CJKV Information Processing, 2nd Edition

8

|

Chapter 1: CJKV Information Processing Overview

Encoding Methods

Encoding is the process of mapping a character to a numeric value, or more precisely,

assigning a numeric value to a character. By doing this, you create the ability to uniquely

identify a character through its associated numeric value. e more unique a value is

among dierent encoding methods, the more likely that character identication will be

unambiguous. Ultimately, the computer needs to manipulate the character as a numeric

value. Independent of any CJKV language or computerized implementations thereof, in-

dexing encoded values allows a numerically enforced ordering to be mapped onto what

might otherwise be a randomly ordered natural language.

While there is no universally recognized encoding method, many have been commonly

used—for example, ISO-2022-KR, EUC-KR, Johab, and Unied Hangul Code (UHC) for

Korean. Although Unicode does not employ a single encoding form, it is safe to state that

the encoding forms for Unicode—UTF-8, UTF-16, and UTF-32—have become univer-

sally recognized. In addition, each one has become the preferred encoding form for spe-

cic uses. For the Web, UTF-8 is the most common encoding form. Applications prefer to

use the UTF-16 encoding form internally for its overall space-eciency for the majority

of multilingual text. OpenType fonts, when they include glyphs that map from ch