> Unicode enabled all sorts of cool things in computing that simply were not possible before or only with awful hacks and workarounds.
Could you give an example? I don't know anything about this stuff.
Could you give an example? I don't know anything about this stuff.
That's because every byte stored in the file, for example byte number 188, either means "¼" (as it does in ISO/IEC 8859-1, aka. Latin-1 or ANSI), or it means "ỳ" (as in ISO/IEC 8859-14) or "ｼ" (in JIS X 0201, one of the many Japanese encodings that were devised over the years.)
How do you know which encoding a certain file uses? In general YOU CAN'T and this was the source of many problems and "solutions" which caused even more problems over the years.
Well then, how did you mix symbols from different alphabets, say in a dictionary or in a post that talks about them, like this very post? YOU COULDN'T, short of doing ugly hacks and other subterfuges, like using GIFs for all foreign characters.
Unicode gave a distinct number (or "codepoint") to every character and symbol known to man (within reasonable limits) and this allowed a lot of things that we take for granted nowadays, including this very post, were I just copied and pasted various symbols from their Wikipedia pages and just expect it to work.
Unicode defines normalization algorithms (is é its own character, or e with a modifier character?).
I can have a document which combines English, Russian, Arabic, and Chinese, and expect it to be readable and editable by many different tools.
English combines with top-down Chinese, and English combines with right-to-left Arabic, but top-down Chinese and right-to-left Arabic don't combine properly in the same document using Unicode -- the Arabic will be written bottom-up instead of top-down when embedded in the top-down Chinese.
I meant something simpler, like: The word 'computer' in English is 计算者 [jì suàn zhě] in Chinese, Компьютер in Russian, and حاسوب in Arabic."
Try that without Unicode.
It's of course possible with TeX, and no doubt other solutions. Which is why I added "and expect it to be readable and editable by many different tools".
(As a real-world use case, look at Knuth's "The Art of Computer Programming" and see how he credits people using their full names, in their own written language.)
計算機 is a calculator, not a computer, in Taiwan: https://tw.images.search.yahoo.com/search/images?p=計算機
Compare with Baidu in the PRC: http://image.baidu.com/search/index?tn=baiduimage&ie=utf-8&w...
Edit: Oops, I didn't even realise that the parent poster asked about 計算著. That one I've never seen.
In this way, new FAQs and other updates spread to all the markets easily.
It's vastly easier to do this stuff when all the documents use the same text encoding. Even if noone can read everything, the fact that everything uses the same encoding means that any pair of languages you can read are technically readable.