UTF-8

In the early days of the internet, opening a foreign website or downloading a document from another country often turned the text into alien gibberish. That's because a computer doesn't know human letters at all. It can only handle number signals made of 0s and 1s. So people made agreement tables that pair each character with a specific number, one to one. This is called a character encoding.

The trouble was that every country and operating system made its own table. The early table made in the United States held only the English alphabet, and Korea built its own number table just for Hangul, the Korean alphabet. If you fed the Hangul table into an American program, the computer read the characters as completely different ones and had no choice but to show garbled text on screen.

To end this mess, "Unicode" appeared: a huge digital dictionary that gives every character in the world one shared, unique number. And the widely used standard for storing and sending those characters in computer files most efficiently is UTF-8.

The biggest strength of UTF-8 is its flexible design, where the amount of data a character takes depends on the kind of character. If you put every character in the world into the same big box, sized for the largest one, you'd waste a lot of storage.

The box size depends on the number a character gets in Unicode. Letters, digits, and basic punctuation sit at the very front (numbers 0 to 127), and they fit snugly into the smallest box, 1 byte (8 bits). Characters with later numbers, like Hangul or most Chinese characters, go into a 3-byte box. And the colorful picture characters, or emoji, that people use on smartphones today are mostly stored in 4 bytes.

So UTF-8 adjusts the length to fit each character, from 1 byte to 4 bytes. Because of this, English documents can stay as light as they were under the old systems. At the same time, any language or emoji in the world can be mixed naturally on a single page.

To be more precise, the only old system that UTF-8 can read as-is, with no conversion, is the original ASCII standard. ASCII is the most basic number table for the English alphabet, digits, and special symbols, made in the United States in the 1960s.

UTF-8 was designed so that its first 1-byte range matches the old ASCII codes exactly. Thanks to this, the vast number of old text documents and system program files written in English decades ago open perfectly in a UTF-8 environment, with no changes at all.

But it's a different story for Hangul documents written in the older Korean encodings, such as the "wansung" (precomposed) style and EUC-KR. Their English letters and digits match ASCII, but they write Hangul and Chinese characters with 2-byte numbers that are totally different from UTF-8. So if you open them with UTF-8 settings, the Hangul parts turn into garbage. UTF-8's backward compatibility applies only to English ASCII, so even today, opening an old Hangul document means choosing the right character set or converting the file.