> We could dig a big further and ask why computer were able to handle alphabet characters before hiragana for instance. JIS X 0201 was released in 1969 and wid

> We could dig a big further and ask why computer were able to handle alphabet characters before hiragana for instance. JIS X 0201 was released in 1969 and widely adopted on mainframe systems in the 1970s. It supports a modified form of Katakana (“half-width”) as well as the Latin alphabet in a 7 or 8 bit encoding. 8-bit encodings supported both, for 7-bit you could switch between Katakana and Latin using control characters. So, by the 1970s you could already do Japanese language computing if you limited yourself to katakana and romaji. Kanji took longer because there are too many characters for a 7/8-bit code and their visual complexity required more advanced display hardware, but already in 1971 IBM was selling their “IBM Kanji System” which contained specialised software, printers and keypunches to support Kanji on IBM S/360 mainframes. In 1979 IBM added support for 3270 terminals (with additional Kanji circuitry) to enable Kanji in interactive as opposed to batch mode punched card computing. Early systems without kanji support generally used katakana rather than hiragana, because if writing everything in katakana looks weird to Japanese speakers, doing the same thing in hiragana looks even weirder. Also, the “blockier” shape of katakana makes it more legible than hiragana on low resolution displays. > Or why no EU company came up with an OS that could fight Mac OS or Windows. In the 1990s, Oracle licensed RISC OS from Acorn in the UK, and sold it in the US as Oracle NCOS (Network Computer Operating System). But the whole “network computer” thing was never as successful as its backers had hoped.