According to the MQM and TAUS DFQ quality metrics in localization, error categories are divided into the following 8 branches or “dimensions”:
Accuracy: issues that arise from the relationship of the TT to the ST, such as omissions, additions, mistranslations, over-translation, under-translation etc.
Fluency: issues that arise from the internal rules of the TT structure, such as grammar, spelling, inconsistencies, etc.
Verity: issues that arise from relationship of the TT to the external world, such as culture-specific elements, suitability, completeness, etc.
Design: issues that relate to the physical presentation of the TT, such as hyphenation, truncation, formatting, etc.
Terminology: issues that relate to the use of specific terminology, such as consistency with termbases, etc.
Style: issues that are closely related to the fluency category and deal with the register of the text.
Internationalization: issues that relate to the preparation of the source content for subsequent translation or localization.
Locale convention: issues that relate to locale-specific conventions, such as address format, postal code, date and number format, etc.
For those of us who are familiar with the more general – and in my opinion more logical – error categories (accuracy/meaning, grammar/syntax, spelling, punctuation, terminology/vocabulary, register/style and readability/naturalness), we will definitely have a hard time identifying whether a “missing step of a technical procedure” should be categorized under “Accuracy”, since omission is one of its sub-categories or under “Verity”, since this dimension addresses issues of completeness.
More information: QT21 and MQM issue types
