Section 4.2. Versions of Unicode

4.2. Versions of Unicode

Unicode versions are numbers much the same way as program versions, using a hierarchic number of the form m.n.p, where m is the major version number (which usually remains the same for years), n is the minor version number, and p is the update version number. For a detailed description, refer to http://www.unicode.org/versions/. The format of citing Unicode and its versions is discussed in Chapter 5.

In practice, the minor version number 0 is often omittede.g., "Unicode 4.1" instead of "Unicode 4.1.0." In this book, "Unicode" means Unicode 4.1.0 unless otherwise stated.

Unicode Version 1.0 used somewhat different names for some characters than ISO 10646. In Unicode Version 2.0, the names were made the same as in ISO 10646. However, the Version 1.0 names (such as "period" for "full stop") are still preserved as alternate names, mentioned both in code charts and in the Unicode database.

New versions of Unicode are expected to add new characters mostly, though changes and clarifications are possible. However, there is a firm policy that no characters will be removed, no code numbers changed, and no Unicode names changed. Annotations, including alternate names, and properties of characters may change.

The growth of Unicode is summarized in Table 4-2, which shows the number of characters (code positions assigned to characters, including 65 control characters) and blocks in each version. (Minor versions that did not add any new code position assignments have not been included.) The issue date is specified in year-month notation. The jump in the number of characters in Version 3.1 is mainly caused by the addition of the CJK Unified Ideograms Extension B block (42,711 characters).

Table 4-2. Unicode versions

Version	Issued	Characters	Blocks
1.0	1991-10	7,161	57
1.0.1	1992-06	28,359	59
1.1	1993-06	34,223	63
2.0	1996-07	38,950	67
2.1	1998-05	38,952	67
3.0	1999-09	49,259	86
3.1	2001-03	94,205	95
3.2	2002-03	95,221	107
4.0	2003-04	96,447	122
4.1	2005-03	97,720	142

Version 5.0 is to be published in the third quarter of 2006. Information on it is available at http://www.unicode.org/versions/Unicode5.0.0/. It is intended to add 1,365 characters, for the needs of some living languages (e.g., in India), for mathematics, and for academic use, particularly for coding cuneiform and other ancient texts.