UTF-8 is a multibyte character encoding of the Unicode character set, made up of one or more bytes. Unlike some other encodings such as UTF-16, the UTF-8 encoding is upward compatible with 7-bit ASCII characters, and can be processed to some degree by applications that are only aware of bytes.
Full support of UTF-8 for searching, collation, word parsing, etc, does require support of Unicode concepts such as characters, normalisation, supplementary characters, etc. Many application and OS problems with "special characters" such as accented European letters, or ideographs such as used in Japanese or Chinese, derive from mismatched character encodings.
Related tags: unicode