Linguistic localization for software

Localizing a software program or application is an indispensable choice for companies aiming to expand on foreign markets, but it is often challenging and complex. Indeed, unlike traditional technical translations, software includes two types of content: text for users and strings of programming code for machines. These two components can, in any case, be separated using advanced technological tools so that translators can focus on creating the foreign language version of the text without creating bugs and other technical issues. IT professionals and software developers localize the code. For year, Landoor has been known as the go-to provider of software localization services with a team of mother-tongue translators, expert editors specialized in the relevant field, software engineers and testers – in continuous contact with the client throughout the entire process. Landoor is also a reliable partner for software houses active on the international scene and in need of consultancy to develop digital products designed from the start to be seamlessly and effectively localized.

Stages of software localization

The ideal software localization process begins with a project that has already been designed from the onset to be localized in various languages. This will prevent the need for excessive technical changes later when the software launches on other markets and will streamline the time-to-market. For example, to code characters, we recommend using UNICODE, which enables us to manage languages written with accents and other diacritical marks, as well as non-Latin alphabets. Interfaces should also be designed from the start to allow for a potential increase in the bulk of the text for languages that take up more space. In industry jargon, functional localization or internationalization is the process in which specialists use specific tools like Gettext libraries. From a practical standpoint, for an app to support multiple languages, the developer must design it so that, in execution, the program selects the correct language resources. Therefore, the code must be able to, for example, check the proper formats for data entry, which directly depend on the user’s local preferences. Landoor operates on nearly all the most popular hardware and software platforms. The client may select one of two options:

filtering the text and providing the translation agency with a text file only;
providing a localization kit containing executable files (.exe, .com, .bat, .dll, .drv or .bin) and the software development environment.

In the case of the former, the client will then re-enter the text in the code of the proprietary environment. In the case of the latter, the team of linguists will use a localization platform that “blocks the code,” isolating the translation of strings of text. Clients may also provide glossaries and translation memories sourced from previous projects. Landoor’s specialists make full use of computer-assisted translation tools (CAT tools) to maintain terminological and stylistic consistency and run spell checks. If clients do not provide glossaries and translation memories, Landoor’s translators extract the terminology using previously existing company documentation and the industry terms, in accordance with international standards. The accurate translation of the strings of text in and of itself is almost never enough, and Landoor’s translators know this. They unwaveringly focus on localizing the text, in order to express exactly the same message as the original version, but in a way that is as consistent as possible with the norms and mores of users of a certain language or country. This entails, for example, adapting the formal aspects of texts, like the way dates and times, currencies and measurements (temperature, length and weight) are written, the format of addresses and telephone numbers, the symbolism of colors and other historical/cultural references, which need to be replaced with the equivalent “local” forms and information without changing the original meaning. The localizers work in synergy with the software developers and the client to adapt other aspects of the page too, like icons, images and layout. For instance, with languages written from right to left (such as Arabic or Hebrew), the sequence of buttons and images must be inverted for readers who are used to reading in a different direction. The text to be translated for software includes the graphics interface (presentations, dialog boxes, menus, FAQs and error messages) and the online guide. As additional documentation usually accompanies a software program, the team also handles other text like readme files, installation instructions, configuration guides, user manuals, brochures and other promotional material – which require a translation that is consistent with the wording and terminology used in the software interfaces. The quality of the translation/localization work must then be checked and, if necessary, the finishing touches must be made, after checking three aspects of product quality:

linguistic check to verify that the terminology and style are consistent and that the text can be viewed properly with respect to spaces and special characters;
check of the graphics interface, which must not only be legible but also aesthetically pleasing within the layout;
functional check to confirm that that the program functions are the same as in the original version that it is compatible with the hardware for which it was designed (processor, graphics and audio cards).

See also Localization – Translation of apps for mobile devices (iOS, Android and Windows)