Unicode 18.0.0

28 points by EvanHahn

Finally we have a pickle emoji!!!

https://www.unicode.org/emoji/charts-18.0/emoji-released.html

lake

and a monarch butterfly! genuinely exciting.
reezer

But where is the seahorse emoji? That would drive LLMs insane and probably evil for good. ;)
oats

I wonder how fonts will distinguish between the existing cucumber🥒 and the new pickle.
fs111

Still no potatoe-chips emoji :(

sinic

Happy to see that most of the issues of Egyptian hieroglyphs in Unicode 16 raised here have been resolved: https://nederhof.github.io/newgardiner/unicode5to16corruption.html

hongminhee

One thing about this release bugs me: the new Seal block (小篆, Small Seal script) gets its own Script property and its own U+3D000..U+3FC3F range, kept separate from Han. Yet it's a direct ancestor of modern Chinese characters, and most of its roughly 11,000 characters have modern counterparts.

Simplified and traditional Chinese routinely differ in abstract shape too (愛 vs. 爱), and Unicode gives them separate code points for exactly that reason, per the basic principles of Han unification. They still stay inside Han, linked through Unihan's kSimplifiedVariant/kTraditionalVariant fields rather than pushed into a new script. The 2014 Small Seal proposal (PDF, L2/14-242) argues the opposite: because many Small Seal glyphs don't map to modern Chinese characters one to one, and the ones that do map have different internal structure, the whole script “shall not be regarded as a typeface of modern Hanzi” and needs independent encoding.

The difference is one of degree. Partial non-mapping plus structural divergence already shows up in plenty of unified Han variants, just less severely. The “own script” line for Small Seal looks less like a technical requirement than a precedent already set by Khitan, Jurchen, Tangut, and Nüshu: sinoform writing systems treated as separate scripts once they're old and divergent enough. A WG2 contributor floated merging Khitan into CJK Unified Ideographs back in 2010 and got pushback along exactly these lines (N3934).

Oracle Bone Script (甲骨文) diverges from modern Chinese characters far more than Small Seal does. It's had a pending proposal since 2015, with tentative code points reserved in Plane 3, and still isn't in the standard as of 18.0. Divergence from modern Chinese characters doesn't predict either the script boundary or which proposal gets finished first.

gerikson

the basic principles of Han unification.

The linked fragment URL (https://en.wikipedia.org/wiki/Han_unification#Basic_principles) doesn't seem to exist anymore.

Han unification seems contentious[1], which might explain why later proposals for Small Seal script decided to sidestep it entirely.

[1] Assume British understatement here
- masklinn
  
  Yeah I suspect the practical consequences of unification, alongside the disappearance of its primary purpose (compression) has shifted the justification burden for splitting versus unifying to favor splitting.
- hongminhee
  
  Good catch, thanks. Fixed it to link the actual Unicode unification rules instead: https://www.unicode.org/versions/Unicode18.0.0/core-spec/chapter-18/#G11484.