121 脡 32843 32844 U+8171 腱 33008 33009 U+8247 艇 33498 33499 U+839A 莚 34212 34213 U+839B 莛 34215 34216 U+8711 蜑 36143 36144 U+8712 蜒 36146 36147 U+8713 蜓 36149 U+8A94 誔 37839 U+8A95 誕 37841 37842 U+8BDE 诞 38491 U+8E3A 踺 39378 U+90D4 郔 40714 40715 U+92CB 鋋 41697 41698 U+92CC 鋌 41700 U+9375 鍵 42039 U+94E4 铤 42682 U+952E 键 42756 U+95AE 閮 42912 U+9706 霆 43485 43486 U+97AC 鞬 43803 U+9832 頲 44037 U+988B 颋 44170 U+99F3 駳 44845 U+9A1D 騝 44919 U+9BC5 鯅 45715 U+9C0E 鰎 45837 U+9F2E 鼮 47223 47224 U+22321 𢌡 60035 U+22325 𢌥 60036 U+23DBB 𣶻 60343 U+2497B 𤥻 60538 U+249E3 𤧣 60567 U+249FB 𤧻 60575 U+27A59 𧩙 61145 U+289DE 𨧞 61296 U+28AC6 𨫆 61334 U+28AE1 𨫡 61343 U+290C0 𩃀 61443 U+2A2B6 𪊶 61582 U+2B127 𫄧 61617 U+2C629 𬘩 61723 ----------- ------ ---------- ---------- ---------- [OK] Found 102 codepoints with 130 glyphs
The import script may have bug and the CHISE database may not be comprehensive, but I think it's pretty close. The list covers all the codepoints you mentioned. Hope it's useful to you.
@tamcy Nice. You had two errors that I corrected in your post above: 筳 was listed as U+7B75, not U+7B73, and 鰎 was listed as U+389F, not U+9C0E. In any case, between our two checks, I think that the scope has been accurately determined, unless someone can identify additional CN, TW, or HK glyphs that include the 廴 component.
Also, I removed U+269DD 𦧝 and U+26FF8 𦿸 from your listed above, and adjusted the figures at the end accordingly. These two ideographs include only JP glyphs.
In HK glyph, based on personal preception: For the character 蘭 it seems like the upper grass root isn't symmetric for both sides For the character 該 the second slant on the bottom right feel kinda strange, maybe moving it a bit higher and more to the right would be better? For the character 賓 is the overlap between the slant and the "shell" component a bit too much? For the character 亦 it now seems to be a little bit strange For the character 無 is the central horizontal line a bit too light?
@c933103 Okay, here we go:
U+862D 蘭 = The HK glyph is a glyph shared by TW/HK, uni862D-TW, and has been unchanged since Version 1.000, released almost 4.5 years ago. U+8A72 該 = The HK glyph is a glyph shared by CN/TW/HK, uni8A72-CN, and has been unchanged since Version 1.000, released almost 4.5 years ago. U+8CD3 賓 = The HK glyph is a glyph shared by CN/TW/HK, uni8CD3-CN, and has been unchanged since Version 1.000, released almost 4.5 years ago. U+4EA6 亦 = The HK glyph is specific to HK, uni4EA6-HK, and was introduced in Version 2.000. U+7121 無 = The HK glyph is a glyph shared by TW/HK, uni7121-TW, and has been unchanged since Version 1.000, released almost 4.5 years ago.
uni5B1D-CN and uni5B1D-TW (嬝): I would like to suggest to adjust the longest horizontal stroke of the 鳥(sans 灬) component in the ExtraLight master such that it is evenly spaced vertically.
In addition, the organization of the components of 裊 in uni5B1D-TW looks a bit weird. It looks to me that uni5B1D-JP is much better composed, so if possible I would like to suggest to redesign the TW glyph base on the current JP glyph.
Adjustments to the CN and TW glyphs for U+5B1D 嬝, uni5B1D-CN and uni5B1D-TW, are now reflected in the table at the beginning of this issue.
After redesigning 廴 (CN/HK/TW versions), next up, every character with the 专 component should be redesigned to "match the TW/HK version of the 辶 component".
We have 传, 抟, 䌸, 䏝, 转, 砖 (all taken from Wiktionary that the font supports at the moment). I don't know if there's more characters containing that component, but they need to be redesigned.
@Marcus98T Dude, why didn't you bring this up in a more timely fashion, like prior to the Version 2.001 release? And, for such suggestions to be actionable, a definitive list needs to be provided. Keep in mind that this is a massive project with nearly a half-million glyphs, and has tons of moving parts and interdependencies thanks to the five sets of mappings and related OpenType features.
While we are open to suggestions, please, please, please, please, don't do this in a piecemeal fashion. I do have critical tasks besides this particular typeface. And, it truly makes me wonder which component you'll suggest for redesign next.
(This should have been posted to #204. My apology)
In version 2.001:
(1) U+9834 頴 (HK)
There's still a problem with the new U+9834 頴 glyph.
(2) U+5795 垕 (HK and TW)
- The composition of components is different. Interestingly the new
uni5795uE0102-KRintroduced in 2.000 is exactly what the TW and HK glyph should look like despite one subtle difference in 𠂆 which is a design choice made in Source Han Sans.
uni8E04-CN: Suggest to make the lower vertical stroke to touch the horizontal stroke on above, so that the design aligns with all other characters with the 步 component:
uni57D7-CN: This is a commonly used HKSCS character. The 土 component on left is shrunk too much, causing the character to look strange. JP glyph looks good, I suggest to use it as a reference to tweak the CN glyph.
Alright, I did a real extensive research... and found loads and loads of characters with the "ㄅ/ㄎ" part that technically should be redesigned to match the 辶 (TC version), 廴, 东 and 乐 components. All SC, TC and HK variants are affected.
In addition, the small bopomofo characters ㆴ (U+31b4) ㆵ (U+31b5) ㆶ(U+31b6) should be redesigned to match their corresponding characters ㄅ(U+3105), ㄉ (U+3109) and ㄎ(U+310E).
Here is the list (Components with 专, 丂, 亏, 与, 互, 㐄, 马, 牙 and 弗):
专传抟转䏝砖
考侤拷洘栲烤铐銬鲓鮳
丂巧窍汅朽攷
亏圬扝污杇肟釫虧夸芌咢雩疞迃
夸垮胯跨挎绔刳桍挎姱誇
粵粤澚
亐(JP glyph default) 汚
兮枍肹盻諡
与欤屿玙写泻
㸦
互 仾(JP glyph default) 冱坘沍枑魱 彑互宐彖彘彜彝䇘
㐄
舛舞儛躌舜僢橓瞬蕣粦僯隣隣撛噒嶙獜潾憐嫾璘橉辚轔暽膦燐磷瞵疄繗蹸鏻鳞鱗驎麟麟亃鄰粼甐斴翷遴桀傑搩嵥滐榤磔謋 曻(JP glyph default) 僲桝荈
韋偉喡幃徫湋愇媁瑋椲暐煒禕稦褘緯諱韓鍏郼韡葦韏闈圍違韑韙颹衛
马 驭 驰 驮 驯 驳 驴 驱 驲 骀 驸 驾 驹 驽 驶 驷 驼 驿 驵 驻 驺 骉 骇 骅 骄 骆 骂 骈 骁 骃 骋 骏 骊 骎 骍 验 骖 骒 骐 骑 骕 骓 骔 骙 骗 骚 骛 骘 骜 骝 骞 骟 骠 骢 骡 骣 骤 骥 骦 骧
马冯吗犸妈玛杩祃码蚂骂笃蓦骉闯羁
牙伢冴讶訝呀岈犽玡枒砑䥺釾蚜谺齖邪鸦鴉雅芽穿笌閕厊庌疨迓颬衺
I almost forgot these glyphs at the last minute (so no image yet): 弗佛 梻(JP glyph default) 坲拂咈彿狒沸怫绋紼柫昲胇炥砩鉘鮄刜艴茀岪笰羛髴费镄費㵒曊鐨靅鄪疿氟
Image (Defaulting to SC glyphs):
I have found four characters defaulting to JP glyphs for SC, TC and HK: 亐仾曻梻 (Source Han Serif have the SC/TC versions of those characters)
Sorry if there may be duplicates.
I know maybe it’s too much and too burdensome to redesign all the hundreds of glyphs I’ve provided, but that’s the best I can do to make the redesign consistent with the components of 辶(TC version), 廴, 东 and 乐.
@Marcus98T I think what you want is a "decroation" at the turning point of all similar stroke type. But I am not sure if the component 辶 (TC), 廴, 东 and 乐 can be regarded as the same type in this particular typeface.
First, 东 and 乐 are different from the other characters you mentioned, because the stroke before the horizontal stroke for 东 and 乐 isn't straight, but slightly curved. In constrast, all the characters you reported have a straight stroke before the horizontal stroke.
Second, I think the characters you mentioned are not directly comparable to 廴 and 辶, not only because of the difference in stroke composition, but also because of the angle of the slanted stroke. Unlike 辶/廴 which is essentially an Z shape (the slanted stroke before the horizontal stroke starts from one end to the other end), the slanted stroke in characters like 牙, 弗 before the horizontal stroke is just slightly slanted. They can be regarded as a variant of the vertical stroke where no decoration is needed at the turning point.
I am not saying my explanation must be correct. There's no problem making them the same, but leaving them different doesn't necessary result in inconsistency (which mostly depends on how one interprets the design), so I don't think it is a "technically must" to redesign the glyphs you mentioned for consistency. To me this is more or less a designer's choice. Otherwise I can easily spot another "inconsistency" in the font which you have not mentioned:
@tamcy I see what you mean. I appeared to have missed out on the 山 component, but for now, it will probably depend on how the designer will see it when the next update cycle begins. I’m just giving suggestions on how to improve the look of the SC, TC and HK fonts. Again, it’s up to interpretation, so basically my suggestion is probably to add a deco when the 𠃋 part is not 90 degrees, whether or not the top part is curved. For 山, I’m honestly not sure about it. But for 专 and others where there is an obvious angle, I honestly felt a redesign is necessary. Like the letter Z in Avenir, Helvetica and San Francisco and even the Latin characters of Source Han Sans/Source Sans Pro, I think it would look better without the pointy angle, although the Z in Futura has a pointy angle.
UPDATE: With that in mind, I propose this redesign of the "ㄅ" part in picture, in addition, modify the 㸦 glyph for better balance.
And at the same time I'm wondering if the 山 radical on the side (which means it doesn't include 山 (U+5C71) itself, and characters where the 山 radical is placed on the top or bottom (e.g. 峯 and 密)) can be redesigned like this (the list of affected characters are not up yet, just a sample):
And then once again I forgot the rest of the characters with the 弓 radical, along with the 齿歯齒, 缶, 旡 and 毋母 radicals.
Funny enough, 既旣 in the 旡 part have the decorated 𠃋 where the others do not.
Anyway you get the idea. I will prepare all the characters with the left-side-山/齒/缶 and those with the 弓/毋母/旡 parts that would be affected (not guaranteed that it will be complete). Looks like a real big redesign if they ever take my suggestion. 🤔
And, it truly makes me wonder which component you'll suggest for redesign next.
Unfortunately, I fell into the trap because I continue to dig this ever increasing rabbit hole with even more radicals to “redesign”. But this time I’ll try to be fast and comprehensive once and for all, so there will be a definitive list when the next update cycle begins.
UPDATE 2: However, on hindsight, after looking through different Chinese sans-serif fonts (Adobe Heiti Std, SimHei), there could probably be exceptions (e.g. 山, 弓, 马, 缶, 毋母, 齿歯齒 and 㐄舛韋) to the decoration rule since my original list of affected radicals could be too burdensome. Perhaps roughly the decoration should only occur when the turning point is somewhere less than 80 degrees, not 90.
That said, I have provided an updated list of only a few radicals that need redesigning for now (and at the same time found some minor errors in this font). As stated before, all SC, TC and HK variants are affected unless stated otherwise.
专 radical 专传抟转䏝砖
丂 radical 考侤拷洘栲烤铐銬鲓鮳 丂巧窍汅朽攷 亏圬扝污杇肟釫虧夸芌咢雩疞迃 夸垮胯跨挎绔刳桍挎姱誇 粵粤澚 汚 兮枍肹盻諡
牙 radical 牙伢冴讶訝呀岈犽玡枒砑䥺釾蚜谺齖邪鸦鴉雅芽穿笌閕厊庌疨迓颬衺牚撐橕㸧 Probably map the HK glyphs to the TW font for consistency (even if out of scope): 伢厊 㸦
旡 radical All Chinese variants affected: 兂旡㤅朁僭谮譖撍噆潛憯橬熸糣鐕嶜簪蠶旤 Do not change: 兓 Only TC and HK are affected: 既旣黖
At the same time I found additional JP only glyphs in the SC/TC/HK version: 圸杣銟岲峠嵄嶃巙弖弫弬弾彁彅毎塰勄缼. I wonder if there is space to make CN versions of the glyphs, for some of them. But all of them I strongly believe are in GB18030. 嵄 (U+5D44) in particular was not given a SC glyph, and that JP glyph should have been marked for removal since it's not in Adobe-Japan 1-6. I suggest replacing JP with a new CN glyph, with or without my proposed redesign. But I will save that doomed JP glyph for the future Kangxi-neutral fork though.
The three glyphs require minor correction. Only SC affected: 疝嵆 - straighten up the 山 part. For SC/TC/HK versions, The 山 of 峇 should not have the feet sticking out.
@Marcus98T As your proposal will potentially affect TW and HK glyphs (and you've stated that the suggestion applies to TW and HK glyphs), allow me to to take the chance to clearly voice my opinion to vote against your suggestion to extend the what I call the “calligraphic touch” feature, as least in its current form, to all other stroke forms having similar turning direction (those L-like strokes) in HK and TW glyphs. Reasons are as follow:
-
Modern Chinese sans-serif typefaces tend to give a more geometric shapes to the glyphs by removing or playing down this kind of decorations. Your suggestion seems to go the opposite way.
-
Such a decoration is basically non-existent on most of the stroke types you proposed in commercial Traditional Chinese typefaces to the best of my knowledge (专 is exclusive to Simplified Chinese so isn't discussed here):
Even for Simplified Chinese I am not aware of any sans-serif font that adds calligraphic touches to such an extent as you have suggested. I also have doubt if such a change is really an “improvement” to existing glyphs at all - for 互, 韓 and 弗, current glyphs clearly look better balanced.
-
Source Han Sans is used as the default UI font in many Linux based OSes and Android phones. It is crucial for the font to be sharp and clear at smaller point size. Your suggestion will risk adding "noises" to the glyphs, as it affects small components.
-
The design of the decoration adopted in Source Han Sans isn't easy to get right, and can get out of control quite easily. Even for the current redesign version of 辶/廴 component I can easily spot rooms for improvement.
By the way, the current design approach of Source Han Sans is similar to that of 方正悠黑 (SC, released in CN) and 蒙納祥鶴黑體 (Monotype XiangHe Hei, TC, released in HK):
I agree to iron out the inconsistencies of the stroke forms of the same component, but disagree with your suggestion as if there's something fundamentally wrong on the design which'd be fixed by adding the decoration everywhere, especially to the HK/TW specific glyphs.
And, for what it's worth, what you are suggesting is quite close to what I saw in Simplified Chinese publications in 1980s (the actual publish date would be earlier). Which is best illustrated via the following font sample:
The year of these fonts can be best guessed by their 襯線 (serif-ness) and 喇叭口 features. They look outdated to my eyes after comparing them with the publications using modern Monotype/Dynalab fonts released in mid 1990s. Even these fonts don't employ decorations to all L-like strokes like in 母/互/山/㐄 etc.
@tamcy Thank you for your constructive feedback. I will try to get them to fix the inconsistencies in 旡, but I insist that characters with the 专 radical (just that one radical) would need to be redesigned as the sharp angle of 71 degrees is just too obvious.
Updated final list:
In addition, the small bopomofo characters ㆴ (U+31b4) ㆵ (U+31b5) ㆶ(U+31b6) should be redesigned to match their corresponding characters ㄅ(U+3105), ㄉ (U+3109) and ㄎ(U+310E).
Add decoration: 专传抟转啭䏝砖 (list may not be exhaustive). No more changes. I will leave it to Dr Lunde to decide whether to accept or not.
However, upon closer inspection, 专啭 actually need to have a slant at the top stroke to standardise with the others which have the slant, which should have fallen under glyph correction.
Standardise with the rest (SC glyphs only, TC/HK not affected; remove the decoration): 兓既旣黖
So if they were to only fix the CN version they would sure add even more unnecessary glyphs; now this is where I can see why my proposal can be very burdensome. Anyway, I will drop my proposal for all other radicals, except for 专.
@tamcy @Marcus98T When you discuss the "decorations" of these strokes, do you both realize the difference between SZ 豎折, ST 豎提 and PZ 撇折 and how they were rendered in calligraphy? Not all L-like strokes are equal, and I think the current Source Han Sans approach works fine. Having a "decorative bevel" on a 豎折（牙 山 亏 etc） just look strange for Source Han Sans that does not employ bevel （or 喇叭）on other stroke ends. The stroke in 衣 or 收 is a 豎提, so it was okay to have a decoration.
(By the way, I'll stop here. Dr. Lunde would probably not like when we make a lot of discussions or arguments here.)
@tamcy @Explorer09 At that time I was proposing the decoration, no, I have not considered the different stroke types. Thanks for the insight, but you didn't mention that 专 is a SZP 竪折撇 type, so while it is a derivative of SZ 豎折, it has an angle of some 71 degrees which I think should have the decoration.
I have to admit modern Chinese fonts like PingFang and Hiragino Sans GB have the sharp angle while the 80s style fonts have the decoration (although they only apply to 专 alone). I will leave it to Dr Lunde to decide whether or not to add the decoration to characters with the 专 part as per my final list above, or maybe just 专 alone.
And with that said, I promise no further changes from now on.
I do not know if it's going to be relegated to v3, but I have another suggestion so that there will be more sharing of glyphs.
And with that said, I promise no further changes from now on.
What I said was only related to the earlier attempt to add the decoration to 专. My other proposed changes have nothing to do with it. With that, let's begin.
For the 戶 part in the JP and KR versions, can the top left decoration be removed, so that they can be safely used with TW glyphs, and then those unnecessary TW glyphs can be removed? I tried finding the exact original issue where someone else brought it up, but was unable to find it so far.
I think the balance and aesthetics of the JP/KR glyphs can be brought over to TW glyphs if the top left decoration, as circled above, is removed.
EDIT: Forget it. Dr. Ken Lunde said it was a non-starter.
For 人, can the JP glyph be adjusted to fit the balance of the CN glyph (on the left)? It probably will fall under correction for this one.
(I know the unifying of the 人 roof is underway with the JP form prevailing, but not until v3 I believe)
@Marcus98T Both overall design adjustments—to unify 戶 and similar shapes, and to unify 人 when used as a roof-like component—were already rejected by the typeface designer.
Oh. I see. But which typeface designer? The Chinese or the Japanese (Ryoko Nishizuka) one? Sorry to bring this up, but I need to clarify.
The latter. There are actually some shapes that I personally would like to unify, but she rejected those, too. And that is okay.
Suggest to tweak the Heavy master of uni6E96-CN for 準 (U+6E96) so that the horizontal strokes in the 隹 component are evenly spaced visually.
@tamcy And in addition, as mentioned before, please remove the protruding bottom left feet from the 隹 part, so that it basically matches the balance and proportions of the JP glyph. Actually your issue should be put under glyph correction.
@Marcus98T
First, for most of the CN/TW/HK glyphs, there exists a foot when 隹 is above other components. So the CN/TW/HK glyphs are mostly consistent in this sense (I believe you are already aware of this). While I think it is a good idea for the CN/TW/HK glyph set to stay closer to design choice of the JP/KR glyph set whenever possible (well-established regional community preference should not be ignored tough), your proposal should either be accepted or rejected as a whole. It would be unwise to only remove the foot for one single glyph, as this will break consistency.
Second, allow me to quote @kenlunde from the redesign issue of the Source Han Serif font (emphasis mine):
This issue is meant for tracking and submitting suggestions for redesigning glyphs, meaning that the glyphs are technically correct in structure, but could benefit from adjustment in order to become more usable for more languages or regions, or simply for aesthetic reasons.
I did report an issue in the Redesign suggestion thread that turned out to be a glyph error, and I did wrongly post a Correction suggestion to this thread. But 準 U+6E96 probably isn't one of them.
@tamcy
It would be unwise to only remove the foot for one single glyph, as this will break consistency.
I have already done my homework and linked to a whole list of 隹 glyphs that are affected (which fall under the correction section). Just read through and you will see them.
https://github.com/adobe-fonts/source-han-sans/issues/204#issuecomment-500100007
Under this section (copy the quoted text, bring up the find bar via Ctrl+F and paste it there):
Then finally, for the 隹 component in all CN, TW and perhaps HK glyphs, the feet must be removed only when 隹 is on top of ⿱, similar to the JP glyphs which have no feet, so as to remove the noise when viewed on lower resolution screens.
In fact I had already linked it on the text “as mentioned before” in my previous comment.
But as to where to put your issue, I believe that the 準 glyph have a minor error in the proportions, thus could fall under correction, but who knows.