I think this is a rare “useful” abuse of font ligatures, and in that respect I think this is a cool project, but it's also a terrible idea that nobody should actually use, because Japanese kanji readings are famously ambiguous and context-dependent. The modal character has two different readings without counting those used in personal names, nor contextual slight changes to pronunciation. Some characters have six different readings, and that's again without counting those used in personal names. There's no way a font can do a good job at it unless you can somehow stuff an entire LLM in there.
For context, it's a hard enough problem that Google Translate's own romanised pronunciation guides are frequently wrong, or disagree with their text-to-speech.
Really nice font! Native Japanese speaker here, and seeing furigana done purely in the font is very cool.
One small thing on the Japanese version of the page: in the "Ligatures do the work" section, 合字 itself only gets じ over 字, and 合 has no reading. It should be ごうじ. I guess 合字 isn't in the dictionary, so it falls back to single kanji, and 合 is skipped because it has several readings.
Automatic furigana is not really possible to be reliable considering readings heavily rely on context that can't be easily inferred, but I'm sure it will help some people who can't read very well.
You can run LLMs inside the shaping engine, so maybe you can use more advanced segmenting and transliteration algorithms? https://fuglede.github.io/llama.ttf/
This is very cool. Unfortunate that it gives the wrong pronunciation when words get split in a line wrap (e.g., 大学 got split between lines and changed to incorrect furigana for me).
This is really nice. I'm able to read Hiragana and Katakana fully after about 4-6 weeks now but then I'm hit with the reality of how many Kanji I have to learn still everytime I want to read something in real life lol
Maybe this is less dependent on rendering engine or easier to use. Or maybe because it could be done and was done.
From practicality standpoint, this would not work on unprepared texts, as pronunciation of Japanese words are a bit arbitrary and so correct pronunciations would need to be manually selected for each words.
i.e. "maize" could be read "corn". I believe LLMs don't necessarily have good intuition for that, being trained on plain texts.
As to how could such a situation could ever manifest: I would blame the fact that the official language of historical Japanese governments having been a really creative interpretation of medieval Chinese[1] sprinkled with labels and jumps and supplements[2][3], resulting in people just accepting wild divergence between literal and spoken Japanese. I'm not a linguist or a language major so I'd happy if anyone could further elaborate...
When you control the content, I agree that ruby tags would be better, since they can easily handle alternative readings and also enable smarter automation to infer the correct reading from context.
But you could set this as the default system font for Japanese and enjoy furigana everywhere, even in places that don't support ruby tags at all.
That’s incredible. I’m always impressed with the utility of the TrueType/open font spec - the flexibility there is really an underappreciated work of engineering genius.
I mean, fonts are Turing-complete [0], so it isn’t surprising that this is possible. Turing-completeness is often by accident rather than by genius [1].
Which makes sense, because human language is apparently Turing complete, so anything intended to represent it faithfully would wind up with that characteristic, too.
Bitmap fonts aren't turing complete. The fact that TrueType is built with an arbitrary virtual machine is insane, makes implementations ridiculously complex, and is a huge security liability. The purpose of the VM wasn't to "represent human language faithfully" but to make fonts look a little better when displayed at a variety of resolutions, via hinting. Hinting doesn't necessarily require a full VM, but somebody let the engineers have a little too much fun.
Human language isn’t a formal language, so this is a category error.
In the present context, an example is that the common reading of a Kanji word isn’t always well-defined, and neither is the decomposition of a sequence of Kanji characters into words.
It's an interesting move to point me to a rigorously written English language definition of a formal language to make the argument that humans don't co-opt our natural languages as formal languages when the need arises.
For context, it's a hard enough problem that Google Translate's own romanised pronunciation guides are frequently wrong, or disagree with their text-to-speech.
One small thing on the Japanese version of the page: in the "Ligatures do the work" section, 合字 itself only gets じ over 字, and 合 has no reading. It should be ごうじ. I guess 合字 isn't in the dictionary, so it falls back to single kanji, and 合 is skipped because it has several readings.
From practicality standpoint, this would not work on unprepared texts, as pronunciation of Japanese words are a bit arbitrary and so correct pronunciations would need to be manually selected for each words.
i.e. "maize" could be read "corn". I believe LLMs don't necessarily have good intuition for that, being trained on plain texts.
As to how could such a situation could ever manifest: I would blame the fact that the official language of historical Japanese governments having been a really creative interpretation of medieval Chinese[1] sprinkled with labels and jumps and supplements[2][3], resulting in people just accepting wild divergence between literal and spoken Japanese. I'm not a linguist or a language major so I'd happy if anyone could further elaborate...
1: https://en.wikipedia.org/wiki/Kanbun
2: https://youtu.be/_k0I6bi2pYs?t=248
3: https://d1ysbfo9sxjda0.cloudfront.net/wp-content/uploads/201...
But you could set this as the default system font for Japanese and enjoy furigana everywhere, even in places that don't support ruby tags at all.
https://www.destroyallsoftware.com/talks/wat
[0] https://litherum.blogspot.com/2019/03/addition-font.html
[1] https://en.wikipedia.org/wiki/Turing_completeness#Unintentio...
In the present context, an example is that the common reading of a Kanji word isn’t always well-defined, and neither is the decomposition of a sequence of Kanji characters into words.
Maybe it is:
With the ligatures to adApt to pronunciation of words.
https://pingtype.github.io/japanese.html
It also support line breaking, so that words don't get cut in half when wrapping text.
confusing when you don't explain which limktations this is