ISO-IR-111^[1] orKOI8-E^[2] is an 8-bit character set. It is a multinational extension ofKOI-8 forBelarusian,Macedonian,Serbian, andUkrainian (exceptҐґ which is added toKOI8-F). The name "ISO-IR-111" refers to its registration number in theISO-IR registry, and denotes it as a set usable withISO/IEC 2022.

It was defined by the first (1986) edition of ECMA-113,^[3] which is theEcma International standard corresponding toISO/IEC 8859-5, and as such also corresponds to a 1987 draft version of ISO-8859-5.^[4] The published editions ofISO/IEC 8859-5 instead correspond to subsequent editions of ECMA-113, which defines a different encoding.^[5]

ISO-IR-111, the 1985 edition of ECMA-113 (also called "ECMA-Cyrillic" or "KOI8-E"), was based on the 1974 edition of GOST 19768 (i.e.KOI-8). In 1987 ECMA-113 was redesigned.^[5] These newer editions of ECMA-113 are equivalent toISO-8859-5,^[5][6] and do not follow the KOI layout. This confusion has led to a common misconception that ISO-8859-5 was defined in or based on GOST 19768-74.^[6]

Possibly as another consequence of this,RFC 1345 erroneously lists a different codepage under the names "ISO-IR-111" and "ECMA-Cyrillic", resembling ISO-8859-5 with re-ordered rows, and partially compatible withWindows-1251.^[7][6] Due to concerns that existing implementations might use the RFC 1345 definition for those two labels, it was proposed that theIANA additionally recogniseKOI8-E as a label for ECMA-113:1985 content,^[7] and the IANA presently lists that label as an alias.^[2]

The following table shows the ISO-IR-111 encoding. Each character is shown with its equivalentUnicode code point.

A modified version named KOI8 Unified orKOI8-F was used in software produced by Fingertip Software, adding theҐ in itsKOI8-U location (replacing thesoft hyphen and displacing theuniversal currency sign), and adding some graphical characters in theC1 control codes area, mainly fromKOI8-R andWindows-1251.^[4][6][8][9]

RFC 1345 erroneously lists a different code page under the name ISO-IR-111, encoding the same Cyrillic characters but with a different layout. It resembles a mixture ofWindows-1251 andISO-8859-5.^[7] Specifically, line A_ corresponds to ISO-8859-5, lines C_ through F_ correspond to Windows-1251^[6] (equivalent to lines B_ through E_ of ISO-8859-5), and line B_ nearly corresponds to line F_ of ISO-8859-5, with the exception of the§ being replaced with a¤.

Certain codes resemble ISO-IR-111 with flipped letter case, which may have contributed to the confusion. The majority differ and are shown below.

• KOI character encodings

• Character sets
• Cyrillic script

• Articles with short description
• Short description is different from Wikidata