One approach to pseudo-translation involves the addition of special characters typical for the locale of the target language (for example a diacritical mark like a German Umlaut 'ä'), as well as changing the number of characters belonging to the text. In that approach, the text is pseudo-translated in a way that allows to recognize the original source text. Another pseudo-translation solution involves the use of machine translation technology, which not only generates the necessary special characters but also gives developers a good indication of the length of a string in a particular target language.
Pseudo-translation precedes the actual translation in the software development process. Its purpose is to test that the software is prepared for translation.
Pseudo-translation allows verifying that:
Verifying no incorrect concatenations have been used requires putting begin and end markers in the pseudo-text. For example: '<' and '>'. It is however not necessary to ensure no truncation in place by using pseudo translation. As the translation from one language to another (for instance, English to German) sometimes requires additional 100% or more space to hold the translated content, it may degrade the quality of the original language (for instance, English) by allocating more wasteful space to just be able to accommodate German version of this item. In this case, subsequent localization process should ensure the translation quality be making sure that the space is adequate for the targeting language.