What Is Software Localization?

Today, there is virtually no area in our private and professional sphere in which no software is used. On conventional PCs, software can be found in the form of an operating system on the basis of which application software is installed—from office applications to multimedia applications to video games. Software can also be found on omnipresent mobile devices, such as smartphones and tablets, and on the Internet in the form of web applications. Additionally, many other devices and machines also contain software: household appliances, cars, industrial production systems, etc.

To increase the user-friendliness and the user experience, such software is often needed in a variety of languages apart from the language in which it was originally developed. This is where software localization comes in, a subject area that is becoming increasingly important.

In the context of the translation of software (as well as marketing texts), the translation process is often referred to as localization. Localization is usually defined as the technical, linguistic, and cultural adaptation of a product—in this case of software—to a regional market. Thus, localization comprises more than the mere translation.

By the way: The term "localization" is often abbreviated as L10N, consisting of the first and last letters of the word and the 10 letters in between.

The localization of software comprises the translation of the program files that make up the software as well as the localization of other software-related components, such as the online help, the product documentation, and the associated web pages.

However, this article focuses exclusively on localization of the software itself.

The software content that needs to be localized mainly comprises texts that are displayed in the user interface, especially the menus and dialog windows (including dialog elements such as checkboxes, drop-down lists, and buttons). Strings that the software displays under certain conditions, such as error and status messages, also need to be localized.

Moreover, the software localization may also comprise keyboard shortcut for executing certain functions as well as version information. The version information contains basic information on the software, e.g. copyright information, the product name, and of course the version of the file or product. This information can be accessed by opening the properties of the respective file.

The software content that needs to be localized mainly comprises texts that are displayed in the user interface.

Localization of software usually takes place using computers and software, such as special software localization tools or conventional Translation Management Systems. Both types of software work in the same way: The software extracts the content to be localized from the program code. Simply stated, the program code consists of the commands that are written in the programming language of the software.

In addition to text, the content to be localized may also contain information on the graphical display of the software, especially of dialog windows and their components. The so-called WYSIWYG mode can be used to ensure that the localization result is correct especially also with respect to the graphical design of the software. The acronym WYSIWYG stands for "what you see is what you get". The term Echtzeitvorschau, or real-time preview, is sometimes used to refer to this approach in German graphical user interfaces. Even while the elements of dialog windows are being localized, you can see what they will look like later on.

Apart from providing a preview and thus a visual context of the elements to be localized, the WYSIWYG mode enables the translator to resize and reposition the graphical elements of the dialog windows. If a button is too small for the target-language designation, the button can simply be resized directly without the need for a software developer to intervene.

Software usually consists of a small or large number of program files. To fully localize software, all these files need to be localized. The files can come in various formats.

Most importantly, the file formats used depend on the respective development platform or programming language. Standard Windows resources and .NET resources are widely used.

In terms of specific file formats, EXE (executable files) and DLL (program libraries) are well known. Other formats include OCX and CPL. The said formats are binary files, i.e. the files have already been compiled.

In addition to compiled formats, uncompiled formats exist as well. In these formats, the content to be localized has been extracted from the respective program files to text files. These are so-called resource script files. Common uncompiled formats include RC, RC2, and DLG.

The advantage of localizing compiled files is that the files are operable immediately upon localization, i.e. they can be used without any overhead for the software developer. By contrast, the advantage of uncompiled files is that the files can be opened and edited with any text editor. However, they need to be integrated in the program code after the localization and be compiled.

Localization content can also come in the form of exchange formats. For instance, XLIFF is a common exchange format for general translation resources. Specific exchange formats from the field of software development include PO (for Portable Object) and JSON (for JavaScript Object Notation).

The advantage of localizing compiled files is that the files are operable immediately upon localization.

Localization of display texts is a software localization subarea that presents a certain number of challenges: field length limitations, widths, proprietary fonts, lack of context information, and lack of file format standardization.

In the localization process for display texts, the strings to be translated are often sent to the translators without any context whatsoever – for fear that they will edit or delete the source code. However, this approach leads to problems, such as translations that are too long or that are unsuitable for the context. To correct these errors, it is necessary to return to the translators again and again, which is costly in both time and money.

Alternatively, the complex process of display localization can be optimally coordinated and rolled out using a translation management system. In our article “Localizing Display Texts”, you will find additional information on this topic as well as tips on how to best localize display texts.

The complex process of display localization can be optimally coordinated and rolled out using a TMS.

If you want to sell your products or services internationally, you will come into contact with the task of software localization at some point. Since software localization involves more effort than a “normal” translation, companies should turn their attention to the topic in due time. The software text should not be extracted and sent to the translators without context information, as doing so can quickly lead to errors. It is therefore recommended that you use a suitable tool to carry out the translation. The Across Language Server allows you to easily have your software translated, thus shortening your time-to-market. In this way, nothing will stand in the way of a successful international launch.

Other articles that you may find interesting:

• All about Translation Management Systems (TMS)
• Optimizing Translation Processes
• Machine Translation for Enterprises