Translation is difficult at the best of times. I thought it was interesting how Google Translate seemingly kept coming up with different translations for the name of the program. Under “Features”, it suddenly decides the name is “Takigami”, as one example. By the end, it even goes so far as to say: “When you start up the cooking program, the following screen will be displayed.”
The translation seemed largely consistent with what Google Translate provided, but some of ChatGPT’s translation differences seemed more plausible to me, and it certainly reads more coherently. It also doesn’t keep forgetting that it’s dealing with the proper name of the program.
I didn’t try Gemini for this, but I imagine it has to be decent at translation too, so I wonder if/when Google will use Gemini to assist, replace, or otherwise complement Google Translate.
As a note, for Japanese text deepl is widely used even by Japanese people. From eng to jpn it may not choose properly nuanced words though, but it largely produces acceptable translations.
Japanese is my favourite written language, I love English but I'm definitely jealous of the beautiful glyphs and the vertical writing. From what I've seen, vertical writing is often poorly supported in software though, which is a shame.
That is quite beautiful. Thanks for sharing. Are you familiar with Japanese grass script? It has quite a different feel than the Mongolian, but it's a type of Japanese cursive that flows really nicely IMHO:
which is actually a fairly legible example. Admittedly, the more flowing styles that you see in old poetry and the like effectively require specialized training to read. Beautiful, though!
Haha can you read Japanese though? It's beautiful for sure and it even feels a little different when reading it as if you're, in a way, sorta sounding about pictures. But man is it a pain in the butt to learn!
Same, kanji is also the hardest for me, I have a much easier time learning new words by sound/hearing. But, I know some people that are the complete opposite & can't learn enough kanji, ymmv
English had beautiful writing, but it was destroyed by technology. First by the printing press, then by typewriters, then by low-resolution computer monitors. All of the human character and calligraphic qualities of the script have been mechanically stripped away in order to better accommodate what are now outdated legacy technologies, but everyone is so used to the status quo that we don't even realize what we've lost, and instead just accept that English script happens to be uglier than Japanese or Arabic or Devangari. In an alternate universe, we could be reading this in a script reminiscent of, say, the Uncial script used in the Book of Kells (which is what inspired Tolkien's beautiful Tengwar script).
Alternatively, we've made things a lot more legible for people from different backgrounds to understand. Deciphering modern fonts is a lot easier than deciphering cursive script, and hand-writing complicated scripts just raises the barrier of entry for people to communicate in written form
Deleted Comment
https://github.com/mitoma/kashiki2?tab=readme-ov-file#%E5%AE...
This kind of effect works especially well for Japanese, with its curved strokes inside square character boxes.
https://github.com/mitoma/kashiki2/blob/main/doc/assets/psyc...
https://github.com/mitoma/kashiki2/blob/main/doc/assets/ar-m...
Also, hotkey(s) for a set of predefined isomorphic camera views would seem useful; maybe I am not seeing it? https://github.com/mitoma/kashiki2/blob/main/kashikishi/asse...
I asked ChatGPT 4o to translate the README: https://chatgpt.com/share/6700bed9-1198-8004-8eed-07f5055d07...
The translation seemed largely consistent with what Google Translate provided, but some of ChatGPT’s translation differences seemed more plausible to me, and it certainly reads more coherently. It also doesn’t keep forgetting that it’s dealing with the proper name of the program.
I didn’t try Gemini for this, but I imagine it has to be decent at translation too, so I wonder if/when Google will use Gemini to assist, replace, or otherwise complement Google Translate.
Deleted Comment
https://en.m.wikipedia.org/wiki/Mongolian_script
https://ja.wikipedia.org/wiki/%E8%8D%89%E6%9B%B8%E4%BD%93#/m...
which is actually a fairly legible example. Admittedly, the more flowing styles that you see in old poetry and the like effectively require specialized training to read. Beautiful, though!
https://www.cjvlang.com/Writing/writmongol/websitesinmongolb...
Example:
http://khumuunbichig.montsame.mn/index.php?home
At first I thought it was a descendant from Arabic, but a Wikipedia detour shows that the most common ancestor is actually Aramaic script.
でもやはり高低アクセントはもっと苦しいと思います。あれは無理ですww
For instance a lot of the obvious brush stroke is gone such as: うえらおや
The upper line is supposed to either look like a droplet of water or like ふ upper part
Certain details are gone: にこ no longer has the half arrow you can see exists vertically on the horizontal line.
ふ lost a lot of its details.
Still I would argue it looks better now for the most part.
Source: https://gogonihon.com/en/blog/japanese-characters/
Why English in particular? English uses Latin script pretty randomly.
-- sincerely your Ptoughneigh