Incorrect shapes of 又 and 叉 parts (T Chinese)

[ButTaiwan]

http://www.edu.tw/FILES/SITE_CONTENT/M0001/BIAU/f04.htm?open

According to Taiwan Standard, 又 should be closed.
When something in 又 (aka. 叉、蚤), 又 should be opened.

Glyph '又' of Source Han Sans looks correct,
however, '設' and '沒' are not well closed. (But, '役' seems correct in light weights.)

'叉' and '蚤' are wrong.

[kenlunde]

Handling 叉 (U+53C9) will be relatively easy, because the mapping can simply be changed from CID+12280 (uni53C9-CN) to CID+12279 (uni53C9-JP). For 蚤 (U+86A4), the Traditional Chinese glyph, CID+36256 (uni86A4-TW), will need to be tweaked.

[ButTaiwan]

more other characters with 又 or 叉...

娶趣輟啜�桑 should be closed. (need to be fix)
釵杈扠 should be opened. (need to be fix)

奴取授叟瘦搜 are not closed well. (better to be fix)

[ButTaiwan]

more research here.

Here are characters containing 又 or 叉 in Big5 range (without HKSCS).
And maybe there are more other characters I lost.

All glyphs with 叉 are wrong. Only 5 code point can be fixed by mapping to JP one simply (blue ones).
All glyphs with 蚤 are wrong. Due to the glyph of 蚤 is JP is different, new glyphs should be made.

All glyphs containing 又 are wrong when second stroke of 又 is a long dot (ヌ).

Some glyphs containing 又 are well closed, and some are not.
It's complicated that some character looks correct in ExtraLight and Light, and wrong in Bold / Heavy.

Colors:
Black - correct
Red - wrong
Brown - ambiguous (maybe correct maybe wrong)
Blue - JP form (usable in TW)
Pink - JP form (unusable in TW)

Tags:
TW - Source Han Sans TW Light
TW2 - Source Han Sans TW Medium
JP - Source Han Sans JP Light
MSJH - Microsoft Jhenghei (as an example of MOU standard here)

all characters are here. (order is changed.)

叉釵杈扠衩汊
蚤搔瑵颾溞慅糔

又友馭燮
叡壑
亟極殛
受綬授閿
度渡踱鍍喥
反叛版板阪返粄破鈑舨昄販皈飯扳
皮玻岥坡陂被皾皺皸皰皽皯
波婆碆菠啵
夃盈楹
及昅級伋吸扱极彶觙忣
服報赧蕔棴
殳投杸設疫股毆轂骰毅殽毄毇羖殶祋軗鑿
般搬瘢槃磐盤媻幋瀊縏搫褩鞶
殷慇
殼嗀榖彀瞉豰縠瀫瀔鷇螜愨
殺樧鎩摋
毀燬譭檓
殿壂臀癜
繫擊罊轚蘻蟿墼
醫翳贀鷖瑿毉嫛黳繄蘙
段葮緞鍛椴腶碫
侵葠浸埐
發潑撥廢癈櫠襏蹳鏺醱鱍
支技吱汥岐枝衼忮妓歧跂鬾
鼓鼜鼕鼛鼘瞽鼞鼚鼙
敲敥敁敼闅
寇蔻滱
叔淑埱椒菽督惄
假遐暇蝦瑕徦椵煆嘏婽猳犌貑瘕
叟瘦搜獀醙溲嗖瞍餿颼廋鎪騪鄋
爰援媛湲愋緩暖鶢
取娶趣聚棸諏陬棷踙焣鯫
最嘬撮蕞
桑輟惙蝃掇腏畷餟棳裰綴錣
圣怪
奴帑怒努駑孥笯弩砮
隻愯雙艭
獲護臒濩蠖韄嚄頀
矍戄攫
曼慢漫蔓嫚幔縵謾僈墁鏝熳獌
堅婜腎豎掔蜸緊賢鋻
餐粲韰瀣
聲馨罄磬謦毊韾漀
亂辭

[tomoe-musashi]

i did double check it , the glyth of 返 and 皯 is definitely wrong
the 又 of 度, 杸, 設, 盤, 蟿, 緞, 浸, 奴, 嫚, 幔, 縵, 謾, 墁, 鏝, 熳, 獌, 縵, 緊 also need to be tweaked in a minor way

[kcwu]

more 又叉 characters you missed in big5 range.
俶剟剫叢呶唚啜嗓垼娵嫂寂寑寖寢廄弢彏役怓慳拏掫擭敪梫楥槾檕歠歿毃毈毲汳沒漃澱煖燦玃琡瑗璨瓁瘙癹眅矱礥祲穀穫笝箙綅緅罬艘芟菆菔菣萲葭蔎蔱蕸薍藂藙蘉蝂蝯蟃蠮蠼裻褑覶觳諔諼謏謢豭貜赮踧躞躩郰鄤鄹酘醊鋄鋟鍜鍰鏗鑊钁雘霞顙饅駸騢騷驟鬘鰹鰻鵩鵽麚黀齱

[yuukiyuuki]

I don't know why 又 should be closed but 叉、蚤 should be opened. All these characters are 又 in fact. It is non reasonable. When the last stroke of 又 changed into 點 from 捺, it is ugly if we have to make it as closed. Maybe it is acceptable in handwriting, but not in the Hei (Gothic) style.

[ButTaiwan]

I totally agree that it's ugly and non reasonable.
However, the rules are expressly provided in the MOE standard.
And, following the standards correctly is a major premise of Source Han Sans.

[kenlunde]

I just prepared a file that I will be using to resolve this issue.

[kenlunde]

I finally had a chance to wade through this issue, and concluded that while we will eventually address this issue, it won't happen in the pending Version 1.002 update. It is likely to be addressed in the Version 2.000 update, which will make several adjustments to the glyph set.

I found that remapping existing glyphs will fix only four of the characters. To address this issue completely, some new (TW) glyphs will be needed, but I also want to explore the possibility of modifying existing CN (or JP) glyphs to address this issue. I will need to consult with our designer and SinoType's designers about that possibility.

[kenlunde]

Consolidated with Issue #99.