The act of matching segments or sentences in translated documents for importation as translation units or sentence pairs in a translation memory tool.
Automatic Content Enrichment (ACE)
A bridge between single language websites and localization, ACE technology associates English words and phrases on web pages with pop-ups containing information in a user’s native language.
Languages that are read right-to-left but which support left-to-right words for European terms.
The characters (and their byte codes) available in an application program. The most common is the ASCII (or ANSI) set of 128 characters, in which CAPITAL LETTER A has the code 41 in hexadecimal, 065 decimal, and 01000001 binary. Various localized character sets are ISO 8859-1 or Latin-1, the most common encoding of characters for Western European languages. ISO 8859-1 through 8859-15 are 256 character sets. The first 128 characters are the same as ASCII, and the upper 128 encode Cyrillic, Hebrew, Arabic, etc. characters.
Computer Aided Translation (CAT)
Use of a translation memory tool by a human translator. The TM tool enables the translator not to have to translate the same sentence twice. More importantly, it guarantees that translations will be identical, when consistency is a requirement for the material, and yet enables a different translation when literary quality and creativity are prized.
The act of using software to create high-quality publications that combine text and graphics in a sophisticated layout following design standards. In localization it also encompasses the act of creating and maintaining the look and feel of the original document once translated into other languages.
The glyphs or shapes that render a character on the computer screen or on a printer. The mapping between character codes and glyphs may be different on different operating systems.
Full match (100% match)
A segment or sentence in a source document for which the translation memory tool has a perfect match in the target language, including font style and various attributes like context. See fuzzy match.
Fuzzy match (partial match)
A segment or sentence in a source document for which the translation memory tool can match some of the words in the target language, including font style and various attributes like context.
Use of machine translation to provide a rough translation of text in order to determine its content.
Term describing the process of designing, developing, and adapting a product for distribution in multiple countries. G11N includes all the strategic and marketing preparation that goes into global deployment of a product.
Global Content Management System (GCMS)
A translation tool designed to automate translation of website and other content that changes frequently.
The acronym GUI stands for Graphical User Interface and is synonymous with UI.
A keyboard shortcut used to access functions in menus and sub-menus. A hotkey is usually represented by a mnemonic which is shown as an underline on the menu line; the mnemonic typically refers to the initial letter of the function (e.g. “Copy” = “CTRL+C”). (See shortcut key.)
A language-specific computer keyboard or a software tool that allows any standard keyboard to input the characters or ideographs of a language. When the symbols exceed the number of keys as do Asian languages, the software presents a selection of possibilities for clicking with a mouse.
Input Method Editor
An IME may be as simple as a localized keyboard, which emits the appropriate character codes for the local language when the keys are struck. It may be a software program that works with a standard keyboard and mouse to select character codes from an onscreen display. It may generate a character code only after a sequence of keys has been struck. More sophisticated input methods include graphic pads that recognize a character drawn with a stylus, and speech recognition systems.
The process of designing a product so that it can be adapted to various languages and locales without engineering changes.
An optional addition to an operating system that enables its keyboard and application programs to work in the character sets appropriate to the language, and to render its fonts. Note that text files produced with a Greek language kit will generally produce gibberish if read on a machine with a Russian or Hebrew language kit. Unicode is meant to resolve this problem.
A combination of a language and a particular geographic region (usually a country) where the culture is distinctive enough to merit the use of different terminology and web page design practices. Locales usually have character sets suited to representing their scripts, and custom fonts for their character set.
The act of customizing software and documentation for a particular country. It includes the translation of the GUI, including menus and messages, into the native spoken language as well as changes in the user interface to accommodate different alphabets and cultural issues.
Machine Translation (MT)
A translation tool that automatically translates text not previously seen based either on linguistic parsing or on similar text stored in a database.
Surrounding text with beginning and ending tags, typically set off in angle brackets.For example: This is bold. Translators must take care to leave the words in the tags alone. Translation memory tools that are markup aware (e.g., Trados, TagEditor) do not change tags, except for certain quoted material inside a tag’s attributes (e.g. HTML, SGML, and XML).
See full match or fuzzy match.
Refers to software that supports more than one language simultaneously, thereby allowing the end user to select multiple languages and formats. This software allows data containing multiple languages to be entered, processed, presented, and transmitted in many different locales.
Multiple Language Vendor (MLV)
A relatively large localization service provider that offers a wide range of languages and other services.
Involves the preparation of usually updated files for translation where the existing files already contain related segments of previously translated data. Only 100% matches are replaced, with the result being a set of files containing both source and target language terminology.
Text elements, especially software strings in the user interface, may expand in some languages (such as German) during translation and may no longer fit inside buttons, menus, dialog boxes, etc. Localization engineers must then resize the UI elements to accommodate the expanded text.
A part of a document, usually limited by punctuation – periods, tabs, paragraph marks, or custom tags. The segment or sentence is the fundamental unit of information stored in translation memory with its corresponding bilingual matching segment.
A keyboard combination used to access functions in menus and sub-menus. The shortcut key combination is essentially an abbreviation of the menu item command. Ctrl+V is a typical accelerator key combination for the Paste command. When a product is localized, this key combination is generally not changed, i.e. the same combination (Ctrl+V) is used in the localized version of a product. (See Hotkey.)
An abbreviation for “simultaneous shipment,” the release of multiple product versions at the same time.
Single Language Vendor (SLV)
A localization service provider that offers translation and or localization services into only one language.
The language in which the product that is to be localized was originally developed.
The language in which the product that is to be localized is converted to (e.g. from US English to German).
Software tool such as Trados MultiTerm which allows easy integration of the Termbase into the translation process.
A terminology list, usually bilingual. The input/output text files or TBX files of Terminology Managers. Usually in a format like comma-delimited or tab-delimited files suitable for use with a spreadsheet.
Some languages are more verbose and have greater average word lengths and therefore designers must leave room for the extra text space needed, in German, for example. Text expansion impacts UI design elements and should be dealt with before translation.
A translation tool that stores text segments (usually sentences) and their translations in a database and automatically retrieves translations for text that is already in the database (usually from a previous version of the text). The tool may also find similar segments and their translations to assist the translator.
Translation unit (Trados)
A segment or sentence pair in source and target languages.
The 16-bit standard capable of encoding the characters of the world’s major language scripts. It is designed to be a universal character set. Version 3.0 contains 49,194 characters and 8,515 code points for private uses and future expansions. The limit to 16-bit encoding is 65,536 characters. Unicode is supported on all the major computer operating systems, as well as by HTML 4.0, XML, and X-HTML.