Character Name Aliases

The names of Unicode characters are intended as stable identifiers and can therefore never be altered once assigned, even if they are found to be misleading or flat out incorrect. In case of egregious mistakes, these code points are given formal name aliases that can serve as equally stable alternatives so that documentation does not have to reference defective names to uniquely identify characters. This alias concept was later expanded to cover other applications as well. Names, name aliases, and named sequences share a common namespace and as such are guaranteed to be globally unique.

There are five types of aliases:

correction

Fixes for severe, unambiguous errors, such as typos or mistaken character identity

control

Names for the C0 and C1 controls, which for historical reasons do not have regular names in the Unicode Standard, i.e. their Name property value is just an empty string

abbreviation

Alternate short names, mostly for control and white space characters, that are commonly found in implementations such as regex engines

figment

A special case for U+0080, U+0081, and U+0099, which unlike the rest of the C1 set were never formally assigned control characters but did have specific semantics in a failed early draft of what would later become ISO/IEC 10646, the Universal Coded Character Set

alternate

A special case for U+FEFF, which for historical reasons is the code point for two completely unrelated characters simultaneously: Zero Width No-Break Space and Byte Order Mark

See Unicode Technical Note #27: Known Anomalies in Unicode Character Names for more information on unfortunate character names, including those that did not warrant a formal alias.

Source: NameAliases.txt