WEBVTT 00:00:00.000 --> 00:00:05.760 Imagine a keyboard with thousands of keys because the language you speak has thousands of characters. 00:00:05.760 --> 00:00:09.840 It turns out the most natively spoken language in the world, Mandarin Chinese, 00:00:09.840 --> 00:00:15.600 does have thousands of characters. But you can't exactly make a keyboard with 20,000 keys. 00:00:15.600 --> 00:00:20.000 So how the heck do our Chinese speaking friends enter characters on their PCs and phones 00:00:20.000 --> 00:00:24.240 without pulling their hair out in frustration? There are several common and efficient ways 00:00:24.240 --> 00:00:30.720 to do this, but it wasn't always so. Back in the mid-1970s, before word processing was widespread, 00:00:30.720 --> 00:00:35.520 there was a rather popular typewriter called the double pigeon. But unlike the typewriters 00:00:35.520 --> 00:00:42.320 you're probably used to seeing, it didn't even have keys. Instead, it had well over 2,000 pieces 00:00:42.320 --> 00:00:47.760 of movable type, with each piece containing either a character or part of a character. 00:00:47.760 --> 00:00:52.640 To pick one, you'd have to move a lever above the character you wanted and then press down, 00:00:52.640 --> 00:00:56.560 which would cause the ARM to flip up and punch the character against the paper. 00:00:56.560 --> 00:01:02.480 Although the pieces of type were arranged in a logical fashion, it was still quite tedious to use. 00:01:02.480 --> 00:01:08.080 But it was around the same time, 1976 to be precise, that inventor Qiubong Fu developed 00:01:08.080 --> 00:01:12.480 what's called the Kang Ji input method for keyboards. And if you're wondering why it's 00:01:12.480 --> 00:01:16.880 called that, it was named after the mythological inventor of the Chinese writing system. 00:01:16.880 --> 00:01:21.920 Anyhow, the Kang Ji system was notable because it used a standard QWERTY keyboard, 00:01:22.000 --> 00:01:26.880 alleviating concerns about whether the post-industrial age would end up leaving Chinese speaking 00:01:26.880 --> 00:01:31.680 areas behind because their language was one that just couldn't quickly be typed out. 00:01:31.680 --> 00:01:36.080 Only a few years after Chu's invention, he released his patent rights, meaning the 00:01:36.080 --> 00:01:40.320 Kang Ji system became open source, making it very popular to this day. 00:01:40.320 --> 00:01:45.600 But consider this, even if you're not trying to create a keyboard featuring every Chinese 00:01:45.600 --> 00:01:50.320 character in existence, it's estimated that an educated Chinese speaker knows between 00:01:50.400 --> 00:01:56.880 3 and 8,000 characters. So how can you reduce all that down to just 87 keys? 00:01:56.880 --> 00:02:01.280 I'm not quite sure why anyone would think this, but Chinese characters aren't just constructed 00:02:01.280 --> 00:02:06.320 at random. Rather, they're often made up of elements sometimes referred to as radicals. 00:02:06.320 --> 00:02:11.600 On a Kang Ji keyboard, each Latin character matches up to a common radical. They're sorted, 00:02:11.600 --> 00:02:16.000 according to the order of the Latin alphabet. For example, the keys for A through G 00:02:16.000 --> 00:02:20.640 roughly correspond to classical elements of nature, such as water or fire. 00:02:20.640 --> 00:02:25.040 And once you've decided which character you want to type, you build them up from these radicals, 00:02:25.040 --> 00:02:30.080 going from top of the character to the bottom, left to right, and outside to inside, 00:02:30.080 --> 00:02:35.120 meaning that you can construct a large number of characters from three or four key sequences 00:02:35.120 --> 00:02:39.920 called Kang Ji codes. Of course, these days, there are plenty of software-based tools to 00:02:39.920 --> 00:02:44.640 make typing easier for the average Chinese speaker. Similarly to how English speakers will get 00:02:44.640 --> 00:02:48.960 suggestions for what word they want as soon as they press the first key, Chinese speakers will 00:02:48.960 --> 00:02:54.400 get suggestions based on which radicals they punch in on both PCs and on phones. Which is 00:02:54.400 --> 00:02:59.280 especially useful, as there are plenty of Chinese words that are composed of multiple characters. 00:02:59.280 --> 00:03:04.480 Kang Ji has been supplanted in popularity by other methods such as WuBi, a similar but 00:03:04.480 --> 00:03:09.760 faster shape-based system more geared towards simplified Chinese, and Pinyin, which involves 00:03:09.760 --> 00:03:14.320 typing in a romanized version of a Chinese word, which is the version that I've been saying the 00:03:14.400 --> 00:03:20.000 whole time in the script, and having software convert it to a Chinese character. But Kang Ji 00:03:20.000 --> 00:03:24.960 remains widespread and marked an incredibly important milestone in Chinese technological 00:03:24.960 --> 00:03:29.200 development. When you think about it, it's not all that different from how we can form tens of 00:03:29.200 --> 00:03:34.560 thousands of English words from just 26 letters instead of having keys that say things like 00:03:34.560 --> 00:03:40.800 coconut, jerky, or attractive YouTube post. I didn't write that. That was just in there. 00:03:40.800 --> 00:03:44.000 Hey, that was a whole tech wiki. Thanks for watching. Like the video if you liked it, 00:03:44.000 --> 00:03:47.600 dislike it if you disliked it, check out our other videos, comment below with video suggestions, 00:03:47.600 --> 00:03:51.600 and don't forget to subscribe and follow for Pete's sake.