Consolidation of Post-V2 Glyph Sharing Suggestions

[kenlunde]

The following table shows the glyphs that have been explicitly targeted for removal due to glyph sharing or other reasons:

Character	Glyph Name	Notes
U+720B 爋	uni720B-JP	The CN glyph, uni720B-CN, will be default for all regions
U+89BE 覾	uni89BE-TW	The CN glyph, uni89BE-CN, will be default for all regions (see Issue #214); remove HK mapping override to uni89BE-TW
U+24A01 𤨁	u24A01-KR	Identical to CN glyph for U+7468 瑨, uni7468-CN (see Issue #202)

[tamcy]

I think there is no need to create a separate set of glyphs for characters with the 尺 components for HK.

• Strictly speaking, this HK version of 尺 also matches Taiwan's MoE standard. Since SHS didn't use a separate set of glyphs for TW in the first place, I believe there's no need to do so for HK.
• As mentioned in the HK Reference Glyph document (P.8), shifting the "㇏" stroke a little bit is seen as a design difference, not a violation of the standard (it's not a standard, but you get me. Not sure how to call it accurately). The document uses 尺 and 家 as examples.

• SHS didn't create a separate glyph set for characters containing "家" for HK.

• For some glyphs like 咫, 沢, 蚇 and 駅 the differences are small at heavier weights.
• This HK version may look nice in script form, but the JP/CN version definitely look better for serif and sans-serif fonts.

Therefore I suggest to remove the HK-specific glyphs for characters with the 尺 components, and use the JP/CN version (where appropriate) instead.

Followings are the codepoints concerned:

Codepoint	Suggestion
U+4F2C 伬	Use CN
U+544E 呎	Use CN
U+54AB 咫	Use CN
U+5C3A 尺	Use CN
U+5EB9 庹	Use CN
U+6CA2 沢	Use JP (also for TW)
U+8687 蚇	Use JP
U+99C5 駅	Use JP (also for TW)

[kenlunde]

@tamcy I am actually inclined to map U+4F2C 伬, U+544E 呎, U+54AB 咫, U+5C3A 尺, U+5EB9 庹, and U+8687 蚇 in the TW CMap resource to the corresponding HK glyphs. Also note that there are three additional cases whose forms are consistent, and used for both the TW and HK CMap resources: U+7C8E 粎, U+9FC8 鿈, and U+26B28 𦬨. These six mapping overrides are now reflected in the table in Issue #202.

[tamcy]

@kenlunde Ah yes, I forgot to also include the codepoints without regional-specific glyphs. Thanks.

[kenlunde]

Because actual glyph removal won't happen until Version 3.000 (unscheduled), I am closing this issue, and will track such cases in Issue #227.

For Version 2.000 dot releases, such cases will be handled via mapping overrides to the appropriate CMap resources (see Issue #202), leaving some glyphs unused.